Demonstration of geolocation database and spectrum coordinator as specified in ETSI TS 103 143 and TS 103 145 ETSI Workshop on Reconfigurable Radio Systems - Status and Novel Standards 2014 Sony Europe Limited 1
Abstract Sony Europe Limited has been actively involved in the development of standards in the ETSI RRS technical committee. Recently, the following two standards have been developed by the ETSI TC RRS working group 1. - TS 103 143 System Architecture for Information Exchange between different Geo-Location Databases (GLDBs) enabling the operation of White Space Devices (WSDs) - TS 103 145 System Architecture and High Level Procedures for Coordinated and Uncoordinated Use of White Spaces The two technical specifications describe the interaction between GLDB, Spectrum Coordinator (SC) and WSD as well as between GLDBs. New concepts for spectrum management (coordinated use of TVWS) are also developed in these two standards as compared to existing TVWS standards (uncoordinated use of TVWS) such as EN 301 598 This demo intends to show the spectrum management functions based on TS 103 143 and TS 103 145. 2
Content 1 Introduction... 4 2 Architecture and scenario... 5 3 Demonstrated functions... 6 3.1 Interface between GLDBs... 6 3.2 Management of non-priority CRS for priority access... 8 3.3 Maximum EIRP calculation based range of activity... 11 References... 13 Biography... 14 3
1 Introduction The spectrum scarcity has become a limiting factor to the wireless communication systems to accommodate more users and incorporate more services. Allocating new spectrum for the additional services and capacity increment is costly since the available spectrum is a limited resource and has been assigned to various wireless systems. These systems known as the licensed users have exclusive use of the spectrum allocated to them. Contrary to the congestion seen on the frequency occupancy by regulatory bodies, the actual spectral load shows that there are large idle spectral bands [1]. Cognitive radio system (CRS) has now been considered as an efficient approach to solve the above-mentioned issue of spectrum scarcity. The CRS allows communication of unlicensed devices, as secondary users (SUs), to access the spectrum that are temporarily unused or underutilized by licensed users, i.e., the primary users (PUs) or incumbents. The commercialization of the CRS was ushered in by the digital switchover happened in UK, Japan and US leaving useable spectrum in the TV band to CRSs. In 2008 the FCC adopted rules to allow unlicensed radio transmitters, called unlicensed white space devices (WSDs), to operate in the broadcast television spectrum at locations where that spectrum is not being used [2]. Earlier 2013, ECC report 186 was issued by the European Conference of Postal and Telecommunications Administrations (CEPT), in which technical and operational requirements for the operation of white space devices under geolocation approach are shown [3]. Recently, OFCOM UK has initiated the trial program of TV white space operation. These activities in the regulatory domain push the industrial standardization acuities. In 2013 the IEEE 802.11af standard Wireless LAN operation in TV white space was published. In Europe, the ETSI Reconfigurable Radio Systems (RRS) Technical Committee (TC) as the center of competence for Cognitive Radio and related advanced Spectrum Management Technologies is actively developing standards for the operation of wireless systems in unlicensed band including TV white space. The developed standards not only consider providing sufficient protection to incumbents but also include spectrum coordination function which makes the utilization of precious spectrum efficient. These features include management of priority access, where the quality of service (QoS) of a CRS can be guaranteed by managing the spectrum utilization on other CRSs, interaction between different geolocation databases (GLDBs) that might belong to different regions under different spectrum regulatory requirements. This demo shows the spectrum management functions based on the following two technical specifications (TSs) which have been under development by ETSI TC RRS. TS 103 143 System Architecture for Information Exchange between different Geo-Location Databases (GLDBs) enabling the operation of White Space Devices (WSDs) TS 103 145 System Architecture and High Level Procedures for Coordinated and Uncoordinated Use of White Spaces 4
2 Architecture and scenario Figure 2.1 shows the scopes of the aforementioned standards, TS 103 143 and TS 103 145. TS 103 145 specifies the interaction between GLDB, spectrum coordinator (SC) and CRS; whereas, TS103 143 specifies the interaction between GLDBs. The GLDB is an entity whose operation is mandated or authorized by a regulatory authority and that provides a WSD in a CRS with location specific information on the available frequencies and associated maximum equivalent isotropically radiated power (EIRP) values that the WSD is permitted to use which allow for protection of the incumbent service and are derived from information provided by the WSD and the minimum required Adjacent Channel Leakage Ratio (ACLR) of the WSD. The GLDB consists of database and geo-location functions. The SC is an entity that coordinates spectrum usage of CRS based on the information obtained from geo-location database as well as supplemental spectrum usage data from different CRSs using its service. Scope of TS 103 145 GLDB CRS-1 Scope of TS 103 143 Other GLDB SC CRS-2 CRS-3 Figure 2.1 system architecture and the scopes of TS 103 143 and TS 103 145 Figure 2.2 shows the over scenario of the wireless system operating in TV band as unlicensed system we have implemented. The SC function is implemented together with the GLDB. The CRSs are instantiated as Wi-Fi network or LTE small cells where they are assumed to be an operator s small cell having an expected QoS. This CRS is considered as the priority-access CRS. The other Wi-Fi cells are considered as non-priority access CRSs. 5
Figure 2.2 Overview of TV white spaces system implemented by Sony 3 Demonstrated functions The GUI of the GLDB with SC implemented by Sony is shown in Figure 3.1 In the following section, we explain the details of the functions implemented. Figure 3.1 GUI of GLDB with SC function 3.1 Interface between GLDBs Different GLDBs may be deployed in different regions that are following different spectrum utilization regulations. Each GLDB can provide protection to the incumbents under its own 6
management region. Therefore, the information exchange between different GLDBs is needed to ensure that a CRS operating under the management of one GLDB will not interfere with the incumbent in the management region of another GLDB (the victim GLDB) as shown in Figure3.2. Figure 3.2 Demo for the interface between to GLDBs As shown in Figure 3.3, before providing information of available channels to a CRS, the managing GLDB shall check whether the use of such available channels by the CRS brings interference to incumbents registered in other GLDBs. If it is true, the responding GLDB is a victim GLDB of the requesting GLDB. Then, the responding GLDB send a result to the requesting GLDB. Also the information on the affected incumbents will be given to the requesting GLDB and such as the most severely affected region and other affected region. The CRS s EIRP calculated will consider the incumbents in its managing GLDB and the other GLDB. Finally the CRSs spectrum usage can be registered into the victim GLDB. Here, we use the information about the most affected region to determine a virtual CRS whose interference effect is equivalent to a group of closely located CRSs to reduce the information exchange burden. If a CRS s transmission will not interfere with the incumbent of the other GLDB, its information will not be registered. We can reduce the information exchange overhead. From Figure 3.1, it can be seen that when we input a CRS at one GLDB, the information at the other GLDB will also be changed. If virtual CRS is used only limited number of CRSs are exchanged between the two GLDBs. Without virtual CRS, the other database will transfer information of all the CRSs to the other database. Comparing two GUIs of the two GLDBs, we can see that GLDB1 only stores virtual CRSs information, whereas GLDB 2 without using the method stores information of all CRSs managed by GLDB 1. 7
GLDB 2 GLDB 1 Victim_Discovery_Request Victim_Discovery_Response Check if there are incumbents within the interference range Affected_Incumbents_Info_Request Affected_Incumbents_Info_Response Modify spectrum usage of CRSs managed by the querying GLDB for protecting incumbents registered in victim GLDB Channel_Usage_Registration_Request Channel_Usage_Registration _Confirmation Figure 3.3 Procedure for CRS calculation to protect incumbent in the other GLDB. In the calculation of EIRP, we also consider the location uncertainty incumbent and CRS. Instead of calculating EIRP considering all possible locations, the algorithm can first determine the critical points of CRS and incumbents. Based on these two points, the maximum EIRP is calculated. The computation complexity can be reduced significantly. Also, if the activity region of a CRS overlaps that range we can see from Figure 3.1, the power shown in EIRP SONY1 of that CRS will be very small and powers other CRSs that are not within the range of the incumbent are increased. 3.2 Management of non-priority CRS for priority access One of the new concept introduced in TS 103 145 is that the GLDB together with SC can realize the coordinated channel usage of white space. One of the examples is the priority access which allows a CRS operate in a channel for a specific reservation period and in a specific area based on particular minimum protection requirements of the CRS. The CRSs assigned such channels with therefore have priorities over other CRSs. The priority access of a particular channel means reducing spectrum usage (such as transmit power) of those CRSs without priority access on the same channel in order to guaranty the protection requirements, for example but not limited to SINR, of a CRS with priority access. The procedure is shown in Figure 3.4. The CRS request operational parameters for priority access. 8
The SC having the information of existing CRSs, will determine which CRS s spectrum usage shall be modified to provide the requesting CRSs intended spectrum usage. Although, the SC cannot directly manage the uncoordinated CRS, these CRSs need to check with GLDB to confirmation spectrum availability. Then, the GLDB will check with the SC and modify the spectrum usage of uncoordinated CRS for priority access. Finally, the SC will provide operational parameters to the coordinated CRS for its intended priority access. CRS1 (Uncoordinated) CRS2 (Priority) SC GLDB 1 Coordinated_Channel_Request Coordinated_Channel_Confirm Requesting CRS channel access procedure 2 Channel_Request Available_Channel_Response 4. Obtaining registration request (Sending device parameters) 3. Determine which uncoordinated CRSs need to update the operation parameters Requesting SC channel access procedure 5 PriorityUsage_Checking_Request PriorityUsage_Checking_Response Priority usage checking procedure 6. Update the current available channel of CRSs 7. Providing available channel (Sending operational parameters) 8. Obtaining registration confirm (Sending channel usage parameters) 9 ChannelUsage_Information_Announcement ChannelUsagae_Information_Confirm Providing uncoordinated CRS information procedure 10 CRS_Reconfigulation_Request CRS_Reconfiguration_Response Reconfiguration request from SC to CRS procedure 11 Channel_Usage_Notification_Request Channel_Usage_Notification_Response Procedure of channel usage notification for subject CRS Figure 3.4 Procedure for priority access 9
Reducing spectrum usage of non-priority-access CRSs brings two effects to the CRS with priority access on that channel. The first effect is a direct reduction of interference from the CRS without priority access to the CRS with priority access. The second is a reduction of the interference from the CRSs without priority access to the incumbent. This allows an increase of the maximum EIRP of the CRS with priority access while keeping protection to incumbent. Therefore, uncoordinated CRSs could be chosen considering both factors wisely such that a minimum number of CRSs are affected while achieving the priority access requirements. From the demo, we can see that the using this method, the number of affected CRSs can be reduced. As shown in Figure 3.1, when setting coordinate value ( QoS flag ) to 1 we can input the BER of the CRS as expected QoS. Using the procedure in Figure 3.4, EIRP Priority shows the result when we reduce the power of all non-priority CRS until the priority CRS s expected QoS is achieved. EIRP Priority SONY2 shows the result when we select the non-priority CRS with high interference to incumbent and interference to priority CRS. EIRP Priority SONY3 shows the results when we allow adjustment of transmit power of the priority CRS. We can see that EIRP Priority method reduces power of all existing on priority CRSs. Whereas EIRP Priority SONY2 adjusts only few CRSs. Moreover, EIRP Priority SONY3 can also increase the power of the priority CRS. We have also made a lab demo as shown in Figure 3.5. The CRS with priority access is making a video transmission while we move the transmitter as the non-priority access. The GLDB with SC will adjust the transmit power the non-priority access CRS to protect the expected QoS of the video transmission of the CRS with priority access. Figure 3.5 Lab demo for video transmission of CRS with priority access in the presence of non-priority access CRS 10
3.3 Maximum EIRP calculation based range of activity The maximum transmit power of a CRS is determined for a particular location under the constraint of incumbent protection. When the location of a CRS changes the maximum transmit power changes. If a CRS always transmit according to this location-specific transmit power limit, the power of its signal varies when the CRS s location changes. Such changes lead to frequent reconfiguration of the CRS s transmission and big fluctuation of the interference power to its neighboring CRSs. This change occurs not only when the CRS s location changes but also when neighboring CRSs locations change as the maximum transmit power limit is dependent on the interference from all CRSs to the incumbent. Start SC receives channel request information from a CRS containing its location and expected geographical area of movement during operation SC sends channel request of channel for all possible locations in the expected geographical area SC examines the distribution of the maximum EIRPs for all possible location and determines a x%-eirp SC checks the actual EIRP for the current location of the CRS Actual EIRP > x%-eirp? Yes The SC informs the CRS to use the x%-eirp No The SC informs the CRS to use the actual EIRP End Figure 3.6 Flowchart of the EIRP calculated based on activity region 11
Knowing the expected geographical range of movement during operation, the SC can obtain a set of all available spectrum, e.g., EIRPs for a CRS s all potential location the EIRP can be obtained. Let the value of x%-eirp be defined as that in which x% of the EIRPs for a set of locations in the range of movement are smaller than this value. For example, if the CRS uses the 10%-EIRP value there is a 90% of time that the actual EIRP calculated for the particular location of CRS is larger than the x%-eirp when the CRS is moving within the reported geographical area. This means that by 90% of time, the CRS can keep a constant EIRP and does not need to perform reconfiguration. The following section gives the procedure of SC operation of managing the CRS to use this x%-eirp as an example. The flowchart of the algorithm is shown in Figure 3.6. First, the SC receives the channel request information from one or multiple CRSs. The messages contain the locations of CRSs as well as the expected geographical area of movement during operation. The SC obtains available channels for all possible locations within the expected geographical area of CRSs. The SC examines the distribution of the maximum EIRPs for all possible location and determines an x%-eirp value. At each actual location, the CRS will access GLDB to obtain maximum EIRP. The SC then compares the actual EIRP for the current locations of the CRSs with the x%-eirp value. If the x%-eirp value is lower than the actual EIRP, the SC information the CRS to use the x%-eirp instead of the actual EIRP by the response message. If the actual EIRP is lower than the x%-eirp value, the SC informs the CRSs to use the actual EIRP values. Note that the same CRS will request channel again when its location changes and it is still within the previously reported expected area of operation, the SC does not need to calculate a new x%-eirp value. Figure 3.7 GUI of a CRS for inputting expected range of activity over a map 12
One use case of such range of activity is that a user would like to use CRS in unlicensed band for a period of time and over a region such as a garden. As shown in Figure 3.7, the user can indicate the range of activity over a map and the device will determine the range of activity as shown in the demo. The SC obtains the maximum x% EIRP of such CRS over this region. Figure 3.1 EIRP SONY4 shows the result for the tablet whose IP is 192.168.0.105. When using this x%-eirp, the reconfiguration complexity can be reduced when the user is moving inside the range of activity. References [1] G. Staple and K. Werbach, The end of spectrum scarcity, IEEE Spectrum, vol. 41, no. 3, pp. 48 52, Mar. 2004. [2] FCC, Second report and order and memorandum opinion and order, no. FCC 08-260, Nov. 2008. [3] ECC Report 186, Technical and Operational Requirements for White Space Devices under Geo-location Approach, Jan. 2013 13
Biography Chen SUN (Chen.Sun at sony.com.cn) received the PhD degree in electrical engineering from Nanyang Technological University, Singapore, in 2005. From August 2004 to May 2008, he was a researcher at ATR Wave Engineering Laboratories, Japan working on adaptive beamforming and direction finding algorithms of parasitic array antennas (the ESPAR) as well as theoretical analysis of cooperative wireless networks. In June 2008, he joined the National Institute of Information and Communications Technology (NICT), Japan, as expert researcher working on distributed sensing and dynamic spectrum access in TV white space. Since then he has been contributing to IEEE 1900.6 standard, IEEE 802.11af standard and Wi-Fi Alliance specifications for Wi-Fi networks in TV white space. He served as the technical editor of the IEEE 1900.6 standard, working group acting chair and received the IEEE Standard Association Award for his leadership in 2011. In 2014 he received the IEEE 802.11af outstanding contributions award. Currently, he works at Sony China Research Lab. Beijing as the head of Wireless Network Research Department. He is a delegate of Sony Europe Limited to ETSI TC RRS and rapporteur for the interface between Geolocation Databases enabling the operation of Cognitive Radio Systems (EN 303 144). Naotaka SATO (NaotakaA.Sato at jp.sony.com) received B.E. and M.E. degree in electrical engineering from Tokyo University of Science, Japan, in 1991 and 1993. He has been engaged in development of CDMA cellular phone and research and development of RF systems for 3G cellular phone system in Sony Corporation, Tokyo, Japan (1993-1995 & 1999-2001), Sony Electronics, Inc., San Diego California, USA (1995-1999) and Sony Ericsson Mobile Communications Japan, Inc., Tokyo, Japan (2001-2005). Since April 2005, he has joined Sony Corporation, Tokyo, Japan, as a senior researcher working on cognitive radio systems and 4G cellular technologies. His research interests include dynamic spectrum access and RF system architecture. He has been contributing to 3GPP RAN1, IEEE 1900.6 standard and IEEE 802.19.1 standard. In 2014, he received the IEEE 802.19.1 outstanding contributions award. He is a delegate of Sony Europe Limited to ETSI TC RRS and rapporteur for the interface between SC and CRS in coordinated use of TV White Space (EN 303 387-1). Sho FURUICHI (Sho.Furuichi at jp.sony.com) received B.E. and M.E. degree in electrical and electronic engineering from Tokyo University of Agriculture and Technology, Japan, in 2010 and 2012, respectively. In 2012, He joined Sony Corporation, Japan. He has been engaged in development of 4G cellular technologies and dynamic spectrum access technologies in TV white space. His research interests include 4G cellular technologies and cognitive radio systems. 14