Digital Dividend Technologies & MHz Spectrum

DIGITAL COMMUNICATIONS KNOWLEDGE TRANSFER NETWORK Digital Dividend Technologies & 470-862MHz Spectrum DCKTN Wireless Technology & Working Group Wireless Technology & Spectrum Working Group Page 1

The Digital Communications Knowledge Transfer Network has been established by an industry-led group of leading players, with funding from the Technology Strategy Board. We seek to bring competitive advantage to the UK by promoting collaboration and knowledge sharing between the users and providers of Digital Communications, and helping to drive innovation in the sector. If you are involved in Digital Communications, and have not yet registered as member of the DC-KTN please visit our web page and register as a member Digital Communications KTN Russell Square House, 10 12 Russell Square, London WC1B 5EE Email: info@dcktn.org.uk Website: www.dcktn.org.uk Telephone: +44 (0) 20 7331 2056 Contact points for positioning paper: Stuart Revell, DCKTN stuart.revell@dcktn.org.uk David Barker, Quintel Solutions david.barker@quintelsolutions.com Rob J Davies, Philips Research rob.j.davies@philips.com Brian Copsey, Copsey Communications bc@copsey-comms.com Andrew Lillywhite, Sennheiser ALillywhite@sennheiser.co.uk Mark Waddell, BBC R&D mark.waddell@bbc.co.uk Steve Hope, Docobo stephen.hope@docobo.co.uk Clive Harding crharding@theiet.org Geoff Varrall, RTT Programmes geoff@rttonline.com The opinions and views expressed within this positioning paper have been reviewed by the members of the Digital Communications Knowledge Transfer Network Wireless Technology & Spectrum working group. The views and opinions do not necessarily reflect those of the individual members of the DCKTN or the Working Group or the organisations that the members represent. Page 2

1 Contents 1 Contents... 2 2 Abbreviations and definitions... 5 3 Introduction... 7 3.1 Purpose of this paper... 8 4 Executive summary... 9 5 Background (Spectrum)... 11 6 Future Digital Dividend Spectrum... 12 7 Digital Dividend compatibility issues... 13 8 Digital Television & Mobile Cellular... 15 8.1 DTV & Mobile Cellular background... 15 8.2 DTV Receiver Selectivity Characteristics... 15 8.3 LTE-800 Interference into Digital Television (DTT)... 16 8.3.1 LTE-800 BS Interference into DTT Ch60 N+1 offset... 16 8.3.2 LTE-800 UE Interference into DTT N+9 offset... 17 8.3.3 Interference resulting from out of band emissions... 18 8.3.4 Interference Discussion Possible solutions... 19 8.4 DTV & Mobile Cellular Network Evolution Proposal... 20 8.4.1 Shared DTV Broadcast and Cellular Topology... 21 9 Hybrid Fibre Cable (HFC) Networks... 24 9.1 Digital Dividend history and impact to Cable & STB industry... 24 9.2 LTE Interference into Cable and DTV STB systems... 25 9.2.1 Set Top Box - LTE Interference testing... 26 9.2.2 Cable Modem LTE Interference Testing... 30 9.2.3 In- Home Cabling LTE Interference Testing... 31 9.2.4 Summary of LTE Interference with Cable systems... 32 9.3 White Space Interference to Cable modems... 32 9.4 Cable Interference - Possible mitigation and solutions... 33 9.4.1 Restrict cable transmissions to below 790 MHz.... 33 9.4.2 Reduce QAM rates to make the system more robust:... 34 9.4.3 Increase HFC signal level... 34 9.4.4 Adapt DOCSIS protocol to improve immunity to interference... 35 9.4.5 Increase immunity of CPE... 35 9.4.6 Connectors... 35 9.4.7 Home Cabling improvements... 35 9.4.8 Pilot Tone... 35 9.5 Cable Summary... 36 10 Communal Aerial Systems... 36 10.1 Communal Aerials Systems and Topologies... 36 10.2 Communal Aerial Interference Issues... 38 10.3 Communal Aerial Customer base... 39 10.4 Communal Aerial Interference Assessment... 39 Page 2

11 Programme Making and Special Events (PMSE)... 40 11.1 PMSE Background... 40 11.2 UHF white space spectrum use by PMSE... 40 11.3 PMSE devices and applications... 42 11.4 PMSE Ch61 69 clearance compliance problem... 42 11.5 PMSE technical requirements:... 43 11.6 PMSE Digital radio microphones... 44 11.7 Implications of new entrants to PMSE... 44 11.8 PMSE Solutions and recommendations... 45 12 White Space... 46 12.1 White Space Applications... 47 12.1.1 Home media distribution... 47 12.1.2 Wireless broadband... 47 12.2 White Space Regulatory and Standards background... 48 12.2.1 FCC... 48 12.2.2 IEEE... 48 12.2.3 CogNeA and ECMA... 48 12.2.4 Ofcom... 49 12.3 White Space Spectrum availability... 49 12.4 White Space Interference challenge... 50 12.4.1 Interference to incumbent services... 50 12.5 White Space topologies... 51 12.6 Spectrum sensing... 51 12.7 Geo-location database... 52 12.8 Propagation modelling... 53 12.9 White Space and Cable TV networks... 53 12.10 Can White Space Leverage other Technologies?... 54 12.10.1 Scanning networks co-located with Mobile Infrastructure... 54 12.10.2 White Space downlink capacity using LTE MBMS mode... 55 12.10.3 TV Receivers providing location data... 56 12.10.4 TV Receivers transmitting beacon for White Spaces... 57 12.11 White Space - Summary... 57 13 Public Safety... 58 13.1 Public Safety background... 58 13.2 Key Issues... 58 13.3 TETRA 2 Radio and Spectrum... 58 13.4 Cellular Networks as an alternative to TETRA 2... 59 13.5 Public Safety Summary... 60 14 Applications above 862 MHz Short Range Devices... 61 14.1 863MHz-865MHz testing (Copsey Communications)... 62 14.2 863MHz 865MHz test set up and results... 63 14.3 863MHz-865MHz test conclusions... 67 15 Opportunities for Innovation (Industry & Academia)... 68 15.1 Content and Services... 69 Page 3

15.2 Network Topologies... 69 15.3 Consumer Platforms... 70 15.4 RF Front End Components... 70 15.4.1 Radio and Telecommunications Terminal Equipment Directive (R&TTE).. 71 16 Summary and Collaboration... 73 Page 4

2 Abbreviations and definitions Abbreviation 3D TV 3GPP BS C/I CATV CAGR CENELEC CEPT Ch COFDM COGNEA CPE DAB DCKTN Dig Div (DD) DOCSIS DSO DTG DTT DTV DVB-C DVB-T DVB-T2 DVD ECC ECMA ECN EDGE EIRP ERP ETSI FCC FM FTTC GI GMSK GPRS GPS GSM GSMA HD HDTV Definition Three Dimensional Television 3 rd Generation Partnership Project Base Station Carrier to Interference (Ratio) Cable Television Compound Annual Growth Rate European Committee for Electrotechnical Standardisation European Conference of Postal and Telecommunications Administrations Channel (Allocated frequency slot) Coded Orthogonal Frequency Division Multiplexing Cognitive Networking Alliance Customer Premise Equipment Digital Audio Broadcast Digital Communications Knowledge Transfer Network Digital Dividend 800MHz Spectrum (790-860 MHz) Data Over Cable Service Interface Specification Digital Switchover Digital Television Group Digital Terrestrial Television Digital Television Digital Video Broadcasting Cable Digital Video Broadcasting - Terrestrial Digital Video Broadcasting Terrestrial (2 nd generation) Digital Video Disc or Digital Versatile Disc Electronic Communications Committee Standards body for Information and Communication Technology and Consumer Electronics Electronic Communication Networks Enhanced Data rates for GSM Evolution Effective Isotropic Radiated Power Effective Radiated Power or Equivalent Radiated Power European Telecommunications Standards Institute Federal Communications Commission (US Regulator) Frequency Modulation Fibre To The Curb Guard Interval Gaussian Minimum Shift Keying General packet radio service Global Positioning System Global System for Mobile communications Global System for Mobile communications Association High Definition High Definition Television Page 5

HDMI HFC HSPDA HSPA+ ICT IEM IF IMT IPTV IRS LBT LNB LTE MATV MBMS MER MIMO NTP OFDM PCB PMSE QAM QoS R&D R&TTE RF RFID RS SAB SAP SIM SMATV SNR SRD STB TACS TETRA TETRA 2 TRP UE UHF UMTS UMTS r99 WSD xdsl High-Definition Multimedia Interface Hybrid Fibre Cable High-Speed Downlink Packet Access Evolved High-Speed Packet Access Information and Communication Technologies In Ear Monitor Intermediate frequency International Mobile Telecommunications Internet Protocol television Integrated Reception Systems Listen Before Talk Low Noise Block (converter) 3GPP Long Term Evolution Master Antennae Television Multimedia Broadcast Multicast Service Modulation Error Rate or Modulation Error Ratio Multiple-input and multiple-output antenna configurations Network Termination Point Orthogonal Frequency Division Multiplex Printed Circuit Board Programme making and special events Quadrature Amplitude Modulation Quality of Service Research and Development Radio and Telecommunications Terminal Equipment Directive Radio Frequency Radio Frequency Identification (RFID tag) Reed Solomon Services Ancillary to Broadcast Services Ancillary to Programme making Subscriber Identity Module (SIM Card) Satellite Master Antennae Television Signal to Noise Ratio Short Range Device Set Top Box Total Access Communication System TErrestrial Trunked RAdio TErrestrial Trunked RAdio (2 nd generation) Total Radiated Power User equipment Ultra High Frequency Universal Mobile Telecommunications System UMTS Release '99 higher speed data transmission in 3GPP networks White Space Device Digital Subscriber Line Page 6

3 Introduction Wireless technology has already reached a level where it is embedded in our daily work, social and leisure activities. The demand for services is now reaching beyond the traditional broadcast, mobile cellular applications and local area network connectivity. The DCKTN Wireless Technology & Spectrum working group believes Wireless Technology is the key enabler for several markets and is seen as the critical element for future development of these segments. Wireless is becoming all pervasive and critical for future development in areas such as: Future Mobile Internet Digital Media & Content Energy & Environment Public Safety & Emergency Services Health Assisted Living Smart Grids / Metering Intelligent Transport Automation M2M applications Future of the internet ecommerce, Payment technologies and Banking The terrestrial broadcast spectrum released by the switch from analogue to digital TV transmission is known as the digital dividend (DD) and will become available for new applications in 2012. This spectrum includes cleared and interleaved space between 470 MHz and 862 MHz. The band has many desirable propagation characteristics which can enable enhanced broadband communications services to a wider population. The broadcast spectrum has traditionally been used solely for TV broadcast and low power radio microphones for programme making and special events (PMSE). New applications targeting the spectrum include mobile telephony, which requires dedicated, cleared spectrum, and white-space devices that can potentially use frequencies that are not used locally for licensed TV or PMSE applications. Reallocation of broadcast spectrum to provide these new services requires careful coordination with existing users to prevent interference. Potential issues need to be identified at an early stage, as fixing these problems retrospectively will become increasingly expensive as the new services are deployed. The existing and new applications for the UHF broadcast band are given below: Page 7

Existing services: 1) Broadcast Television Digital TV (470-862MHz) 2) Programme Making and Special Events (PMSE) (also referred to as SAB or SAP 1 ) 3) Cable Television and home media consumer platforms 15-862MHz 4) Communal Aerial Systems (distribution of terrestrial broadcast to multiple dwellings) 5) Short Range Devices (Adjacent band 863-870MHz) New services targeting the DD spectrum: 1) Mobile Cellular (791-862MHz) 2) White Space Devices (470-790MHz) 3) Emergency Services The challenges and opportunities this represents provides the UK with some significant opportunities for our R&D community to address locally and benefit from global exploitation. 3.1 Purpose of this paper The purpose of this paper is to provide the basis for discussion amongst all interested stakeholders as to how a deeper collaboration regarding, Digital Dividend Spectrum, Technology development and Service End Users can stimulate new exploitable opportunities for the UK R&D community. The document and associated presentations shall be used at appropriate Knowledge Transfer events to inform the wider industry, DCKTN members and as vehicle to stimulate discussion with Government bodies. This paper discusses the compatibility issues between existing services and new applications of the Digital Dividend spectrum. The purpose of the paper is to highlight the known issues and suggest areas requiring further investigation by industry and academia ahead of new deployments. The DCKTN working group believe these can be solved through further R&D activity / investment, therefore providing the UK eco-system an excellent opportunity to lead in development and deployment of world leading wireless technology and services. 1 Services Ancillary to Broadcasting or Services Ancillary to Program making Page 8

4 Executive summary This document will look at each of the potential issues and suggest some solutions and areas for further exploration and/or analysis. The executive summary is intended to provide an overall summary with the supporting sections providing more detailed information for the reader to use if required. Table 2 in section 7 provides a good overview of how potential issues interact with different services and users of the spectrum. Table 1 below is a summary of the issues detailed in this document. The vertical axis represents the users or service being provided and the horizontal axis represents the potential interferers. Each box summarises the potential solutions or action required. Page 9

PMSE (470-862MHz) Existing Users Digital TV (470-790MHz) Cable (15-862MHz) Table 1: Positioning paper recommendations New Entrants (Potential interferers) Mobile (791-862MHz) Long term solution: Equipment design recommendations for Cable Receivers Short term solution: High probability LTE in close proximity will interfere. R&D opportunities exist to solve these issues, industry challenge. Encourage Industry and Ofcom to conduct further analysis. Encourage Cable and Mobile industries to conduct further collaborative engineering trials. Collaborative R&D opportunity Adjacent channel ch60 and image channels +9 White Space (470-790MHz) Possible White Space device (in close proximity), will interfere. Encourage Industry and Ofcom to conduct further neutral analysis. Collaborative R&D opportunity R&D opportunities exist to solve these issues, industry challenge. Long term solution: Equipment design recommendations for DTV receiver conformance parameters. Short term solution: High probability LTE in close proximity will interfere. R&D opportunities exist to solve these issues, industry challenge. Encourage Industry and Ofcom to conduct further analysis. Cognitive and/or data base driven usage not permitted. Collaborative network topologies innovation opportunities exist to solve co-existence problems R&D opportunities exist to solve these issues, industry challenge. Encourage Broadcast TV and Mobile industries to conduct further collaborative engineering trials. Potential interference to ch60 assignments - Encourage Industry and Ofcom to conduct further neutral analysis. Band management required Dynamic control to ensure PMSE assignments protected from White Space assignments. Encourage Industry and Ofcom to conduct further analysis. R&D opportunities exist to solve these issues, industry challenge. White Space and PMSE industries to explore possible solutions. R&D opportunities exist to solve these issues, industry challenge. Page 10

5 Background (Spectrum) The UHF spectrum from 300MHz to 3GHz is particularly attractive for mobile and portable applications. Above 3GHz, high building penetration losses prevents use indoors and below 300MHz, the antenna size become impractically large for portable devices. The development of mobile services has relied on spectrum vacated by established services with most of the current cellular spectrum formerly used either by the military or for fixed links which have since migrated to higher bands or alternative technologies such as fibre. The figure below shows how the spectrum available for mobile has grown in the last 40 years, especially in the higher frequency bands. Further growth in cellular applications, especially for broadband wireless data, is resulting in increasing demands for new spectrum to be released. 500MHz 750MHz 1GHz 1.25GHz 1.5GHz 1.75GHz 2GHz 2.25GHz 2.5GHz 2.75GHz 1970 500MHz 750MHz 1GHz 1.25GHz 1.5GHz 1.75GHz 2GHz 2.25GHz 2.5GHz 2.75GHz 1990 500MHz 750MHz 1GHz 1.25GHz 1.5GHz 1.75GHz 2GHz 2.25GHz 2.5GHz 2.75GHz 2010 Military Aeronautical Broadcasting Fixed Links Cellular Other (e.g. GPS, WiFi) Figure 1: Evolution of Spectrum allocations up to 3GHz from 1970 to 2010 Note figure above Cellular = Mobile applications. Digital TV broadcasting provides greater efficiency allowing an increased number of TV services to be delivered in less bandwidth compared to analogue delivery. This has enabled the top end of the TV spectrum to be cleared for use by the cellular industry. Different regions of the world are reallocating different parts of the TV spectrum for mobile as shown below: Page 11

Figure 2 Regional digital dividend spectrum allocation (Source: GSMA) 6 Future Digital Dividend Spectrum Another area for consideration is the Digital Dividend spectrum potentially to be cleared at a later date. In this area we would encourage all parties to review the process we have gone through for the 790 862MHz clearance and attempt to address the issues highlighted above at an earlier stage. Ofcom is currently consulting on the 550MHz to 606 MHz band usage 2. The list below details some potential ideas that need to be explored by all parties prior to making any decisions on how these bands should be allocated and potentially used for Mobile applications, please note items 2) and 3) below are mutually exclusive: 1. DD technical recommendations from 61 to 69 apply to any new cleared bands at lower frequencies, therefore lessons learned need to be applied in the decision making process. 2. Future usage of 7 channels Ch 31(542-550MHz) to Ch 37 (598-606MHz) Ensure adequate guard band is allocated to protect adjacent services. Therefore avoiding the 1MHz guard band issue we have in the upper cleared spectrum. Dedicating Ch 31 and Ch 37 guard bands to PMSE usage. 2 Digital dividend: 600 MHz band and geographic interleaved spectrum http://www.ofcom.org.uk/consult/condocs/600mhz_geographic/ Page 12

3. Retain Ch 31 to Ch 37 as a reserve to allow further upper clearance. Therefore moving TV broadcast into these bands at a later date. Ch 53 to (718-726MHz) to Ch 60 (784-792MHz) to be cleared for cellular usage (8 channels) therefore providing alignment with some US frequencies to exploit economies of scale. Can more than 8 channels be cleared? Potentially explore going to Ch 49 which would mean 12 channels from 49 (686-694MHz) to 60 (784-792MHz) Benefits: Harmonising with larger global eco-system. 7 Digital Dividend compatibility issues The digital dividend is providing many new challenges for the wireless industry to address. The combination of interested parties is unique due to the attractiveness of the spectrum. New licensed-exempt devices and mobile cellular services using paired and unpaired spectrum assignments are vying to operate alongside traditional broadcast and radio microphone applications. The potential problems are particularly acute in the UK where a successful broadcast industry makes extensive use of the band and serves 60% of TV households. Traditionally the wireless industries have always differentiated licensed-exempt and licensed spectrum by assigning different bands. White Space technology will introduce a third mode of operation whereby under-utilised, licensed spectrum will be used on a regional basis to support transmissions by new licensed-exempt devices. Careful management of these devices will be required prevent interference to the licensed services PMSE and broadcast industry. The UK PMSE industry has been addressing this challenge for a number of years by maintaining a geographical database and manually setting the frequency of operation. The operation has been managed well because of the relatively low volumes and the operation of such devices by professional users. The new white space opportunity is looking at consumer volumes so a manual mode like this will not be practical or manageable. Further discussion on PMSE and white space is provided in dedicated sections later in this document. The cable industry has been using the entire analogue UHF TV band and additional spectrum at lower frequencies. Typical systems operate between 15-862MHz within wired hybrid fibre optic and coaxial cables (HFC) and provide high quality TV and high speed (up to 100 Mbps) broadband connections to the consumer. The deployment of mobile cellular between 790-862MHz can inject interfering signals into the in-home set top box and cabling, due to co channel interference and imperfect screening; this can disrupt the quality of the video and broadband cable services. Interference between different types of service is usually controlled by guard bands which are used to separate the different assignments. For new mobile cellular Page 13

Services impacted deployments planned in the upper cleared band (790-862MHz), the guard band has been reduced to just 1MHz. The close proximity of cellular and digital TV assignments will result in interference to digital TV services in some areas, particularly near the edge of coverage. This will require co-ordination between the broadcasters and the mobile operators to manage the problem. This document will look at each of these potential issues and suggest some areas for further exploration and/or analysis. The table below shows a matrix of the potential interactions between the different technologies: Table 2: Interaction of Digital Dividend potential users Potential Interferers Spectrum users Cable (15-862MHz) Cellular (791-862MHz) Digital TV (470-790MHz) PMSE (470-862MHz) White Space (470-790MHz) Cable (15-862MHz) Cellular (791-862MHz) No interference Yes potential co-channel breakthrough into cable systems and CPE Close proximity to broadcast transmitter / repeaters Yes responsibility of Cellular to solve Negligible Although this has always been possible, but statistically very low No -PMSE will vacate 61-69 Yes White Space devices in close proximity to Home STB Yes - potential issue around 790MHz adjacent channel Digital TV (470-790MHz) PMSE (470-862MHz) White Space (470-790MHz) Short Range Devices (863 to 870MHz) No interference No interference No interference No interference Yes Adjacent channel ch60 and image channels +9 Potential interference to ch60 assignments Potential interference to ch60 assignments Yes - potential interference Not applicable DTV own spectrum Not applicable DTV own spectrum No interference No interference controlled by PMSE band manager Yes White space devices will need to avoid PMSE assignments PMSE use this spectrum for licensed exempt Yes Geolocation and sensing to control access Yes PMSE & White space will need to interact to minimise impact No interference The following sections will look into all of these technologies in more detail Page 14

8 Digital Television & Mobile Cellular 8.1 DTV & Mobile Cellular background The 790-862MHz spectrum band has been proposed as a harmonised band for additional broadband mobile cellular applications (typically LTE-800) across Europe and other parts of the world as a benefit of digital switch over (DSO). This spectrum band is proximate and adjacent to DTV broadcasts which will continue to occupy spectrum up to 790MHz. The transmitter powers, sites, network delivery infrastructure, antennas and network topologies are quite different for broadcast DTT and mobile cellular networks and as a result of these differences and the very small guard band, a number of interference scenarios are emerging. These need to be considered, understood and ultimately resolved to ensure compatibility between cellular and broadcast technologies. The LTE-800 mobile band has been designed to support 2x30MHz paired LTE FDD deployment in either 5MHz or 10MHz bandwidth modes. The proposed channel plan was developed within CEPT SE42 and is shown in Figure 3 below for the 5MHz LTE mode. For 10MHz LTE modes, the 5MHz carriers would be aggregated in pairs. The plan includes a 1MHz guard band between the edge of the broadcast channel 60 and the first LTE base station downlink. A reverse duplex scheme was chosen, with the base station in a lower frequency band to the mobile uplink, as this was felt to reduce some of the interference effects between handsets and the broadcast service. Figure 3 Frequency plan for LTE-800 deployment The close proximity of the LTE base stations (BS) to the broadcast channels and the finite selectivity and image rejection performance of DTT receivers results in a number of interference mechanisms. 8.2 DTV Receiver Selectivity Characteristics The selectivity performance of domestic DTT receivers is specified in the UK by the DTG, who define minimum performance targets in a publication known as the D-book. Page 15

The selectivity performance is specified in terms of a protection ratio which is the minimum carrier to interference ratio required for the onset of picture failure quality at a given frequency offset. Performance targets for LTE interferers have not yet been defined, but since DVB-T and LTE base stations both use COFDM technology, the C/I characteristics for LTE into DTT are expected to be similar to the DVB-T into DVB-T targets currently specified. Figure 4 shows the C/I targets specified by the DTG, and a graph of typical performance for the average of 7 receivers using traditional discretecomponent, Tin-Can tuners: Figure 4 DTG C/I performance targets for DTT receivers and mean performance for 7 receivers Using this data, it is possible to predict the LTE signal levels that will provoke the onset of picture failure. 8.3 LTE-800 Interference into Digital Television (DTT) The following sections provide information regarding the main interference mechanisms. 8.3.1 LTE-800 BS Interference into DTT Ch60 N+1 offset Given the 1MHz offset between the broadcast service in UHF channel 60 and the first LTE downlink in channel 61, interference can be expected when the adjacent channel protection ratio target is exceeded. This will typically occur when the LTE signal level is 27dB greater than the DTT signal and is shown in figure 5. Page 16

Given a received DTT signal level of -72dBm at the coverage edge, the maximum level of LTE signal that can be tolerated will be -45dBm. The proposed base station power levels are expected to be 60dBm, suggesting a minimum isolation requirement of 105dB. This corresponds to a separation of 5km between the base station and the victim DTT receiving antenna for the worst case where line of sight propagation between the base station and a 7dBi TV antenna applies. Figure 5 LTE-800 Base station (BS) interference to DTT CH60 8.3.2 LTE-800 UE Interference into DTT N+9 offset The DTT receiver characteristics shown in figure 4 show a steadily improving performance as the offset between the interferer and wanted signal increases. This is to be expected given the characteristics of the receiver s IF and RF filters. There is however a noticeable reduction in performance for interferers at N+9 offset (72MHz) due to the finite image rejection performance of tin-can tuners with a first IF of 36MHz. This gives rise to an interference mechanism whereby a handset at 72MHz offset from the wanted DTT channel can produce picture break up. This scenario is illustrated in Figure 6 below: Page 17

Figure 6 LTE handset (UE) interference to DTT CH57-60 For a typical handset EIRP of 25dBm, and a received DTT signal of -72dBm at the coverage edge, the maximum permitted level of handset signal for the onset of failure would be -41dBm, suggesting a required isolation of 66dB between DTT antenna and LTE UE. This corresponds to a minimum separation of 50m between handset and DTT antenna. In practice, the interference may be worse still, as tests on LTE handsets suggest the time domain characteristic of the signal results in degraded receiver protection ratio performance resulting in picture breakup with lower level interferers. 8.3.3 Interference resulting from out of band emissions Problems resulting from the finite selectivity of the receiver are only a part of the story. In practice, a mobile base station will radiate significant energy into the broadcast band, and this will result in interference, even if additional filtering is used at the receiver in an attempt to reject the base station signals. This interference cannot be addressed at the receiver. This effect is shown in figure 7 below and interference levels up to 0dBm cochannel with TV channel 60 are permitted by the recommendations made by CEPT SE42. The permitted levels of out of band would result in a 20dB degradation in receiver sensitivity at 500m from the cellular base station, and could dramatically reduce broadcast coverage. Page 18

Figure 7 LTE base station OOB causing interference to DTT CH60 8.3.4 Interference Discussion Possible solutions The interference mechanisms described so far are an inevitable consequence of the non-ideal behaviour of receivers and transmitters. Although the RF transmissions are designed to occupy their intended bandwidth, RF energy leaks into adjacent spectrum due to finite implementations of the transmitter devices, and the selectivity performance of the receiver (i.e. receivers will also pick up energy outside of their intended receive bandwidth). Interference problems will be a particular issue towards the edge of DTV coverage and TV channel 60 appears to be particularly vulnerable. Two solutions are emerging: (1) Use fast roll-off filtering at the LTE800 base station to ensure additional protection or (2) Re-transmit the DTT signal from the cellular base station site, thus ensuring that the link budget differential between DTV and LTE800 at the DTT receiver falls within the limits defined by the protection ratios. The following sections discuss how collaboration between the cellular and broadcast industries might further address these problems to ensure and guarantee harmonisation and minimize pathological interference scenarios. Furthermore, such collaboration may result in additional opportunities creating value to customers through working together at the infrastructure, technology and business levels. Page 19

8.4 DTV & Mobile Cellular Network Evolution Proposal The terrestrial TV broadcast industry has evolved at a somewhat slower pace to the cellular industry particularly in terms of its network topology. Unlike the cellular industry domestic TV installations are deemed expensive to upgrade (e.g. the fixed rooftop aerial, cabling and feeds for multiple TV sets). DTT network planning has many constraints due to domestic antenna frequency groupings, the directive properties of the antennas and the need to coordinate with international neighbours. As a result, DTT still uses high-power/high-tower network topologies designed using frequency planning and co-ordination rules dating back to the 1961 Stockholm Agreement. Terrestrial DTT is also seeing slowly growing competition from alternative delivery means such as Satellite (Sky, FreeSat), Cable (VirginMedia), and even broadband IP (IPTV delivery), but nevertheless remains the dominant platform in the UK. Given this, one can sympathise with the fact the terrestrial broadcast industry is losing spectrum as part of the digital dividend rather than exploiting it. Competing platforms are able to offer HDTV and many more channels. IPTV and cable also enable interactivity, which are becoming increasingly important to consumers. Many countries across the world have terrestrial TV delivery platforms. This includes UK, France, Spain, Italy, India, and China as examples. Cable and/or Satellite TV delivery platforms are more popular in other countries such as US, Canada, Germany and Japan, although these still having a significant terrestrial TV component, particularly important for portable and mobile services. Cellular networks have evolved at a more rapid pace. Over 25 years, cellular operators now have acquired 3 spectrum bands and this is expected to increase to 5 bands soon with the introduction of 800 and 2.6GHz services. The networks have gone through many generations of technology for the radio interface (TACS, GSM, GPRS, EDGE, UMTS, HSPA, and soon HSPA+ and LTE), and have adopted network spectral efficiency techniques such as cell-splitting, variable tilting antennas, frequency hopping (GSM) and now an evolved heterogeneous topology of Macro, Micro, Pico and soon Femto cell layers. This has been possible as the consumer is prepared to replace their terminal equipment, almost on a yearly basis, which is certainly not the case for broadcast TV receivers. Terrestrial DTV and Digital Dividend Cellular share the prime UHF radio spectrum for delivery of content and information. Given this, the solution of re-broadcasting DTV from a cellular site to mitigate cellular interference to a DTV receiver has been proposed to permit co-existence of the two networks and mitigate against interference. A remaining question is why the natural evolution can t be to arrive at a shared network topology and infrastructure. The next section explores what could be achieved if this infrastructure sharing could be carried out. Page 20

8.4.1 Shared DTV Broadcast and Cellular Topology The proposed thinking brings together Broadcast and Cellular networks at three levels to exemplify the benefits of working together at the network topology level: a) Solves the near-far adjacent spectrum interference conditions discussed earlier for all DTV-Cellular cases. b) Re-broadcast (via cabled connection or on-air repeater) of a quasi-synchronous transmission of DTT signals from a subset of mobile operator cellular sites under the footprint of the main DTT broadcast antenna. This enhances DTT coverage significantly, allows realistic Mobile DTV, could even remove the need for rooftop aerials (which could in turn permit faster evolution of the DTT industry), allows significant power reduction from the main broadcast transmitter which in turns offers significant spectral efficiency in spectrum re-use, thus permitting more channels/content and/or HD content, and even a further dividend spectrum. c) Assuming we keep the rooftop TV aerial in b) above then the domestic rooftop UHF Yagi antenna could also be connected to an LTE800 device in the home. The fact the Yagi antenna is at 10m height, outdoors, above the rooftops/clutter, and has UHF directivity/gain can equate to something like a 10-15dB gain in received signal quality strength for LTE (over an indoor handheld device). This allows the mobile operators operating say an LTE800 network to offer 10 s of Mbps fixed broadband delivery and thus competing against fixed broadband operators using xdsl, or Cable, and if the LTE800 signal is on-air repeated in the house then this even may mitigate the need for domestic femtocells. At a business level one could conceive the DTT operators teaming up with Mobile Operators to offer Quad-play services; a mix not contemplated previously due to dissimilar infrastructure. In the UK DSL lines have average distance to the DSLAM of about 3km. This in turn leads to 2-16Mbps broadband rates delivery using Shannon s laws. The average distance to the nearest base station is nearer 1km in suburban/urban areas, and again using Shannon s laws (which LTE approaches and exceeds with MIMO) we could expect 10-30Mbps broadband delivery. It must be noted that comparisons between fixed and mobile technologies can be misleading, as the same generation of fixed connections generally have higher bandwidth and less contention compared to wireless. An example of this is BT s superfast broadband product (branded as BT Infinity for the FTTC version) which is in Page 21

the process of being rolled out. The downlink is up to 40 Mbit/s and the uplink is up to 10 Mbit/s. Comparing fixed and wireless technologies can only really be compared if you require Mobility (wireless) and/or the economics/business model prohibit deployment. Figure 8 Broadcast and Cellular Services repeating DTT The figure above depicts a main broadcast transmitter serving rooftop aerials with DTT services. A nearby cellular base station with sector pointing (approximately) at the rooftop aerials is used to re-broadcast the TV signal. The cellular base station delivering an LTE800 mobile service can be received on the rooftop aerial, therefore both spectrally enhanced TV and enhanced signal level for LTE800 signals are received in the home thus offering the potential delivery of HDTV, more TV channels, and an LTE800 service which can be connected to a Wireless Router, as a fixed broadband replacement. Moreover, another incarnation could capture the fact that the LTE signal could be literally on-air repeated inside the home thus providing enhanced in-building coverage; there may also be sufficient isolation between rooftop aerial and re-broadcast to permit a simple device (as opposed to a device which requires smart control of the rebroadcast of the LTE800 signal). The domestic aerial cabling may be an issue in some cases, but the acid test is; if a 64QAM DTV signal can be picked up then normally the house-owner has optimised the installation, and should be fine for LTE800. For rooftop aerials with LNA s then this is fine for delivery of the LTE800 downlink channel, but not for the uplink channel. An LNA with cut-off at around 825MHz may be required rather than 862MHz, or the indoor unit performs this filtering and relies upon a slower LTE800 Page 22

uplink connection (i.e. one that has to go the usual route through the walls of the house to the base station); this may be sufficient as it is the LTE800 downlink channel which is probably the more important link for data rates and user consumption experience. The network topology for delivery is one where a main broadcast transmitter is at the centre and the broadcast transmission is re-transmitted (as a repeater) from a network of cellular base stations where the time difference between main broadcast antenna and repeated signal is within the Guard Interval (GI) of the DVB-T. Furthermore, retransmission from the cellular network sites doesn t have to be all cellular sites, and needs only to be from one sector (sometimes two sectors), in order that the domestic TV rooftop antennas do not have to be re-pointed. Figure 9 Benefits of repeating DTT from Cellular Base Stations The image above depicts the shared topology across a 40km radius area. The left image depicts the coverage to rooftop aerials for TV service; the image on the right depicts the coverage when delivered (repeated) via a grid of cellular sites. Beyond a certain distance (20km), the cellular sites will always provide better service coverage. The FCC plan 3 also proposes the use of the broadcasters transitioning to a cellular architecture to solve many of the interference issues and broadcast/cellular coexistence; and hence a useful reference for collaborative thinking. 3 FCC National Broadband Plan, Connecting America. http://www.broadband.gov/download-plan/ Page 23

9 Hybrid Fibre Cable (HFC) Networks Some five million homes use the HFC networks within the UK to receive a wide range of services which include Digital TV (both standard and high definition), video on demand, interactive and time shifted services, in addition to high speed broadband. HFC networks support the Government plans for wide scale high speed broadband with current speeds of between 50 and 100Mbits/s, and the ability of DOCSIS 3 to bond up to at least 32 channels mean that the final speeds to be achieved will be well in excess of the current offerings. HFC head ends are fed with incoming material via fibre optic links from the broadcasters or studios, and make use of some off air signals and stored media, enabling a single head end to serve in excess of one million customers. A head end distributes its signals using DVB-C and DOCSIS modulations via a fibre optic and coaxial cable distribution system terminating at the customers premises with an isolating unit. Cabling from this unit to the various devices inside the premises is of high quality dual, triple or quad screened coax cable. Telephone services are also carried alongside the fibre/coaxial network and terminate in a conventional copper pair at the customer premises. Handbook providing in depth information is available on the DCKTN website at: http://www.dcktn.org.uk/ 9.1 Digital Dividend history and impact to Cable & STB industry The cable industry was not aware of the changes in spectrum use until early 2009, since then it has taken part in a range of activities including; European Commission,ECC and standards bodies along with Ofcom UK. Testing has shown that there is little in the way of filtering or other simple add-ons which will dramatically improve the situation in the UK. The following sections will investigate the interference scenarios that have been established. Page 24

9.2 LTE Interference into Cable and DTV STB systems A testing program carried out by Cable Europe Labs (the trade association of the Cable industry) and supported by Virgin Media has carried out an investigation into the interference mechanisms which can be expected from the new Electronic Communication Networks (ECN) and other use of cleared spectrum, (copies of this work can be found at : http://www.dcktn.org.uk/ The cable networks are operating between 15-862 MHz, which is today split into two sections: upstream from approximately 15 65 MHz, the return path from the subscriber for data and interactive services, and downstream from 85-862 MHz providing the delivery of programs and data from the head end. Figure 10 Main line COAX & Connector with one pound coin Each head end, which is capable of serving in excess of 1 million people, may have different frequency plans dependant on local requirements. The testing program concluded that the HFC network itself was a robust structure and that the vast majority of interference would be experienced within the home. Where an ECN terminal device (or any other transmitter within the frequency reception range of the CPE) was in proximity to the customer premises equipment (Set Top Box, recorder, modem etc) at ranges of up to 3-4 metres. An interference range of some 3-4 metres is for a terminal unit operating at the maximum power quoted in ECC Decision of some 316 mw. These conclusions are similar to those found during other testing carried out Page 25

by the Dutch Administration and within Germany. The various tests suggest that interfering signals generated in one premises will affect equipment in adjacent locations. Individual household configuration of cable input (access), equipment location (position) and configuration of end equipment vary widely driven by customer preferences. The most common UK scenarios are listed below and section 9.2.1shows the configuration used for these tests: Scenario 1 The first scenario is representative of a simple installation in a dwelling with only two devices, which is within 125m of a street-based tap. Scenario 2 The second and third scenarios represent a dwelling either: a. Within 125m of the street-based tap with more than two devices (e.g. two set top boxes and one cable modem); or b. Further than 125m from the street-based tap, and with either two or more devices. This scenario makes use of a single isolator at the Network Termination Point (NTP), providing feed to the devices via a 4-way distribution amplification unit. This scenario is likely to represent 10% of consumers falling into scenarios 2 and 3. Scenario 3 The third scenario is likely to represent 90% of consumers falling into scenarios 2 and 3. The initial isolator at the NTP is replaced by three isolators, one before each device. 9.2.1 Set Top Box - LTE Interference testing The UK components were assembled on a board as shown in the figure below. The scenario in use here is entitled Scenario 1 as provided by Virgin Media, and makes use of a Tratec isolator, a two-way splitter with equal 3.6dB output attenuations, a 2m trishield cable for the STB, and an HDMI cable as provided with the STB or long SCART cable. Where an HDMI port was present on the STB, it was used in preference over a SCART port. Where no option existed, a 15m SCART cable connected the STB port to a television in the control room. Where coaxial ports remain unused on victim equipment, they have been left un-terminated, whilst all unused ports or cables attached to the cable network are terminated. Page 26

Figure 11 British scenario with a set top box, representative system outlet and cabling Specific configuration data is contained in the following tables. Table 3: Accessible signal level control point configuration Affect on signal Point level Tratec Ecoline 3-way splitter Tratec tap Prior to drop cable Tratec isolator Tratec Ecoline 2-way splitter Page 27 7dB attenuation 25dB attenuation 3dB pad 0.4dB attenuation 3.6dB attenuation Table 4: Cables used in installation Cable Length From To Times Fiber RG6 siamese White tri-shield CPE cable HDMI cable provided with STB Scart cable 70m 9m 2m 15m Trunk network System outlet System outlet STB output CPE installation Set top box HDMI-fibre optic converter Television in control room

Signal standard Table 5: Wanted signal characteristics Input level (to Modulation Symbol rate the STB) Frequency range DVB-C 256QAM 6.875 Ms/s -10dBmV 675-859MHz Signal levels present at the input port of the STB are approximately -10dBmV and provide a Modulation Error Ratio (MER) of approximately 30, as shown on the following screenshots. Figure 12: Sunrise Telecom CM2000 screenshots showing (left) an SLM mini scan (red arrow indicates the interferer s frequency) and (right) QAM constellation. This signal level provided an MER of approximately 30dB on each STB. Such an MER caused approximately one pre-reed Solomon (RS) error per second; the RS filter is able to correct these errors such that there were zero post-rs errors. This level is considered worst case because a minor variation to any of the factors affecting MER could result in a post-rs error, and a visual manifestation of the problem, most likely in the form of macro blocking. Graph of results below:- Page 28

LTE transmit ERP (dbm) On-channel 35 DVB-C STB tests: Maximum LTE level permitting stable audio and video on 256QAM channels from 675-859MHz. STBs in representative networks, -10dBmV DVB-C input level (LTE centre frequency 795 MHz, calculated separation distance 0.5m) The red 316mW line is the nominal pow er (CEPT) of the terminal unit. STB 1 STB 2 30 STB 3 25 STB 4 20 STB 5 STB 6 15 STB 7 10 STB 8 STB 9 5 STB 10 0 STB 11 STB 12-5 STB 13-10 STB 14-15 STB 15 STB 16-20 STB 17-25 STB 18 STB 19-30 STB 20-35 675 683 691 699 707 715 723 731 739 747 755 763 771 779 787 795 803 811 819 827 835 843 851 859 256QAM channel centre frequency (MHz) STB 21 Figure 13: LTE-radiated European scenarios, STBs 1 to 21 LTE (10MHz BW) at 316mW ERP Page 29

9.2.2 Cable Modem LTE Interference Testing A similar test configuration to that shown in 9.2.1 was used for cable modems. Table 6: Wanted signal characteristics Signal standard Modulation Input level (to the STB) Frequency range DOCSIS 256QAM -10dBmV 795MHz Signal levels present at the input port of the cable modem are approximately -10dBmV and provide an MER of approximately 30dB, as shown on the following screenshots. Figure 14: Sunrise Telecom CM2000 screenshots showing (left) an SLM mini scan (red arrow indicates the DOCSIS and interferer s frequency) and (right) QAM constellation. Page 30

LTE ERP (dbm) On-channel Maximum LTE level permitting stable data on a 256QAM channel at 795MHz. CM in representative networks, -10dBmV DOCSIS (Annex B) input level from cable (LTE centre frequency 675-859MHz, calculated separation distance 0.5m) 40 CM 1 CM 2 30 CM 3 20 CM 4 10 CM 5 0-10 -20-30 CM 6 CM 7 CM 8-40 675 691 707 723 739 755 771 787 803 819 835 851 256QAM channel centre frequency (MHz) The red 316mW line is the nominal power (CEPT) of the terminal unit. Figure 15: LTE-Radiated European Scenario, CM 1-9 CM 9 LTE (10MHz BW) at 316mW ERP Points below the red line were affected by maximum power levels proposed for LTE terminal units. Graph demonstrates the effect these power levels will have at a distance of 0.5m and shows that the receivers display significant problems on-channel, on both adjacent channels, and on a number of off-channel frequencies. 9.2.3 In- Home Cabling LTE Interference Testing The UK cable operator uses high quality dual, triple or quad screened internal cables (light blue on graph Figure 16) when they install however the customer is not legally obliged to use the operator when extending or changing their installations. A number of tests were carried out on various qualities of domestic cables as the Dutch and other reports give in-home cabling as a major cause of ingress in their countries the indicative results of the Cable Europe testing show: Page 31

LTE ERP (dbm) On-channel On-channel On-channel Test 24: Maximum LTE level permitting stable audio and video on a DVB-T channel (centre frequencies 786, 818 and 850MHz). -60dBm input level to outside receiver; 25m cables. Calculated separation distance 0.5m. 40 30 Pro quad-shield (black) 20 Good domestic (black) 10 Standard domestic (brow n) 0 Low -end RG59/u (black) -10-20 786 818 850 LTE transmit and DVB-T receive centre frequency (MHz) LTE (10MHz BW) at 316mW ERP The red 316mW line is the nominal pow er (CEPT) of the terminal unit. Figure 16: Test 24, LTE-radiated 25m coaxial cables 9.2.4 Summary of LTE Interference with Cable systems Testing demonstrates that the tested combinations of modulation and power levels with a STB or cable modem in an in-home installation are susceptible to on-channel (where the STB uses the same radio channel as the interfering signal) LTE interference at or below the 25dBm Total Radiated Power (TRP) level proposed for LTE in this band. Seven out of eight tests revealed susceptibility on adjacent channels (where the receive frequency is one 8MHz Ultra-High Frequency (UHF) channel offset from the interferer s transmit in addition to various image/if channel interference. 9.3 White Space Interference to Cable modems Whilst much has been written on the potential for white space devices geographically using spectrum not allocated to, or interleaved with TV transmissions in that location, the technical characteristics are still unclear, and work is underway within CEPT to determine the technical parameters which will be put forward for their use. Page 32

Suggested uses have ranged from a full coverage Wi-Fi system to provide broadband access, to individual in home wireless media extensions. Irrespective of the use, the interference potential and mechanisms will need to be carefully considered. To date the mitigation techniques proposed have been based on the high power television transmitter network. Two main options are being discussed: A geographical database which the device s inbuilt GPS receiver refers to; or A listen before talk (LBT) circuit within the device. Whilst these mitigation techniques may work for main TV transmitters with high field strength other users such as communal aerial systems and Cable Networks, will not be protected. Dependant on the density of use and output power of white space devices, these systems will suffer similar on channel interference to that from ECN s. Further information: http://www.ofcom.org.uk/consumer/2009/11/wireless-waves-and-white-space technology/ 9.4 Cable Interference - Possible mitigation and solutions A number of mitigation techniques have been suggested: 9.4.1 Restrict cable transmissions to below 790 MHz. Whilst this would seem to provide an instant answer, unfortunately it has a range of practical objections: The cable networks already use spectrum above 790 MHz; Requirements for higher speed broadband will need up to 32 bonded 8 MHz channels; Transmission of High Definition and interactive TV is spectrum hungry; 3DTV is just starting to come onto the market, and again is spectrum hungry; This will not provide an answer for the interference expected in the next stage of DD clearance to 600MHz or the use of white space devices (and indeed could compound the issues); Page 33

Costs of reconfiguring the networks and CPE to below 790MHz would be prohibitive, and the loss of existing and future revenue to enable competing technologies and delivery mechanisms appears to be anti-competitive. 9.4.2 Reduce QAM rates to make the system more robust: Testing demonstrates that the tested combinations of modulation and power levels with a STB in an in-home installation are susceptible to on-channel (where the STB uses the same radio channel as the interfering signal) LTE interference at or below the 25dBm Total Radiated Power (TRP) level proposed for LTE in this band. Seven out of eight tests revealed susceptibility on adjacent channels (where the receive frequency is one 8MHz Ultra-High Frequency (UHF) channel offset from the interferer s transmit frequency), and six out of eight revealed wider susceptibility to in-band interference. Amendments to the configuration revealed the following: A regression to a more robust form of modulation 64QAM from 256QAM provides a marginal (between approximately 0 and 5.4dB) improvement in immunity (this is reflected in the German test results); More beneficial is a 10dB increase in input power level (please see 9.4.3), reflecting a change in real-world scenarios from acceptable to excellent input level (providing an improvement of between approximately 7.2 and 22.7dB); A very considerable difference in susceptibility (a maximum of approximately 41.7dB) is also present between the two STB units tested, deliberately chosen as the best case and worst case examples from the selection of CPE we have tested to date. Other implications of converting to lower QAM rates include: A spectrum requirement increase of some 30% to maintain current Cable TV (CATV) and data services; Where the DOCSIS 3 protocol is in use up to some 32 bonded channels may be required in order to obtain maximum data rates, causing the overall spectrum requirement to increase to maintain CATV services; Major re-engineering of the network. 9.4.3 Increase HFC signal level With the wide range of two way signals present at frequencies between 15-862MHz a careful balance of signal levels is necessary to compensate for the transmission losses whilst not overloading amplifiers and especially CPE. Page 34

9.4.4 Adapt DOCSIS protocol to improve immunity to interference This is under review but is a long term solution and will not solve the issue of onchannel interference. 9.4.5 Increase immunity of CPE Again a long term solution, work is underway in the ETSI/CENELEC EMC Joint Working Group on these issues and with individual CPE and components manufacturers. However the short time scales before the introduction of ECNs suggest that new designs will not be commercially available in time. 9.4.6 Connectors There is some evidence from a limited range of testing that barrel connectors provide better immunity than PCB mount connectors (more work is required to clarify issue) Figure 17: Barrel connector Figure 18: PCB mounted connector 9.4.7 Home Cabling improvements Encourage customers to use either the operator or the highest quality coaxial cable for any changes to their installation. 9.4.8 Pilot Tone Page 35

A tracking low level pilot tone transmitter attached to each piece of CPE would serve a number of purposes Cause the ECN terminal unit to avoid that frequency when communicating Provide sufficient level of RF power to be detected by white space devices 9.5 Cable Summary Testing has shown that there is little in the way of filtering or other simple add on s which will dramatically improve the situation in the UK. Time is required for new CPE to be developed, the prime point by a range of agencies for use of ECN s is to provide rural broadband, roll out of ECN s in rural arrears first would achieve Government objectives and allow time for the cable industry to both develop and fit new CPE. 10 Communal Aerial Systems Communal aerial systems are used to deliver television and radio signals received at a single location to a number of users. Typically, they are found in large houses, blocks of flats, anything from about four flats upwards, tower blocks with over 100 flats or housing estates with 1,000 homes, student halls of residence, prisons, hotels, etc. The source of the Communal Aerial Systems section of this document has been supplied by the Confederation of Aerial Industries Ltd and has been used by permission granted via Copsey Communications. 10.1 Communal Aerials Systems and Topologies Communal aerial systems can be partitioned into three categories: MATV (Master Antenna Television) systems, which deliver analogue and digital terrestrial television signals IRS (Integrated Reception Systems), which deliver satellite services as well as terrestrial television SMATV (Satellite Master Antenna Television.), similar to MATV with the addition of locally modulated signals, commonly satellite programmes. An IRS can be regarded as an MATV system with additional components to deal with satellite signals. MATV systems have two principal parts: Page 36

The head end, in which there is at least one aerial to receive signals off-air and often more than one to receive radio, and a launch amplifier to overcome the loss of signal power in the distribution network The distribution network, which is an arrangement of cables and splitters designed to carry signals from the launch amplifier to outlets in each dwelling. MATV head ends can themselves also be partitioned into two categories: Wideband, in which the head-end passes all signals from the frequency band used for terrestrial television Processed, in which frequency selective components are used, so that the system does not pass all terrestrial television frequencies. These components include equalisers and channel amplifiers, designed to apply different amounts of gain to different channels, and channel changers, designed to move a signal from one channel to another. As a rule, small MATV systems tend to have wideband headends, and large systems will have processed head-ends, although this varies somewhat geographically and with the age of the system The antenna array receives off-air terrestrial broadcast signals from both the main (i.e. local) television transmission station and, in many cases, transmission stations from an adjacent area. Signals range from the FM broadcast band through DAB to analogue and digital television signals. Many systems also have a satellite reception capability, receiving either Freesat channels or, depending on license arrangements a combination of Freesat and subscription channels. In addition to the signals mentioned above, some systems will also have the introduction of on-site and locally produced programmes including security cameras modulated into free channels in bands 4 and 5. The addition of satellite to the system comprises of a dish complete with LNB. The LNB has four outputs being low and high band with two polarities for each being horizontal and vertical. Each of the cables carries an IF range of signals between 950 and 2150 MHz. The head end amplifiers enable these four sets of signals to be distributed around the system as well as the terrestrial. Each distribution point has a satellite switch and connects via a single cable with an outlet plate in the flat where commands from the satellite decoder cause the switch to deliver the required polarity. A system such as this with one satellite being distributed is known as a 5 cable system. The addition of a further satellite would result in a system with 9 cables or 3 satellites, 13 cables and 4 satellites, 17 cables. The back bone of the system run between the head end and each switch and repeater amplifier is therefore that number of cables. Distribution components are readily available for up to 17 cable back bones. Page 37

Systems delivering services such as Sky plus and HD would have a minimum of two cables feeding into the flat. In the case of retro fitting to provide the extra feed for Sky plus etc., a converter called a stacker may be used. A stacker will up convert one satellite output from the switch and add this to another output. In the flat, a down converter will separate out the two sets of signals. Stackers result in the distributed bandwidth increasing locally to the flat to between 88 and 3500 MHz. The interference immunity of the head-end varies widely dependant on age and installation design. The best systems will have channelized filtering to prevent interference from other transmissions. Other systems will have the terrestrial aerial connected directly to a broadband amplifier which may have a bandwidth as wide as 30-862MHz, with gains of up to 40dB or more. Distribution may be achieved through a variety of methods, the most common being star distribution and cascade. The selection of distribution will depend on the geography and size of the site concerned. Distribution of the signal throughout the building or buildings will be achieved by wideband repeater amplifiers, normally without any filtering. The gain of these amplifiers will vary by up to 40dB, dependant on the cable losses incurred up to that point. Termination in customer premises will normally consist of an unfiltered plate which may have two or more sockets diplexing FM radio, DAB and UHF television. Cabling of these systems will vary widely from multi-screened cables to single screen. Wall sockets will also vary, the older systems having unscreened outlet and modern systems complying with EMC requirements. The link cables (fly lead) between the wall socket and customer equipment are normally provided by the customer. 10.2 Communal Aerial Interference Issues Many communal aerial systems located in major conurbations will be on high tower blocks where the head end may well be co-located with other radio services including GSM, 3G and in the future LTE systems. With the physical co-location of LTE base stations and communal aerial systems, the amplifiers used in a communal aerial system are likely to receive direct breakthrough from the LTE signal into the wideband amplifiers. Once in the system, the LTE signal may well overload the amplifier or, at minimum, cause interference to any co-located channels or adjacent channels. The interference effect will depend on the bandwidth of the interfering signal and the amount of out-of-band energy produced by the base station. Page 38

10.3 Communal Aerial Customer base It is difficult to determine exact numbers of viewers as there is no single authority; communal aerial systems may be provided by the building owner, local council or a local company. However for the UK the following figures are the best estimate from the Confederation of Aerial Industries (CAI): There are in the order of 750,000 MATV / SMATV / IRS systems serving 5-6 million homes. (Source: Mr D Hodges of the Confederation of Aerial Industries Ltd) 10.4 Communal Aerial Interference Assessment From initial research, it is reasonable to state that interference will occur and is due to the physical proximity of the base station to the head-end equipment and to the wideband nature of amplifiers used for distribution of the signal within a building. Once the signal has entered the distribution chain it is unlikely that conventional RF filtering will resolve the interference issues. Cabling of communal aerial systems is unlikely to be generally accessible as, in some cases, it may have been put into conduit or, more likely, built into the fabric during the construction of the building and therefore replacement with better screened cables is unlikely to be economically viable. For the future, greater consideration of the cabling needs to be made. Communal aerial systems and viewers will also be vulnerable to any other changes in the environment brought about by additional services being introduced in the television bands. These may include: White space devices either due to non-local television signals being distributed throughout the system or for breakthrough into local wideband distribution amplifiers New mobile or similar services introduced in the second phase of clearance above 600MHz The proposed increases in terrestrial television transmission powers may also have an effect on these systems as a minimum rebalancing of the network would probably be required. Page 39

11 Programme Making and Special Events (PMSE) Programme Making and Special Events (PMSE) is the only incumbent user of the UHF white spaces in the UK. The various industries that collectively make up the PMSE sector are responsible for delivering the content for professional distribution. 11.1 PMSE Background PMSE or 'the entertainment industry' in the widest sense, has been using the 'white space' spectrum in the UHF bands for more than 30 years, the BBC for example have been using it since at least 1978 4. There has been a steady increase in demand for wireless devices across all sectors of this mature and well established industry over this time. In international spectrum regulatory circles the terms Services Ancillary to Broadcasting (SAB) and Services Ancillary to Production (SAP) are more commonly used to refer to the types of radio services in question. In the UK this use has been managed and licensed by various means over the last three decades. Instances of interference to UHF TV reception, analogue or digital, from PMSE, although a theoretical possibility, are unheard of. Since 1997 a contracted agent, JFMG Ltd. 5, who receive licence fees and issue licences to individuals and organisations on behalf of Ofcom, has managed the PMSE sectors use of the radio spectrum in the UK. At the same time the amount of UHF white space spectrum available to the industry has been reducing as terrestrial broadcasting has evolved; the introduction of Channel 4 (1982), then FIVE (1997) and then six multiplexes of DTV (1999) operating alongside five analogue channels for the last decade, have all resulted in dramatic reductions in the amount of available clear, white space or interleaved UHF spectrum available to PMSE. The, Digital Dividend. is yet another significant reduction in usable spectrum for PMSE, yet demand for, and use of, the UHF wireless technologies which are tools of the content production trade continues to increase. 11.2 UHF white space spectrum use by PMSE Use of the UHF white space spectrum by PMSE is not confined to the UK and similar activity can be found in most parts of the world, though the levels of formality and regulation vary widely. Hence there is a global market for many forms of PMSE wireless 4 http://downloads.bbc.co.uk/rd/pubs/archive/pdffiles/engineering/bbc_engineering_111.pdf 5 http://www.jfmg.co.uk/pages/about/about.htm Page 40

equipment. Even so, the global market volumes are still comparatively small when compared to consumer devices such as cellular handsets. Since 1994 the UK PMSE sector has been fortunate to have exclusive, UK wide, access to Channel 69 (854MHz to 862MHz). This has lead many people to assume, incorrectly, that the entire needs of the PMSE sector in the UK have been taken care of and, that by extension, simply replacing this single 8MHz band of UHF spectrum with another 8MHz band of UHF spectrum (channel 38) will satisfy the needs of the industry. Whilst channel 69 has been very important to certain sections of the PMSE community it is by no means sufficient to support the spectrum needs of many of the UK s most high profile and best loved entertainment and sporting events. Consequently whilst its replacement by Channel 38 (606MHz to 614MHz) is essential to the day-to-day activities of the PMSE sector on its own it is totally insufficient to support many of the programmes and events that people now take for granted. For example: A typical musical theatre production in London s West End, or on tour, expects to use an average of forty radio microphones. A single isolated 8MHz band can only support between eight and sixteen radio microphones; the solution lies in the use of the interleaved or white space spectrum. A reality TV talent show such as the X Factor uses between eighty and ninety radio microphone frequencies. Clearly this is significantly more than can be accommodated in a single 8MHz band; the solution is to use the interleaved or white space spectrum. Sporting events such as Cheltenham Gold Cup, Formula One, The Open, Wimbledon as well as televised Football, Rugby and Cricket all rely on the use of UHF white space spectrum. Conference events, for example the Police Federation conference, frequently use in excess of sixty wireless microphones and IEM frequencies. This can only be achieved by using, interleaved or white space spectrum. The technology used by the PMSE industry in the UHF white spaces is almost entirely low powered (10 50mW ERP) wide band FM. Equipment development has historically focused mainly on improving performance and reliability as well as on reducing the size, weight and power consumption of portable devices. Higher power frequency allocations in the UHF interleaved spectrum up to as much as 25W can be arranged but are comparatively rare. Page 41

11.3 PMSE devices and applications The most common examples are the class of devices known as radio microphones or wireless microphones. Essentially a short range high quality unidirectional audio link conveying a single audio channel, these devices take several forms to suit the varying needs of different aspects of the entertainment industries, ranging from the invisible as required by film & TV drama productions to the (sometimes highly) visible hand held devices used for live performances by some vocal artistes. It is vital to understand that these devices are used at the Figure 19: Example Microphones start of the audio chain and any interference to them will be experienced by the entire audience regardless of whether that is just the people in one room or worldwide radio and TV audiences in the case of a live international event. Other PMSE devices using the UHF white spaces include In Ear Monitors (IEM) (also known as Personal Monitors ) and radio talkback systems. Like radio microphones IEM s are short range high quality unidirectional audio links but they are capable of conveying stereo, or two channels of, audio. Whereas a radio microphone conveys the audio from a performer to the mixing console(s) or recording device(s) IEM s are used in the reverse direction. IEM s are used extensively in live music performance to help enable performers to stay in time and in tune and also by TV presenters to enable them to receive cues or instructions. They are also used in location film and TV production. Radio talkback systems in the UHF white spaces provide bi-directional full duplex audio communication. They are used to coordinate crew and out of vision activities, often in situations where clear uninterrupted and timely communication are vital not only to the smooth running of an event or production but for the health and safety of performers, crew and the public. The audio fidelity requirements are not as onerous as for radio mics and IEM s however the reliability, dependability and timeliness of such systems is extremely important. 11.4 PMSE Ch61 69 clearance compliance problem The proximity of channel 69 to two other areas of spectrum which can currently be used for certain types of PMSE activity may present a problem for any future new users of the 800MHz band. Radio microphones and IEM s manufactured in the last ten years or so typically have a tuning range of 24MHz or more. This means that equipment that can operate in channel 69 can also often operate in TV channels 67 and 68 and also in the Page 42

licensed exempt band between 863MHz and 865MHz. It is believed that many radio microphones and IEM s are being operated in spectrum between channels 66 and 69 without appropriate licenses. In the absence of a license it could be difficult to ensure that a user complies with the requirement to vacate the 800MHz cleared spectrum. Since much of the same equipment can still legally be operated in the licensed exempt band between 863MHz and 865MHz after the sale of the 800MHz band possession alone will be insufficient proof of an offence. Since the upper part of the 800MHz band will most likely be used as uplink frequencies from handsets to base stations in any FDD system interference to existing PMSE equipment which continues to be used illegally is less likely in the short term than in the lower part of the band. Even equipment which is currently being operated legally in the spectrum between channels 61 and 69 could be sold on by the current owners and could therefore continue to present an interference problem if suitable arrangements are not made to compensate current owners for the loss of utility caused by the sale of the 800MHz band and ensure that the equipment is removed from use. 11.5 PMSE technical requirements: Highest Audio Quality demands, serve better than studio/cd quality 44 ksa/s,16 bit. Lowest latency < 3ms roundtrip (microphone to mixing console to In Ear Monitor) No interruptions, availability 100% of time Calculation: Mic Audio SNR: 100 db (orchestra even does 140 db) Compander gain: 40 db (analogue, no significant latency!) FM process gain 10 db (20 khz Audio on 200 khz channel) Results in 50 db C/I on RF channel Actual Technology: Analogue transmission, digital source coding would cause too much latency FM modulation Constant Envelope provides long operation time for wireless microphones and body pack instrument transmitters Proprietary digital systems have recently entered the market. Page 43

Comparing PMSE and Cellular technical requirements, see table below. Table 7: Comparing PMSE with Cellular PMSE Cellular Audio Quality Highest for content production Only speech Analogue CD quality used for Audio Rate comparison purposes: 44 ksa/s, 16 8 ksa/s, 13 bit bit 704 kbit/s. In some cases higher 104 kbit/s than CD quality is used. Audio Rate 352 kbit/s 12 kbit/s Compression Analogue Digital source Comp Audio Rate 352kbit/s 12kbits/s Compression 15 channels in 20 MHz 75 channels in 5 MHz Raw Audio related spectral efficiency 0.5 bit/s/hz 1.56 bit/s/hz Compressed Audio related spectral 0.25 bit/s/hz 0.18 bit/s/hz Transmission Analogue FM Digital GMSK...128QAM Crest 0dB 14 db with OFDM Interruptions None Short e.g. 20 ms RRM Mobility Fixed power, fixed frequency, fixed modulation 54 km/s e.g. Starlight Express Resource allocation Power control, Handover, adaptive modulation and coding 250 km/h e.g. GSM, UMTS 11.6 PMSE Digital radio microphones The belief that digital technology will somehow alleviate the pressures of spectrum scarcity for the PMSE sector in anything but the very long term is mistaken at best. Current digital radio microphone products have so far failed to deliver proven improvements in spectral efficiency added to which the power consumption, and therefore battery consumption, of many digital devices is considerably greater than that of current analogue ones. Latency in digital audio transmission is also a barrier to acceptance in many applications; remember here that we are talking about a need for latency figures of (much) less than 3mS. 11.7 Implications of new entrants to PMSE The potential for interference to PMSE equipment from other future white space devices is a very serious concern to existing PMSE users. Secondary use of the UHF white space by PMSE has been successful over the last three or more decades without Page 44

causing interference to the primary service, TV broadcasting, precisely because the risk of interference to PMSE is taken very seriously by users. The reciprocal nature of interference has meant that because of PMSE users need to avoid interference from terrestrial TV broadcasting, even at levels which are so low that they might be deemed acceptable for many other forms of radio service, the risk of interference to terrestrial TV reception is automatically eliminated in many cases. Add to this a licensing regime and the fact that PMSE and terrestrial TV frequently represent the two ends of the same broadcast chain and it is easy to see why the two services are able to coexist in the same spectrum. Other services entering the white spaces may be more tolerant of interference from terrestrial TV than PMSE because, for example, they may be able tolerate a lower grade of service, service interruptions or high levels of latency. By extension therefore these new services will not have the same inherent interest in avoiding interference to the incumbent services. PMSE and terrestrial TV broadcasting are complementary services, other future services and devices may be competing with each other and with broadcasting both commercially and for the attention of viewers. 11.8 PMSE Solutions and recommendations No solutions to the interference risks posed by new use of the UHF white spaces by potential new licence exempt services or devices to existing PMSE services are apparent. Other new licensed uses of the UHF white spaces whilst representing the 'thin end of the wedge' for PMSE do at least have the potential to be managed in such a manner that the essential requirements of the PMSE services are not impacted and to ensure that peak PMSE demand can still be met as and when required. PMSE must be given pre-emptive priority in any UHF white space frequency allocation database system. All services and devices using the UHF white spaces must employ best RF practices in order to maximise the benefit to users and to avoid any interference impact on all other services. Page 45

12 White Space TV White space (TVWS) technology works by using unoccupied radio channels called white spaces within the TV bands from 470MHz to 790MHz. In recent times, the concept of spectrum white space has emerged, bringing with it the idea that, while usable spectrum is almost completely allocated to licensed users, only a fraction of it is actually in use at any given location and time. This means that there is a lot of empty, or white space, spectrum potentially available for use by other, licensed-exempt services. TV White space (TVWS) technology attempts to exploit this concept by using unoccupied TV channels within the TV band. Following several workshops, of industry participants investigating future usage, held in the US and the UK (DCKTN White Space events), the two beachhead uses for whitespace are likely to be wireless broadband and home media distribution. Whitespace benefits from the better propagation characteristics of TV frequencies relative to frequencies at 2.4 GHz, so making it a possible solution for public Wi-Fi-like systems. Meanwhile, the use of whitespace for home media distribution offers a convenient and robust alternative to the installation of new wiring. The idea of exploiting this underutilised licensed spectrum has broad commercial appeal, but clearly presents a challenge to ensure the licensed-exempt white spaces users do not interfere with or block licensed spectrum users. However, regulators in the US have laid down regulations that aim to provide just such protection, and the UK regulator is considering very similar arrangements for the UK. Concepts from cognitive radio and location based services are being investigated to find a low cost solution that both satisfies the regulatory requirements and enables the creation of a consumer market. Two different approaches have been identified: geolocation and spectrum sensing. Spectrum sensing would allow a self-contained solution and is therefore an attractive approach. Recent studies have concluded that spectrum sensing requires the detection of potential victim transmissions at levels several magnitudes lower than required for reception. This poses a considerable challenge for the whitespace receiver and, while progress has been made with detection of DVB-T transmissions at low levels, development is ongoing. Geo-location requires the position of the whitespace device to be determined to within 50 metres and an on-line database to be consulted to identify available frequencies in the area. GPS is capable of providing the necessary resolution although indoor use is challenging. It also represents extra hardware costs for the whitespace device as, potentially, does the interface to the on-line database. Nevertheless, despite these Page 46

challenges, it is considered that the geo-location approach is capable of meeting the regulatory requirements and can be implemented with current technology. The presence of a database also opens the possibility of new business models, for example, aggregating spectrum access with other location-based services. 12.1 White Space Applications Applications in TVWS benefit from the superior propagation characteristics of the band, when compared to other bands such as the ISM band. As with standards in the ISM band (e.g., IEEE 802.11 family), it has the potential to support a wide range of applications and we can expect further development in this area as people learn what the technology is capable of. Two applications of immediate interest are home media distribution and wireless broadband. 12.1.1 Home media distribution With the proliferation of digital media sources, including internet streaming, digital video recorders, residential network storage systems, DVD/Bluray players and so on, there is an increasing demand to be able to route data from anywhere to anywhere in the house. This can be, and is, already done with existing WiFi networks but they will struggle to provide the bandwidth and quality of service for high definition streams. TVWS technology, with its ability to penetrate walls, is an attractive solution. TVWS is not the first to address this need, however and other standards are being proposed. TVWS will need to be able to show some clear benefits if it is to compete. 12.1.2 Wireless broadband With governments around the world announcing initiatives to provide universal broadband coverage, wireless broadband becomes an interesting component technology. We might usefully differentiate between two classes of wireless broadband. In one class, wireless broadband technology provides the last mile between network and home. The actual distance covered may vary in practice from up to a hundred metres or so, to a few km in very rural areas. These rural broadband systems are generally expected to be fixed, planned systems and so may differ somewhat in their requirements for cognitive features. In the second class, TVWS technology could be seen as an alternative to urban WiFi networks, which often have limited coverage. Wireless broadband based on TVWS might provide more reliable coverage and anyone could own the cell, be it a coffee shop or a private individual. TVWS technology may also complement cellular systems in rural areas. Cellular service in rural areas can be scarce is because the costs of providing coverage at cellular Page 47

frequencies is too prohibitive for operators to take an interest. Self-help systems using lower frequencies may be the quickest way for rural users to get on-line. 12.2 White Space Regulatory and Standards background Regulatory and standardisation activity began in the US in 2003 12.2.1 FCC In the US, the FCC issued a first Notice of Inquiry on the subject of TV white space in 2003, following up with a Notice of Proposed Rulemaking (NPRM) in 2004. Over the intervening time there has been much discussion between the FCC and the various stakeholders, culminating in several documents, which sets out the FCC s current position. FCC Document FCC 08-260 6 proposes basic system parameters for TVWS devices that aim to provide a viable opportunity for TVWS while causing minimum interference to incumbent services. This includes parameters for both sensing and geolocation methods. 12.2.2 IEEE Within the IEEE framework, there are a number of different activities. The IEEE 802.22 working group on Wireless Regional Area Networks (WRAN) was established as a response to the FCC s NPRM and aims to provide a standard for fixed point-tomultipoint WRANs using the TV band. The IEEE 802.11 working group has begun work on an amendment to 802.11 that will allow WLAN operation in the TV band. More recently, the IEEE 802.19 Wireless Co-existence Working Group has begun development of a coexistence standard for wireless devices operating in the TV bands. 12.2.3 CogNeA and ECMA The Cognitive Networking Alliance (CogNeA) was announced at the end of 2008 with the intention of developing a standard for high speed networking in the TV band. In March 2009, an early version of the specification(v0.8) was transferred to ECMA committee TC48-TG1, which further developed it and published a first edition in December 2009 as ECMA-392 7. ECMA-392 specifies a PHY layer and a MAC sublayer, both optimised for TVWS use. In particular, the standard provides protocols for the protection of incumbent users, transmitter power control, and supports data and network security. 6 FCC 08-260 http://hraunfoss.fcc.gov/edocs_public/attachmatch/fcc-08-260a1.pdf 7 ECMA-392 http://www.ecma-international.org/publications/files/ecma-st/ecma-392.pdf Page 48

12.2.4 Ofcom In the UK, Ofcom has been managing the TV digital switchover (DSO) since 2005. An important aspect of that has been the consideration of uses of interleaved, or shared, spectrum within the digital TV spectrum post-dso. In February 2009, Ofcom published a consultation document on proposed parameters for licensed-exempt cognitive devices using the digital TV spectrum 8, following this up with a second document in November 2009 on geo-location 9. 12.3 White Space Spectrum availability The following chart describes the proposed use of the former UHF analogue TV band. A number of channels from the analogue era will be lost to other services while 32 channels will remain for digital TV usage and also for other, licensed exempt applications on a geographically-interleaved basis. Figure 20- Interleaved Spectrum Availability (Source: Ofcom 10 ) The greater efficiency and tolerance of the digital modulation modes means that six or more digital multiplexes can be located in less spectrum than the four or five analogue channels that were available previously. However, the reduced bandwidth means less room for TVWS devices, especially when taking into account the needs of PMSE users, who are also required to adapt to changed circumstances. 8 Digital Dividend: Cognitive Access http://www.ofcom.org.uk/consult/condocs/cognitive/ 9 Geo-location for Cognitive Access http://www.ofcom.org.uk/consult/condocs/cogaccess/ 10 Digital dividend: 600 MHz band and geographic interleaved spectrum http://www.ofcom.org.uk/consult/condocs/600mhz_geographic/ Page 49

How much spectrum is available in any given area depends on a number of factors and estimates therefore vary. Available channels will vary from region to region and, in some regions, more spectrum will be available than in others. One study suggests that, on average, around 100 MHz may be available for TVWS but notes that this varies considerably with location. Other studies note that the amount of spectrum that could be available depends very much on such factors as the level of protection that needs to be applied to adjacent channel, itself a matter of some debate. The power transmitted by white space devices will also have an impact not only on the active cell size but also the zone of potential interference. In some circumstances, particularly when considering adjacent channel operation or operation near other low power systems, like PMSE equipment, the ability to use a lower transmit power may have a significant influence on spectral availability. In general, there is still work to be done to better understand and quantify spectral availability in the new digital TV bands and it may be that field trials are the best way of resolving real availability. Nevertheless, most studies agree that there will be room for TVWS services. 12.4 White Space Interference challenge As potential incoming users of spectrum already used by incumbents, white space devices need to ensure that the channels they use are genuinely white space channels and, thus, that their transmissions will not cause interference to other services. Existing licensed users of this spectrum are entitled to protection and the onus is on the newcomers to provide it. Two approaches are generally discussed: Spectrum sensing Geo-location database. A third approach is emerging looking at Propagation Modelling. 12.4.1 Interference to incumbent services The existing users of the spectrum in question fall into three camps: Broadcast Television Digital TV (470-790MHz) Programme Making and Special Events (PMSE) Cable Television and home media consumer platforms (5-862MHz) In the UK (and Europe), digital terrestrial TV services are broadcast, usually at high power, using DVB-T or DVB-T2. The remaining analogue transmissions will cease in the near future. Licensed PMSE systems are able to operate at fixed locations on a Page 50

locally-free TV channel and this is expected to continue during and beyond digital switchover. PMSE users are arguably the incumbent white space users because their allocations are made to fit around the TV transmissions available in the area. Both have the potential to suffer from interference from other white space transmissions. Cable TV networks are not users of over-the-air spectrum but might suffer from ingress from nearby transmitters. This problem is a special case and is discussed separately below. 12.5 White Space topologies In general, a whitespace device may only transmit on a channel it has good reason to consider free. ECMA-392 describes two types of network: peer-to-peer and masterslave. A master device is assumed to have the necessary tools required to evaluate channel availability, whatever those tools are. A slave device is not assumed to have such tools and must therefore act on the guidance of a master device. That is, it cannot independently transmit but can only do so on the instruction of a master device. ECMA-392 also makes provision for peer devices operating in a peer-to-peer network. Amongst other things, peer devices can share information on spectral availability, can be otherwise un-tethered and mobile. 12.6 Spectrum sensing A classic method of avoiding interference to other spectrum users is to listen on the channel before sending, sometimes called Listen before talk (LTB). If an incumbent signal is detected above a certain threshold, the transmission is suppressed and steps taken to find an alternative channel or an alternative time to transmit. This principle is already embodied in varying forms in existing standards and coexistence guidelines and is a major component of cognitive radio systems. The challenge with all such systems is that it is the incumbent receiver that would suffer interference as the new device is only able to detect the transmitter. This leads to the classic hidden node 11 problem in which a device is unable to hear a transmitter that is hidden by distance, obstructions or other causes of attenuation and thus inadvertently causes interference at a nearby receiver that is listening to the distant transmitter. This has been studied for the case of digital television with the result that a detection threshold has been proposed that is considerably less than the level required to successfully receive a DVB-T signal. Nevertheless recent work has shown that it is possible already to detect (i.e. sense) a DVB-T signal at levels close to the specified 11 Randhawa, B.S., Wang, Z., Parker, I., Analysis of hidden node margins for cognitive radio devices potentially using DTT and PMSE spectrum, ERA Technology Limited, ERA report 2009-0011, January 2009, http://www.ofcom.org.uk/radiocomms/ddr/documents/eracog Page 51

threshold. Sensing at levels down to the threshold might also be viable with improvements in technology. Some PMSE operations can impose special requirements on PMSE devices that can be most easily met by analogue FM transmission techniques. In particular, the requirement for essentially zero latency is difficult to meet with economical digital techniques and so, unlike broadcast TV, analogue PMSE devices are likely to remain available for some time to come. This means that sensing a PMSE transmission can mean detecting an FM signal. Recent work shows that this is indeed possible but performance against specified thresholds is still to be studied. It should be pointed out that while PMSE devices are similar in their transmission characteristics and usage in the US, UK and rest of Europe, the PMSE sensing parameter proposed by the UK regulator is an order of magnitude lower than in the other regions (12 db lower than US, 10 db to 7 db lower than being discussed in SE- 43). Harmonising the PMSE sensing threshold with other regions will benefit white space users in the UK due to the availability of cognitive radio based applications such as peer-to-peer, high mobility and wholly indoor. 12.7 Geo-location database The fact that broadcast TV transmitters are static and well-documented and that PMSE usage is recorded on a database leads to the idea that, by knowing one s location, one can find out which channels are in use locally and which ones are free to use. This breaks down into two distinct questions: how should one discover one s location and how should one consult the database. The current assumption is that location can be found to a sufficient precision of around 50 metres via GPS. GPS is well-established but has known difficulties in indoor scenarios but it is thought that improved technology may help in some cases. If GPS is not sufficient, a number of alternative strategies are possible, including use of the last known position (assumed acquired via GPS outside the present indoor location) and the use of other radio-based positioning technologies. Discussion of database usage tends to assume two important preconditions: the existence of a database and the availability of a connection to the database. For wireless broadband applications, the availability of a suitable internet connection can probably be taken for granted. For home networking applications, this might not always be true and provision of a suitable internet interface would represent an extra cost for devices that do not require one as part of their normal operation. As for the existence or otherwise of a database, the current assumption is that at least one commercial third party would be charged with maintaining and running a database though it remains to be seen who would run such a database and what sort of business model they would use. Page 52

There has been some discussion about how a database might operate, with Ofcom providing a public consultation document in late 2009. In a suggested approach, the database responds to a request with a list of available channels, maximum power and, possibly, maximum duration. Ofcom has suggested that the white space device updates this information at least once every two hours, which suggests that the database itself needs to be updated at least as frequently. It needs to be decided from whence the necessary information is obtained and the level of geographical resolution required. The need to consult the database so frequently also needs to be understood, as does the amount of data traffic that will be generated as a consequence. If operation of a white space is dependent on database access then both it and the link to it have to be reliable. In particular, failure of the database will render all whitespace devices unusable. In the U.S., Google proposes that the FCC designate Google to be an administrator of a TV bands geo-location database. Google proposal 12 on Clearing House Model outlines the provision of a TVWS device database manager offering repository services supporting the successful operation of one or more TVWS geo-location databases. 12.8 Propagation modelling A recent presentation given by Steepest Ascent 13 at the DCKTN White Space event (April 2010) described the link planning undertaken as part of the work on installing a wireless broadband test bed in a rural area in Scotland. The links described were relatively long and this example notably included modelling of UHF propagation over tidal water, a notorious cause of reception difficulties in conventional TV work. Given that this was a fixed, planned network, the question arises whether the cognitive approaches of spectrum sensing and geo-location also have a role to play in these systems. From a regulatory point of view, it is also not clear whether such systems will, in general, need to be licensed or licensed-exempt. 12.9 White Space and Cable TV networks Cable television networks use frequencies that cover the digital TV spectrum but they are not formally considered to be users of the spectrum because their transmissions are confined to cable networks and are not sent over the air. Nevertheless, as studies described elsewhere in this report show, there is a potential interference issue due to ingress of nearby white space transmissions into the consumer premises equipment. 12 Proposal by Google Inc. To provide a TV band device database management solution. http://www.scribd.com/doc/24784912/01-04-10-google-white-spaces-database-proposal 13 Steepest Ascent presentation, Prof Bob Stewart, DCKTN White Space workshop 28 th April 2010. https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=737699&folderid=865015&name=dlfe- 7012.pdf Page 53

Owing to the special requirements of cable usage, reassigning spectrum is not possible and, because of the full spectrum usage, it would be impossible for a white space device to avoid active cable frequencies even if cable signals could be detected. However, current American usage has not uncovered a major issue and it remains to be seen how serious an issue it is in practice in the UK. Potential mitigation techniques include better screening of consumer equipment, use of more robust DVB-C modulation modes and increasing the power of signals delivered to the cable receiver. Although there is still work to do, white space and cable networks need not be incompatible with each other. 12.10 Can White Space Leverage other Technologies? Another possible technique that could be investigated is using the white space in a combination with other existing services, potentially moving the Scanning/Sensing Processing away from the White Space device. 12.10.1 Scanning networks co-located with Mobile Infrastructure This concept proposes using a geographic network of ultra-sensitive scanning receivers mounted at elevated positions and using high gain, directional and/or beam scanning antennas. The receivers will have many db s of additional sensing gain over the case of sensing at the terminal. The height gain (e.g. say at 20m) could provide 10-15dB of gain, the use of directional and/or scanning antennas should provide a further 15dB or more gain, and the fact the sensing is done outdoors (many White Space devices may be indoors) will provide a handful more db s of gain. Additionally, as the scanning would be done at a fixed infrastructure location this can benefit from using more sensitive receivers, and being elevated above clutter helps to remove the hidden node problem. The scanning function is carried out across the network of sensors and communicates the available spectrum information to devices within the local area; potentially from the same network. The use of cellular base station site s infrastructure could be ideal as this would provide high density scanning of the spectrum. Many variations are possible such as coordinated beam scanning between sensing sites as to increase sensitivity and also determine location of localised White Space spectrum usage. In effect, such a network can create a spectral usage map across the country of White Space devices and incumbent users. This permits low-cost White Space devices to be introduced, and if delivered from the cellular infrastructure, then cellular operators themselves could in fact exploit the White Space spectrum to augment spectral capacity, perhaps in the form of future LTE spectrum plans. Nonetheless, it may be worthwhile to study and research such co-operative sensing and scanning techniques listed in sections 12.10.3 and 12.10.4. Page 54

12.10.2 White Space downlink capacity using LTE MBMS mode Another option to consider in co-operative mode with a Mobile Cellular network; is to use the White Space spectrum not being used in a particular area as incremental downlink capacity to provide extra bandwidth for a licensed Mobile Cellular operator. The LTE standard includes a detailed specification for a new generation of Multimedia Broadcast Multicast Service (MBMS), to be deployed into existing cellular radio spectrum and/or potentially into future allocations. Therefore the network could actually work in both conventional (uni-casting) and MBMS modes. The figure below illustrates the concept presented at the DCKTN White Space Event in April 2010 by Ericsson. BC Tele White spaces LTE-MBMS low tower low power Mobile broadband LTE/HSPA/GSM low tower low power Content Aggregated / packaged content Can be locally produced, or from any provider in any country and distributed in many networks DVB network high tower High power Figure 21 - LTE, White Space and Broadcast - The 'New TV Collaborative Landscape' (Source: Ericsson) 14 14 Ericsson Presentation by Lasse Wieweg at the DCKTN White Space Workshop 28 th April 2010. https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=737699&folderid=865015&name=dlfe- 7007.pdf Page 55

The LTE specification supports Single frequency network mode of to enable efficient multi-cell transmission for the MBMS services, transmission of identical signals at the same time from multiple cells; seen as one transmission by terminal devices. The network allocates radio resources in all base stations to transmit MBMS packets over the air, including time / frequency resources, modulation, coding etc. Content synchronization is ensured between base stations by means of a Broadcast Multicast Service Centre. The concept could operate under licensed conditions with the frequency arrangement planned in relation to an existing DVB plan, therefore negating the need for sensing and/or geo-location aware user platforms due to the fixed location of the transmitting base stations. Variable bandwidth schemes could be used to match the various sizes of white spaces. Aggregation of LTE channels could be possible to increase bandwidth and be dynamically assigned based on demand. Interaction, via other bands (for the uplink) would be possible and the user platform would benefit as the radio resource would be re-used as this would be part of the user equipment already. 12.10.3 TV Receivers providing location data This concept may be a longer term initiative for White Space, but potentially promises significant increases in available spectrum usage. White Space engineering in the forms described above (database, cognitive devices, scanning co-operative base stations, crowd-sourcing, etc.) are all attempting to avoid transmitting on a channel which is used for example by TV broadcast, or PMSE device. Regarding the TV broadcast service this means White Space devices can t use co-channel (and some adjacent channel) spectrum across huge areas of geography, i.e. the spectrum served by a TV transmitter. If however, domestic TV receivers (inc. set top boxes for DVB-T and Cable) could in principle also communicate to a database their location and include whether the TV was on, and what TV channels are being consumed then we have the equivalent of a Smart Grid concept but applied to the utility of spectrum, rather than gas, electricity, water, etc. This would then permit the entire UHF spectrum band to be used in an area so long as the area didn t have any active TV sets at that time, and hence consumption of TV broadcast services. Even if some TV receivers were consuming spectrum, this would typically be one 8MHz DVB-T channel at any time and location. If for example White Space devices are to be used across areas of scale of 10 s of metres, such as home video distribution then using this TV receiver location feedback method could exploit the entire UHF spectrum in time, space and frequency at a small scale. Page 56

12.10.4 TV Receivers transmitting beacon for White Spaces Potentially TV Receivers could intentionally transmit spectrum consumption data as a protective measure, rather than the actual TV receiver device having a means of communicating positional and spectrum consumption information as described above, could future TV sets, and set-top DVB and cable boxes instead include a small intentional transmission at a known frequency and of microwatts power which conveys this information to White Space devices? In this mode, the White Space device listens for such intentional transmissions and when it receives such, the White Space device knows that it can t use certain channels because a domestic TV receiver is consuming this spectrum within range. The White Space device can also communicate this information to a database so other White Space devices can benefit (crowd-sourcing). This could be viewed as assisted sensing and analogous to Secondary Surveillance Radar or Electronic Protection Measures in an Electronic Warfare context, etc. Perhaps by a certain date, e.g. 2020 if TV receivers (and PMSE devices) don t have these transmitters then there interference free TV can t be guaranteed. 12.11 White Space - Summary TVWS technology offers an exciting new way of accessing prime spectrum that would otherwise be unavailable for licensed-exempt use. Two applications of immediate interest are home media distribution and wireless broadband, where the favourable propagation properties of TV spectrum compared with GHz frequencies used by other services will prove attractive. The major concern with TVWS is interference to broadcasters, PMSE users and other incumbents and much work is underway in the regulatory and technical arenas to ensure that these services are protected. This paper provides some exciting challenges and opportunities that if solved would provide a significant global opportunity. It is clear from the studies and inputs of this report that this will only happen with joint co-operation with the broadcast network operators, component companies, standards organisations and regulatory bodies. Page 57

13 Public Safety 13.1 Public Safety background Public Safety Organizations, in the UK, Europe and the rest of the world in general are extensive users of TETRA or in the US P25 and variants thereof. In the UK in particular the Emergency Services Police Fire and Ambulance are all tending towards use of the Airwave TETRA Network. Airwave is already rolled out to the Police Forces and supporting first responder services such as the Highways Agency and a rollout programme for the Ambulance and Fire and Rescue Services is underway. 13.2 Key Issues In the Emergency Services Voice is key for on-scene incident management and is also a critical safety requirement particularly for the protection of the Emergency Services personnel in high risk situations. However, increasingly the Wide Area Incident management, for example dispatching First Responders to an incident requires the use of data services in order to provide the maximum information about the incident, maps, and building plans etc to prepare the crews. Interoperability between the various Emergency Services is essential at critical levels of Command as is the ability to share a Common Operations Picture to aid the management and effectiveness of a multi- Service response to an incident. In addition, Resilience requirements such as the heightened threat levels in respect of potential terrorist attack, and the perceived increase in natural disasters, flooding, snow etc has meant that there is an increasing requirement for high bandwidth data services e.g. mobile video for remote threat and incident assessment. However the current generation TETRA is unable to support high bandwidth data services without severely impinging upon the capacity of the network to support critical voice services. 13.3 TETRA 2 Radio and Spectrum TETRA 2 could support higher bandwidth services and the Public Safety agencies have been lobbying at a European Level for more spectrum to be allocated for Public Safety particularly Harmonised frequencies for TETRA 2 Public Safety Data Services and there is potentially a mandate from the European Commission for Harmonisation not necessarily specifying TETRA 2. However, this does not imply that in the UK there will be any funding to implement this whether or not a mandate is issued and the opposite may be the case in that in the past there have been suggestions that the Public Safety organizations should bid against Commercial users and potentially against other Public Safety agencies for their Page 58

spectrum. Whilst this is supposedly seen as a way of making the organisations use their spectrum efficiently, additional spectrum from the proposed Digital Dividend has been temporarily allocated for support of the Olympic Games in 2012 as it perceived that there is insufficient spectrum from current allocations to manage this. This temporary spectrum allocation is then anticipated to be part of a commercial auction process after the Olympics. This does offer the opportunity to determine priorities for more general spectrum release for Emergency Service use as it can be assumed that the UK will host major events and is likely to be subject to major incidents in the future. In addition it remains to be seen whether the existing Airwave network will have the capacity to support the additional users from the Ambulance and Fire and Rescue Service with existing spectrum allocation in the event of a major incident. It is in the National interest to ensure the Emergency Services have access to mobile communications that are fit for purpose and this requires access to the RF spectrum and an infrastructure to deliver the necessary essential services. Clearly these services must be available in crisis or emergency situations so must be resilient and able to meet projected future capacity demands. Whether this needs to be provided through a dedicated infrastructure, as is the case with TETRA, is open for debate as it is possible to conceive of ways to share infrastructure and services with others whilst maintaining the Emergency Services requirements for coverage, security, availability and reliability. This may lead to more efficient use of available spectrum, lower build and operating costs and a quicker rollout whilst perhaps delivering other worthwhile socioeconomic benefits through the provisioning of wireless broadband services to rural areas. 13.4 Cellular Networks as an alternative to TETRA 2 Much discussion is ongoing about the use of the Commercial Cellular Networks as an alternative to TETRA 2 for Public Safety Wireless Broadband Data Services. It would seem to be anticipated by Government that the release of the Digital Dividend band for auction for commercial services will facilitate this, potentially offering scope for rollout of LTE in the 800 MHz band allowing higher bandwidth services with better in-building penetration and therefore greater land-mass coverage. It is unfortunate that current thinking about the terms of the license requirements and the economic situation raises doubt as the viability of this and is in danger of being an opportunity lost even at this early stage. Extensive use of Mobile Cellular devices is in fact already made by the Emergency Services for Voice Services but it is still not regarded by the Cabinet Office as a primary critical communications method from a Resilience perspective because of the perception they have that there are coverage issues and the network is not load protected. This is not the case. The use of Roaming SIMs for example in rural areas Page 59

means that coverage issues are greatly reduced and the Reliability of the Cellular network can be demonstrated to be comparable with the Airwave Network. However one of the issues that must be considered is that many of the services used by the Public Safety organizations are critical uplink services. Mobile Video is taken at the scene of an incident and uploaded to the control room. Remote Telehealth services potentially require a critical uplink path from the patient monitoring equipment to the Triage Centre for emergency situations. Many other low bandwidth M2M services such as sending data from security movement sensors are also uplink services. However the current and future generation Wireless technologies are Asymmetric with a bias towards downlink services. When paired spectrum is allocated the capacity of the uplink should be considered in respect of the dimensioning and spectrum requirements. Alternatively consideration should be given to mixed modulation schemes to provide greater symmetry in the Uplink and Downlink bandwidth capabilities which may mean addressing spectrum allocations in an unconventional way for Public Safety Applications. LTE may facilitate this situation with the fact that it can potentially offer many low bandwidth uplink services although asymmetric in its capacity characteristics. Another issue for Public Safety is QoS Reservation. In the current generation of cellular services critical downlink services could be competing with non-essential entertainment services which could potentially seize the available capacity. The capability for prioritization exists within the 3G specifications but no manufacturer has currently implemented it and the benefit offered by operating at the lower frequencies of increased range and building penetration could actually act against one in making cell division to gain greater capacity to give greater QoS by default is potentially much more difficult. 13.5 Public Safety Summary The dimensioning of Public Safety and other critical services such as Remote Telehealth which have a strong uplink component has to be carefully considered in respect of Spectrum allocation It is possible that a forum of interested parties could be formed to guide and influence suppliers as it would be difficult for Emergency Services to influence the large-scale market suppliers in isolation. A non-competitive group could provide the mechanism for building supplier confidence in the operational requirements and market size especially if a shared infrastructure approach were adopted. Confidence in both these aspects is necessary to secure their commitment to invest. Alignment with the stated needs of other European organisations that already use TETRA is an important consideration as costs to the UK will be higher if the UK market is fragmented and is dissimilar to the rest of Europe. Page 60

14 Applications above 862 MHz Short Range Devices The 863 870 MHz band is allocated on a harmonised basis throughout Europe for use by licensed-exempt Short Range Devices (SRDs). As this band lies just above the mobile transmit sub-band part of the 800 MHz digital dividend spectrum, there will be potential for interference to arise from mobile terminals into SRDs in the adjacent band. The SRD band is sub-divided into three application-specific sub-bands, which are illustrated below. The entire band may also be used by generic SRDs which may include conventional narrow band systems or wideband systems using spread spectrum technology. Figure 22-863MHz to 870MHz Spectrum usage Although licensed-exempt SRDs are expected to operate in a non-protected interference environment and many devices incorporate interference mitigation techniques such as very short duty cycles or listen before transmit protocols, existing standards reflect the current relatively benign adjacent band interference environment, with high power TV transmitters at fixed locations and a limited number of PMSE deployments. Operation of mobile terminals in close proximity to SRD receivers may lead to interference or blocking in some scenarios, particularly at the lower end of the SRD band. Common SRD uses include: Wireless headphones for consumer markets. These will commonly be plugged into the headphone outlet of televisions, radios, Hi-Fi systems, games consoles and similar devices; Consumer radio microphones, used with karaoke systems, camcorders and similar devices; Radio microphones for semi-professional markets; Radio microphones for professional markets; Portable wireless microphone and receiver for professional markets; Butt transmitters, these are plug-on units for existing audio microphones; In Ear Monitoring systems (IEM) enable performers to hear the combined audio output of other musicians or singers; Tour guide systems. These systems allow a tour guide to provide a commentary on the subject of the tour. In some cases there will be more than one frequency in use to allow additional languages to be used; A voice quality 7 khz audio bandwidth is common in these devices; Baby Alarms consisting of a sensitive microphone in the vicinity of a baby with the receiver adjacent to parents or other adults. Page 61

These devices may use FM or digital modulation with an output power below 10mW. The receivers are capable of receiving signals in the 863-865MHz band. At the time of writing Ofcom has commissioned two studies to ascertain the extent of SRD deployment in the 863 870 MHz band in the UK and to assess the immunity of SRDs to interference from mobile terminals below 862 MHz. A separate study is also investigating the potential impact of mobile terminals on systems operated by the emergency services in the 862 863 MHz band. 14.1 863MHz-865MHz testing (Copsey Communications) A variety of six professional, semi-professional and consumer devices operating in the license-exempt frequency band 863-865MHz conforming to the harmonised standards EN 301 357 or EN 300 422 were tested by Copsey Communications for susceptibility to a 10MHz bandwidth LTE signal transmitting from 700MHz to 862MHz emulating both the base station and terminal unit: 50% of devices were adversely affected by a 10MHz wide signal at 857MHz (this is the highest centre frequency for a 10MHz wide terminal unit); A third of devices were susceptible to LTE transmissions below 830MHz; One device out of six was susceptible to LTE transmissions below 779MHz. Terminal units can use up to 316mW (ECC Decision (09)03 and draft EC Decision). Those devices susceptible to interference on the highest planned LTE terminal unit channel, centred at 857MHz, will start to show interference at powers above 0.15mW ERP (LTE signal transmitted from a 3m distance). This is a conservative figure with the cordless audio transmitter and receiver within 1.5m of each other, whereas normal operating conditions for the cordless audio equipment would probably be in the range 3-5m. One unit also showed interference from the base station transmit frequencies, which could well affect a large percentage of existing cordless headphone users. In addition whilst the base station out-of-band emissions are stringent at -49.5dBm EIRP, there is the possibility of interference from base stations under some circumstances, especially with the older cordless audio and baby alarm systems. The figure below reproduced by kind permission of UK Ofcom and Copsey Communications provides a graphic view of the energy dispersion over the band; this is obtained from the block edge mask contained in CEPT Report 30. Page 62

Figure 23 - excerpt from presentation MFCN base station emission limits for different 800 MHz licensees, UK Ofcom. The objective was to determine if Mobile Networks operating in the reallocated band 790-862MHz using LTE modulation will cause interference to incumbent devices transmitting and receiving in the 863-865MHz band. LTE modulation schemes are likely to include data dongles which may well be in close proximity to domestic equipment. As the cordless audio devices consists of two parts (a transmitter that broadcasts to a receiver or receivers in the vicinity), it was necessary for both components to be operating normally in the presence of an interfering signal in the 790-862MHz band to determine if there is potential for interference to the transmitter and/or the receiver. Various tests have been carried out within Europe and LTE terminal units have been shown to cause interference up to 4m away, this means that terminal units in adjacent flats or rooms may well generate interference into domestic setups. As these devices migrate into Mobile phones the possibility of these being in close proximity to radio microphones or even on the user of the cordless audio device will increase. 14.2 863MHz 865MHz test set up and results The cordless audio devices were set up in an anechoic chamber at a 3m separation distance from the interferer antenna. Initially the cordless audio transmitter was located centrally and held stationary whilst the output of the receive unit was monitored as it was being moved around in proximity to the cordless audio transmitter. The locations were exchanged, and testing repeated. Moving the cordless audio units enables the impact of any nulls or peaks in the transmit power of both the LTE and cordless audio signals to be overcome. The devices were either mains-powered or powered by fully-charged or new batteries. Page 63

A 1kHz audio tone was supplied to the transmit device, and the output of the receive device monitored on headphones in every case. The interferer was a 316mW ERP LTE signal transmitted in a 10MHz bandwidth, using a Frequency Division Duplex scheme. The characteristics of the signal are contained in Table below. Table 8- LTE modulation parameters Radio format Bandwidth Standard rev Access technique Modulation Waveform name FDD Uplink 10MHz 2008-09 SC-FDMA Full-filled QPSK LTE_FDD_UL_10 Test set up and monitoring data capture detailed in the figures below. The field strength was monitored by a NARDA probe. Figure 24 a tour guide system comprising of transmitter (square black device) and receiver (blue headphones with integral receiver) during test preparation. The 1kHz tone generator can be seen behind the transmitter, and a NARDA field strength probe can be seen behind that. Page 64

Figure 25 - Test support board layout. In this example a portable transmitter broadcasts from a central location and a portable receiver is moved around the board. Receive 3 metres Transmit Bilog antenna Test support Movement of receive unit 1.5 metres Chamber Control room Signal generator Cable 3 Cable 6 RF amplifier Figure 26 - Test environment and arrangement of equipment Page 65

FM Wireless headphones Semi-professional radio microphone Professional radio microphone system 1 Professional radio microphone system 2 Professional portable wireless microphone and receiver Tour guide system The following results, were obtained from 4th-12th May 2010. Table 9 - Cordless Audio devices (863-865 MHz). Susceptibility to LTE operating from 700 857MHz. LTE generator centre frequency (1MHz steps) 700-769 770-779 * 780-789 ** 790-799 ** 800-809 ** 810-819 * 820-829 * * 830-839 ** ** 840-849 ** * 850-856 ** ** 857 - highest potential 10MHz LTE Centre Frequency ** ** *** Key: * Mild LTE noise present ** Moderate LTE noise present *** Strong LTE noise present Page 66

Three of the six devices tested where susceptible to signals radiated within the proposed 790-862MHz band. ECC Decision 09/03 provides for a terminal unit power of 23dBm +/- 2dB. For the purposes of testing a 25dBm signal has been used and stepped down where interference was observed, in an attempt to discern the power boundary where interference ceases. Table below shows the signal levels where interference to the cordless audio equipment starts to be observed. Table 10 - maximum 10MHz bandwidth LTE signal power Radiated power (ERP, dbm) Radiated power (ERP, mw) FM Wireless headphones 13.96 24.88 Semi-professional radio microphone 15.3 33.88 Professional radio microphone system 1-8.14 0.15 14.3 863MHz-865MHz test conclusions System susceptibility varied dependant on the physical location of the receiver relative to both the interfering transmission antenna and the cordless audio transmitter: nulls and peaks were detected during lateral movement around the 3m separation point, which in most cases had a reasonable or dramatic effect on the scale of the interference. Where the operating frequency could be selected, 861.3MHz was set as the frequency of transmission/reception for the cordless audio unit. Where selection could not be made, transmission was within the 863-865MHz band. A 10MHz bandwidth LTE signal with a centre frequency of 857MHz will occupy the range 852-862MHz. When listening to audio over a system any noise or interference is unacceptable. Where a radio microphone is in use, any interference will ruin a production or cause the audio amplification to overload with possible harm to the audience. Page 67

15 Opportunities for Innovation (Industry & Academia) Traffic growth is growing exponentially in the fixed and wireless networks and being able to use spectrum more efficiently is of paramount importance for future services and applications to the UK economy. The demand for new services and bandwidth is driving the industry to looks at new technologies and methodologies to use this spectrum more efficiently whilst ensuring services that are already operated continue without a detrimental effect to the quality of service and coverage. The DCKTN working group is providing a unique forum for the different industries to come together to network, transfer knowledge and identify potential issues. Identifying potential issues at an early stage is critical to ensure new services and networks are planned correctly and can be implemented in a cost effective manner. The real benefit for the community engaged with this activity is to treat these potential issues as opportunities for R&D in the UK to solve the problems locally and exploit the outputs globally, this can be achieved by member companies working individually or collaboratively. Wireless is a crucial element of delivering the medium and long term goals for advanced network and broadband services. Recognition and increased future investment is critical to ensure we can benefit from this opportunity. Some notable data points recently published by CISCO are listed below: Mobile phones and laptops will drive over 80% of global traffic by 2013 - indicating the increasingly mobile nature of mobility, both on a consumer and professional level. iphone and Blackberry have created exponential growth in mobile data traffic resulting in a predicted CAGR increase of 131 per cent by 2013. Annual global IP traffic will exceed two-thirds of a zettabyte (667 exabytes) by 2013, By 2013, The internet will be four times larger than it is at present The UK benefits from having all of the elements of the value chain, from world-leading content providers, through to infrastructure and platform development, with vital component companies feeding multiple levels. At the very top of the value chain we have some of the most advanced digital consumers in the world rapidly adopting new services willing to pay for high quality content. The DCKTN Wireless Technology and Spectrum group is working on bringing different parts of the Digital value chain together and being the catalyst for industry co-operation across sectors that have traditionally not worked together in the UK and globally. Page 68

The opportunity to take a leading role in the shaping the future of wireless broadband is available to the UK eco-system to enhance our local services and provide a platform to take incremental market share worldwide, generating export opportunities and encouraging global corporations to inwardly invest in the eco-system. The eco-system is not dominated by one player and/or market segment and therefore it is critical to identify these opportunities at an early stage to encourage and accelerate collaboration and innovation. The paper has highlighted several levels of potential innovation and the following sections provide a high level breakdown. 15.1 Content and Services Understanding the evolution of advanced content delivery to multiple screen sizes and equipment has enabled the potential for new converged services to the consumer, with content specially adapted for each platform and the possibility to synchronise the different content streams and interact in real time. Bringing organisations together, which have traditionally worked independently, is critical for the UK to realise these new concepts and services. In this case, the Digital Dividend Spectrum not only brings together these industries to address the technology challenges it also provides the platform for potential new content and services innovation to take place utilising these new technologies and working across multiple platforms. 15.2 Network Topologies The issues identified in this paper provide significant opportunities for the industries that currently use this spectrum and new entrants that wish to exploit the new opportunities to solve the problems and therefore create new technology and business models. Several co-operative solutions have been proposed and with current technology capability the use and value of the spectrum will not be truly realised without the industries jointly addressing the opportunities. Whilst the DCKTN Wireless Technology and Spectrum working group cannot provide the leading candidate for co-operation several opportunities need to be explored in further detail: Broadcast & Cellular White Space & Broadcast White Space & Cellular Broadcast, Cellular & White Space Fixed v Mobile Wireless solutions Home Base Station, Small Cell and Macrocell All of above. In combination with SRD, PMSE and Cable industries Page 69

15.3 Consumer Platforms Future Consumer platforms such as STB, Home media, Televisions etc will require further connectivity and interaction to benefit from new services and to ensure QoS is maintained. The concept of passive Home Media solutions, not interacting with other networks and dynamically adapting will be replaced by intelligent hubs leveraging these benefits. The Network topologies section above has shown the type of interaction and networks that need to be operating in a connected mode. The Content and Service section provides one example of how future business models could be realised in providing incremental revenue and services not available today. The possibilities of Mobile consumer platforms offering increased bandwidth and interactivity are significantly enhanced by the number of options that can be provided with new network topologies. The combination of Mobile and Fixed Home media solutions can offer significant advantages with seamless connectivity, interaction and services crossing multiple platforms and boundaries. 15.4 RF Front End Components The DCKTN Wireless Technology & Spectrum working group have been highlighting many of the challenges identified in this paper at several events and positioning over the last 18 months. Within the Mobile Space, design and manufacturing teams are being asked to produce mobile phones, lap tops and base stations that cover at least five frequency bands and these are set to increase in the future. Recently at the DCKTN Internet Everywhere energy and bandwidth challenges, Vodafone presented a hypothetical 10 band global platform 15 and highlighted the need for further innovation and investment into front end radio design. The economics of designing these solutions within the desired power, cost and size budget drive the industry to look for the economies of scale to make them viable and hence harmonisation of spectrum and defining solutions that have global appeal is critical for success and investment. 15 Technology challenges in using new wireless spectrum, Trevor Gill, Vodafone Group R&D. https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=737699&folderid=870487&name=dlfe- 7188.pdf Page 70

The Digital Dividend not only adds to the challenge of new bands but also puts some very strict RF characteristics such as Receiver sensitivity, out of band emissions and sharp filtering. Some devices, particularly larger than smart phone form factor devices are also expected to receive DVB-T or ATSC TV broadcasts and the possibilities discussed in this paper regarding broadcasting and cellular solutions co-existing and working in a cooperative mode add to the possible innovation opportunities. The 700 and 800 MHz UHF bands introduce specific innovation opportunities both in terms of operational bandwidth and size constraints including a need to implement adaptive matching and other adaptive techniques in order to increase operational bandwidth beyond traditional good practice limits. Additionally the transceivers have to co-exist at a minimum with the 850 MHz and 900 MHz bands, 1800 and 1900 MHz bands, 1.9/2.1 GHz (Band 1) and other radio functions, for example Wi-Fi and Bluetooth at 2.4 GHz and GPS and FM receivers. In the longer term there will be a need to support 2.6 GHz, possibly 3.5GHz and additional regionally specific allocations. Integrated White Space cognitive transceivers have also been proposed in this paper. A requirement to support multiple standards introduces significant complexity, for example a need to support higher order modulation options and symbol orthogonality implies a need to control linearity and minimise noise and distortion in all parts of the transmit and receive chain. Present design solutions for multi band have used tried and trusted architectures and RF design options that have relied on discrete switch paths for each band. This results in component duplication but also introduces additional insertion loss and poor isolation. Other options exist but need to be carefully implemented to realise performance gain within acceptable cost and size parameters. The interaction of radio technology potentially to be used will also present new challenges in Receiver compliance. The following section provides a recommendation to Ofcom and BIS. 15.4.1 Radio and Telecommunications Terminal Equipment Directive (R&TTE) The DC-KTN encourages Ofcom and BIS to revisit the use of mandatory receiver specifications as a mechanism for increasing the economic value of spectrum, especially in those bands where spectrum trading is envisaged. Page 71

The DC-KTN spectrum group members believe that the introduction of receiver requirements would give a measure of surety to potential new users of radio bands that they would not be required to make "unreasonable" allowance for existing legacy equipment or incumbents when introducing new services and to allow them to make a more comprehensive business plan regarding technology adoption, network planning and user equipment. Page 72

16 Summary and Collaboration Wireless technology development, within the Digital Dividend Spectrum, is a major opportunity for the UK to lead the way globally in deployment, technology innovation, policy and regulation. The key to unlocking this potential is collaboration at all levels. If this can be achieved we will be able to realise significant societal benefits whilst enabling our industry to benefit across the whole of the value chain. The DCKTN Wireless Technology and Spectrum working group believe we can act the catalyst for this collaboration and is keen to take this to the next phase. This document has highlighted some major opportunities for industry to address by solving the issues through technological innovation. To further enhance this eco-system we believe a more holistic UK approach is required involving not only industry but also Ofcom, BIS and the Technology Strategy Board. Driving Innovation INDUSTRY DCKTN Wireless Technology & Spectrum Working group Funding R&D to solve the major challenges INDUSTRY Identifying & prioritising challenges Feasibility R&D to ensure technology can meet market, spectrum and economic challenges Policy and Regulation Flexibility to implement R&D trials POLICY & REGULATION Figure 27 - Collaboration enabling UK Wireless Innovation To help focus the request to OFCOM, BIS and the Technology Strategy Board a table of key priorities is listed below. The output of the DCKTN working group is biased towards innovation and technology challenges and therefore this should be read as opportunities for innovation and best use of spectrum from a technical perspective. Page 73