A BCI Control System for TV Channels Selection

A BCI Control System for TV Channels Selection Jzau-Sheng Lin *1, Cheng-Hung Hsieh 2 Department of Computer Science & Information Engineering, National Chin-Yi University of Technology No.57, Sec. 2, Zhongshan Rd., Taiping Dist., Taichung 41170, Taiwan *1 jslin@ncut.edu.tw; 2 ddtpojack@gmail.com Abstract In this paper, we proposed a wireless Brain-Computer Interface (BCI) with Steady-State Visually Evoked Potentials (SSVEP) to control a television for channels selection. In this system, we used EEG acquisition chip to extract SSVEPs from EEG signals and transform them by using of FFT into frequency domain. Then, these SSVEPs can be converted into commands to control television through a on a mobile device and an infrared emitter for patients. In this system, several flickering patterns with different frequencies were generated. EEG chip were used to capture EEG signals from location Oz on occipital lobe. The patients gazed these flickering patterns to generate SSVEPs, and then these SSVEPs were extracted from location Oz on their occipital lobe. These EEG signals can be transformed by FFT into frequency domain and then transmitted to the BCI system through interface. The advantages of the proposed BCI system are low cost, low power consumption and compact size so that the system can be suitable for the paralytic patients. The experimental results showed that feasible action can be obtained for the proposed wireless BCI system and control circuit with a practical operating in living space for paralyzed patients. Keywords BCI; EEG; SSVEP; Occipital Lobe Introduction Lots of patients lead to congenital or physiological damage due to a major accident which results in individuals cannot fully control their willpower in their life. EEG is the activity on the scalp that indicated by Ullah et al. in 2011. Usually, EEG signals are recorded through a simple system named BCI with multiple electrodes placing on the scalp and applied conductive adhesive. BCI is a low-cost and widely used noninvasive EEG capture technology that can be appropriately applied in a variety of medical auxiliary equipment. BCI is a system combined with hardware and software so that people can directly communicate with external devices through the neuromuscular pathway [Liavas, 1998; Wang, 2008; Bin, 2011; Chang, 2010; Wang, 2010]. BCI system is also a promising tool that can help the paralyzed people to complete several actions such as controlling medical assistant devices. In recent years, different EEG signal characteristics such as mu/beta rhythm, the P300 event-related potentials and visual evoked potential (VEP) have been widely used in the field of BCI. The VEP system has some advantages including higher information transfer rate (ITR), small amount of training samples, low user variable and easily use. The SSVEP signal is natural responses to visual stimulation at particular frequencies ranging from 3.5 Hz to 75 Hz [Wu, 2011; Ortner, 2011; Lin, 2012]. When the eyes are excited by a visual stimulus signal, the brain then generates same reaction at the same frequency of the visual stimulation signal. Frequency coding method has been widely used in the SSVEP-based BCI systems. In such a system, each visual target is flickering with a fixed frequency. The system can identify the primary frequency of SSVEP for the subject's gaze target. The brain electrical activity will produce resonance frequency in the visual cortex that can be applied to achieve visual-control peripheral devices with such brainwave response model. Chang et al. [2010] constructed a wireless SSVEP-based BCI system to remote control riders. They used three different flashing-frequency blocks displayed on the LCD with 13Hz, 14Hz, and 15Hz respectively. These three blocks, gazed by user, were corresponded actions such as left, forward, right, and front for a remote control car. Wang et al. [2010] developed an SSVEP-based BCI system to remote control a car in 2010. Ortner et al. proposed an SSVEP-based system to control a hand orthosis for persons with tetraplegia in 2011. In 2012, Lin et al. proposed a wireless BCI with EEG and eyeblinking signals for controlling electric wheelchair. In 2013, Lin and Huang constructed a new type of 71

International Journal on Communications (IJC) Volume 3, 2014 electric wheelchair BCI system with a core hardware using FPGA architecture. Shyu et al. also used FPGA hardware to construct a BCI-based hospital bed nursing system in 2013. In this paper, an SSVEP-Based control system for TV channels selection was proposed. Five different flashing-frequency patterns were displayed on a tablet PC. The used frequencies were 7Hz, 9Hz, 11Hz, 13Hz, and 15Hz to correspond five actions including Power, Volume, Sequential Channel, Channel Table, and Channel #, to select a suitable channel. System Architecture An SSVEP-based BCI system to control a television for selecting a suitable channel was proposed in this paper. The purpose is to solve the inconvenience of patients in their daily lives in a living space. The SSVEP-based EEG signals of patients are used as input source of the BCI system. A patient looked at the flickering pattern on a tablet PC to generate SSVEPbased EEG signals. We can then extract EEG signals from the location Oz on occipital lobe through the EEG acquisition system. Then, these EEG signals can be transmitted through a interface to the analysis platform on a tablet PC and transferred by FFT and converted commands to control a television. The SSVEP-based BCI system, as shown in Fig. 1, was designed to extract EEG signals from occipital lobe of a patient when he gazed flickering blocks with different frequencies. The extracted frequencies can be converted to several control commands. The control commands were used by a patient to control a television in a living space. The proposed system used a Tablet PC named AUSU PadFone as the main controller in which we developed an Android App and constructed RVS screen for visual stimulation with difference frequencies. The controller communicated with a television by using of a and an infrared emitter interfaces. Through this App, it can allow patients to operate the control systems by gazing the flickering blocks with eyes on the screen. In this system, electrodes are attached point Oz on the occipital lobe for stimulated flickering blocks with different frequencies and point A2 as a reference like shown as in Fig. 2. In the EEG acquisition module shown as in Fig. 3, the EEG signal was extracted and processed by a NeuroSky EEG chip. The circuit diagram is shown as in Fig. 4. Then, EEG signals were transmitted through a interface in the platform of ASUS PadFone. In addition to the interface, the signal processing unit occupied FFT module to transfer EEG signal from spatial domain to frequency domain for recognizing frequencies. EEG Signals Tablet PC Infrared Emitter FIG. 1 SYSTEM ARCHITECTURE Television FIG. 2 ELECTRODE LOCATIONS OF INTERNATIONAL 10-20 SYSTEM NeuroSky EEG chip Module Regulator FIG. 4 CIRCUIT DIAGRAM OF EEG EXTRACTING MODULE 72

Infrared Carrier AT89S51 FIG. 5 CIRCUIT DIAGRAM OF INFRARED EMITTER MODULE The block diagram of TV control system is shown as in Fig. 8. A suitable frequency can be extracted from EEG signals and transmitted to the tablet PC when a patient gazes a flickering block. When AT89S51 received a control command from tablet PC through module, it can get the corresponding code on the lookup table. Then infrared emitter module emits the code to start a function for the controlled television. FIG. 3 EEG ACQUISITION MODULE; (A) EEG CAPS; (B) ELECTRODE PADS; (C) NEUROSKY CHIPS; (D) BLUETOOTH MODULES; (E) POWER SUPPLY For the purpose of controlling a television, we need to design an infrared emitter circuit module. The infrared emitter circuit is shown as in Fig. 5. Infrared is not visible or a common way of wireless communication for universal remote control interface in home appliances. The circuit of 38KHz-carrier frequency of infrared is shown as in Fig. 6. When the infrared receiver receives carrier signal, shown as in Fig. 7, the output is set Low level otherwise the output is set High level. In general, the transmitted infrared data including four basic parts with different time intervals such as start signal, end signal, data 1 and data 0. In Fig. 6, the pin I/O is controlled by AT89S51. First, we have to create a look-up table in AT89S51 to match the code table of remote-control functions for infrared emitter in television. When AT89S51 receives a command from module, it will find an infrared-emitter code from look-up table and emit the code to television to generate a suitable operation. FIG. 6 CIRCUIT DIAGRAM OF INFRARED-CARRIER GENERATOR There are five flickering blocks with different frequencies to enter next pages including Power, Volume, Sequential Channel, Channel Table, and Channel #. When the next page is Power, the patients can choice Power on or Power off block to turn on/off the TV power. If Volume is selected, the Volume + or Volume- can be selected to adjust volume. In the Sequential Channel page, flickering blocks Previous Channel or Next Channel can be gazed by patient to display the previous/next channel 73

Gazing with Eyes International Journal on Communications (IJC) Volume 3, 2014 sequentially. In the Channel Table page, the channel names are displayed on the screen. And, the patients can also select a channel using flickering blocks by Previous Page or Next Page with binary searching method. Finally, the digits of channel number can be entered by using of gazing flickering blocks sequentially to select a suitable channel in the Channel # page. Every subject was requested to complete a functional operation for TV system. The subjects had 10 minutes to learn the functions for TV control system before experiments. During the functional operation, three times were required for the selected function. The consuming time of a successful operation and incorrect number of operations were recorded. The Information Transmission Rate (ITR) is generally used to estimate the performance of the communication and control for BCIs [Wolpaw, 2013]. The higher value of ITR indicates the more performance. The proposed systems is also used ITR to assess the performance of the system. The ITR is defined by (a) Infrared carrier signal (b) Output of infrared receiver FIG. 7 THE CARRIER OF INFRARED SIGNAL AND OUTPUT OF INFRARED RECEIVER Power Volume Sequential Channel TV Control Interface Channel Table Channel # Change Page Tablet PC Android App: FFT Control Commands SMS AT89S51 TV Control System (RX) Infrared Emitter FIG. 8 TV CONTROL SYSTEM WITH SSVEP-BASED BCI Experimental Results Brain EEG Signals EEG Extracting Module NeuroSky Chip (TX) In the proposed integrated system, we used the ASUS PadFone as the main controller to connect with the control system by using of interface. The application program was developed on the Android system by using of Java language. The flickering blocks with different frequencies for repetitive visual stimulus (RVS) were generated on the panel. Then, a command can be converted by the AT89S51 to find a suitable infrared code in its look-up table and transmitted to the controlled television. In order to test the television control system, seven men aged between 20 and 25 years old were invited. where N is the total number of commands, P is the probability of correct selection, and CTI, expressed as Command Transfer Interval, is the assuming time for a command. An operation of television control system is set to complete one functional operation to open the power and select an indicated channel. It needs ten commands to power on and select a channel. The experimental results for TV control system are shown as in Table 1. The subjects 3 and 6 complete a TV operation with 96.29% accuracy. And, the most consuming time is 70 seconds to select a channel by subject 3 at test 3. The average CTI and ITR for these 7 subjects are 4.8 seconds per command and 31.8 bits per minute, individually. From Table 1, the CTIs and ITRs are fluctuated with small values on their average values. From these experimental results, it can be indicated that the proposed TV control system based on BCI is quite stable. Conclusions In this paper, a television control system to extract SSVEP-based EEG signals through a BCI interface in a smart living space for patients was proposed. In the proposed system, the patients just gazed the flickering target to select a suitable operation. This operation can be transferred by extracting the EEG signals from the location Oz on the occipital lobe. And, EEG signals were transferred from spatial domain into frequency domain to take a matched frequency. Then, the extracted frequency directly was converted a command and committed to the infrared emitter module to find a suitable infrared code through (1) (2) 74

by the AT89S51. This SSVEP-based control system can be easily and stably used by patients to turn ON/OFF a television and selecting a suitable channel. TABLE 1 THE EXPERIMENTAL RESULTS FOR TV CONTROL SYSTEM Subject Test 1 Test 2 Test 3 Avg. Time (sec) Accuracy (%) CTI ITR 1 43 44 43 43.33 100 4.33 35.79 2 43 45 47 45.00 100 4.50 34.47 3 46 45 70 53.67 96.29 5.37 25.38 4 44 43 47 44.67 100 4.47 34.72 5 49 50 47 48.67 100 4.87 31.87 6 68 45 45 52.67 96.29 5.27 25.86 7 43 47 45 45.00 100 4.50 34.47 Avg. 48.0 45.6 49.1 47.6 98.9 4.8 31.8 ACKNOWLEDGMENT In this paper, the research was sponsored by the National Science Council of Taiwan under the Grant NSC103-2221-E-167-027. REFERENCES Bin G., Gao X., Wang Y., Li Y., Hong B., and Gao S. A highspeed BCI based on code modulation VEP. J. Neural Shyu K.-K., Chiu Y.-J., Lee P.-L., Lee M.-H., Sie J.-J., Wu C.- H., Wu Y.-T. and Tung P.-C. Total Design of an FPGA- Based Brain Computer Interface Control Hospital Bed Nursing System. IEEE Transactions on Industrial Electronics 60(2013): 2731-2739. Ullah K., Rizwan M. Ali, M., and Imran M. Low-Cost Single-Channel EEG Based Communication System for People with Lock-in Syndrome. Paper presented at IEEE 14th International Multitopic Conference 120-125, Dec. 2011. Wang H., Li T. and Huang Z. Remote Control of an Electrical Car with SSVEP-Based BCI. Paper presented at IEEE International Conference on Information Theory and Information Security (2010): 837-840. Wang Y., Gao X., Hong B., Jia C., and Gao S. Brain computer interfaces based on visual evoked potentials IEEE Eng. Med. Biol. Mag 27 (2008): 64 71. Wolpaw J. R., Birbaumer N., McFarland D. J., Pfurtscheller G., and Vaughan, T. M. Brain computer interfaces for communication and control. Clinical Neurophysiology 113(2002): 67 79. Eng (2011) doi:10.1088/1741-2560/8/2 /025015. Wu, H.-Y. Lee P.-L., Chang H.-C., and Hsieh J.-C.,. Chang H.-C., Deng H.-T.,. Lee P.-L, Wu C.-H. and Shyu K.- K. Real-Time Control of an SSVEP-Actuated Remote- Controlled Car. Paper presented at SICE Annual Conference (2010): 1884-1887. Liavas A. P., Moustakides G. V., Henning G., Psarakis E. Z., and Husar P. A Periodogram-Based Method for the Detection of Steady-State Visually Evoked Potentials. IEEE Transactions on Biomedical Engineering 45 (1998): 242-248. Lin J.-S. and Huang S. M. An FPGA-Based Brain-Computer Interface for Wireless Electric Wheelchairs. Applied Mechanics and Materials, 284-287(2013): 1616-1621. Lin J.-S. and Yang W.-C. Wireless Brain-Computer Interface for Electric Wheelchairs with EEG and Eye-Blinking Signals. International Journal of Innovative Computing, Information and Control 8(2012): 6011-6024. Ortner R., Allison B.Z., Korisek G., Gaggl H., and Pfurtscheller G.. An SSVEP BCI to Control a Hand Orthosis for Persons with Tetraplegia. IEEE Transactions on Neural Systems and Rehabilitation Engineering 19(2011): 1-5. Accounting for Phase Drifts in SSVEP-Based BCIs by Means of Biphasic Stimulation. IEEE Transactions on Biomedical Engineering 58 (2011): 1394-1402. Jzau-Sheng Lin received the B.S. degree in Electronic Engineering from National Taiwan University of Science and Technology in 1980, the M.S. and Ph.D degrees in Electrical Engineering from National Cheng Kung University in 1989 and 1996, respectively. He is currently a Professor in the Department of Computer Science and Information Engineering, National Chin-Yi University of Technology, Taiwan. His research interests include brain computer interface, neural network, pattern recognition, image processing, and medical image analysis. Cheng-Hung Hsieh received the B.S. degree in Computer Science and Information Engineering from National Chin-Yi University of Technology in 2011. He is currently a graduate student in Computer Science and Information Engineering of National Chin-Yi University of Technology. His research interests include brain computer interface and circuit design. 75