Analysing Cognitive Radio Physical Layer on BER Performance over Rician Fading Amandeep Kaur Virk, Ajay K Sharma Computer Science and Engineering Department, Dr. B.R Ambedkar National Institute of Technology, Jalandhar, India anu_virk10@yahoo.co.in sharmaajayk@nitj.ac.in Abstract This paper analyzes the Bit Error Rate performance of Cognitive Radio Physical layer over Rician channel with AWGN noise under different channel encoding schemes, digital modulation schemes and channel conditions. The system outperforms with Reed Solomon along with convolution encoding for modulation technique as compared to other digital modulation schemes and the system is highly effective to combat inherent interferences under Rician fading channel. The system shows improved BER on using encoding schemes with error rate reduced by 10% using Reed Solomon encoding, 91% reduction on using convolutional encoding and 99% error reduction on applying Reed Solomon with convolution encoding. It has been anticipated from the simulation study that the performance of the communication system degrades with the increase of noise power. Keywords - Cognitive Radio, Bit Error Rate, Rician Fading, Reed Solomon encoding, Convolution encoding. 1. Introduction According to Federal Communications Commission (FCC), temporal and geographical variations in the utilization of the assigned spectrum range from 15% to 85% in current spectrum allocation policies [1]. So, we need to find out ways to allow wireless devices to efficiently share the airwaves. Cognitive Radio (CR) has been proposed as a potential solution for spectrum inefficiency problems [2]. CR promises to increase spectrum usage by supporting unlicensed users to share licensed bands [3]. Licensed users of CR are called primary users and the unlicensed users are called secondary users. CR supports user access to the licensed spectrum as a secondary user when and where channels are detected idle [4]. CR is quite different from the traditional wireless radios, so the cognitive radio layers perform additional functionality along with functions of the conventional wireless radios. This paper presents physical layer analysis of Bit Error Rate (BER) performance over a Rician channel with AWGN noise. Time Division Multiple Access (TDMA) system is used for licensed users along with Orthogonal Frequency Division Multiplexing (OFDM) and non- persistent Carrier Sense Multiple Access (CSMA) system is used for unlicensed users who opportunistically use the idle licensed spectrum. Various digital modulation schemes used are,,,, and. Reed Solomon (RS) and convolution encoding schemes are used with varying values of Signal to Noise ratio (SNR) to compute BER. The paper is organized as follows: Section II describes the physical layer of cognitive radio, Section III gives the channel characteristics, Section IV presents the simulation results and finally, the paper is concluded in Section V. 2. CR Physical Layer The physical layer in cognitive layer is responsible for: 1) Spectrum sensing 2) Reconfiguration 3) Change of operation parameters Spectrum sharing is detecting unused spectrum bands. Reconfiguration is done when a licensed user appears while unlicensed user is using its frequency band and operation parameters (such as frequency, power and modulation) are changed to adapt to new operating environment [5]. Figure 1 shows the various layers of cognitive radio along with the physical layer functions. 1697
attempts to correct errors that occur during transmission and recover the original data. A Reed-Solomon code is specified as RS(n,k) with s- bit symbols. This means that the encoder takes k data symbols of s bits each and adds parity symbols to make an n symbol codeword. There are n-k parity symbols of s bits each. A Reed-Solomon decoder can correct up to t symbols that contain errors in a codeword, where 2t = n-k. RS algebraic decoding procedures can correct errors and erasures. An erasure occurs when the position of an erred symbol is known. A decoder can correct up to t errors or up to 2t erasures [8]. 3. Figure 1.Cognitive Radio layers 3.1 Additive White Gaussian Noise (AWGN) In communications, the AWGN channel model is one in which noise is additive, white and noise samples have a Gaussian distribution [6]. Additive noise means received signal equals the transmit signal plus some noise and the noise is statistically independent of the signal. White noise has flat power spectral density and so the autocorrelation of the noise in time domain is zero for any non-zero time offset. The model does not account for the phenomena of fading, frequency selectivity, interference, nonlinearity or dispersion. AWGN is commonly used to simulate background noise of the channel under study, in addition to multipath, terrain blocking, interference, ground clutter and self interference that modern radio systems encounter in terrestrial operation [7]. 3.2 Rician Fading Rician channel is a transmission channel that may have a line of sight component and several scattered off multipath components. The fading assumed to be flat and hence has a multiplicative effect on the channel input. 3.3 Reed Solomon Encoding and Decoding Reed-Solomon codes are block-based error correcting codes. The Reed-Solomon encoder takes a block of digital data and adds extra redundant bits. The Reed-Solomon decoder processes each block and 3.4 Convolution Encoding and Decoding Convolution encoder takes k bits at a time and outputs n encoded bits. Convolution codes are usually described using two parameters: the code rate and the constraint length. The code rate, k/n, is expressed as a ratio of the number of bits into the convolutional encoder (k) to the number of channel symbols output by the convolutional encoder (n) in a given encoder cycle. In this paper, 1/3 convolutional code is used. Viterbi algorithm is used for decoding [9]. 4. Simulation Results The simulations are carried out using MATLAB. The performance is simulated and evaluated for,,,, and modulation techniques. Forward error correction (FEC) encoding schemes used. The system is observed by separately using convolution and RS encoding and then using them in combination by interleaving. Simulation results are shown for TDMA, CSMA and overall channel. Overall channel shows the results for TDMA and CSMA system combined together. The simulation result is evaluated on BER vs. SNR for Rician fading channel with AWGN when the number of data packets is 500 with each packet of size 128 bytes and the BERs are obtained by varying the values of SNR in the range of 2 to 16 db In all the cases, the system provides degraded performance in 16 PSK and satisfactory performance in modulation. Figure 2 through 5 shows the BER performance of data through CR physical layer under six types of digital modulation schemes on Rician fading channel with AWGN. Figure 2 below gives the BER obtained at various SNR values over Rician channel for overall system, TDMA system and CSMA system. It is observed that and 16 QAM modulations perform best among other modulation schemes. The BER values at SNR 2 db are 8.6348 x () and 10.3039 x () for overall channel, 4.1101 x () and 5.1281 x () for TDMA system and 1698
4.5247 x () and 5.1758 x () for CSMA system. Figure 2.BER simulations through CR physical layer over Rician channel for overall channel for TDMA system for CSMA system Figure 3 below gives the BER obtained at various SNR values over Rician channel with RS encoding for overall system, TDMA system and CSMA system. It is observed that modulations perform best among other modulation schemes and the. BER is reduced by 10%. The BER values for at SNR 2 db are 9.266 x for overall channel, 5.0738 x for TDMA system and 4.1922 x for CSMA system. 1699
Figure 3.BER simulations through CR physical layer using RS encoded system for different modulation schemes over Rician channel for overall channel for TDMA system for CSMA system The simulation results for 1/3 convolutional encoded Rician channel are shown in figure.4 below. Figure 4 shows the improved results for overall channel with the improved value of BER at SNR 2 db for as 8.788 x. Improved value for TDMA system at SNR 2 db is 4.788 x and for CSMA system is 4 x as shown in figure 4 and respectively. Convolution encoding removes bit errors and the system shows 91% reduction in errors. 1700
Figure 4.BER simulations through CR physical layer using 1/3 Convolutional encoded system for different modulation schemes over Rician channel for overall channel for TDMA system for CSMA system Figure 5.BER simulations through CR physical layer using RS encoded and 1/3 Convolutional encoded system for different modulation schemes over Rician channel for overall channel for TDMA system for CSMA system Figure 5 shows the simulation results for RS and convolution encoded Rician channel. The BER values at SNR 2 db for modulation scheme show improvement with value for overall Rician channel as 0.566 x. RS encoding eliminates block errors and convolution encoding removes bit errors and the system shows 99% improvement. Table 1 shows the BER values obtained for various modulation schemes using RS and Convolutional encoding. The table highlights the lowest BER values obtained in each case. 5. Conclusion In this paper, the BER performance of data through CR physical layer is shown using RS, Convolutional channel coding and different digital modulation schemes. A range of system performance highlights the impact of digital modulations under RS and Convolution coding under Rician fading channel with AWGN. The system shows improved performance on using encoding schemes. The enhanced BER is 10% for RS encoding, 91% for convolution encoding and 99% for RS combined with convolution encoding scheme. In the context of system performance, it can thus be concluded that the implementation of modulation with RS and convolution channel coding technique together provides satisfactory result among the digital modulation schemes with limited SNR. 1701
BER values TABLE 1 RICIAN CHANNEL RS ENCODING 1/3 CONVOLUTION ENCODING RS WITH CONVOLUTION ENCODING SNR = 2 Overall TDMA CSMA Overall TDMA CSMA Overall TDMA CSMA Overall TDMA CSMA 17.0163 x 8.7643 x 8.252 x 22.9229 11.6924 11.2305 38.1758 19.4758 x x x x x 18.7 x 37.1377 17.6064 19.5313 x x x 10.3039 x 5.1281 x 5.1758 x 9.266 x 5.0738 x 4.1922 x 8.788 x 4.788 x 4 0.566 x x 0.566 x 0 23.0213 11.6932 11.3281 32.5537 16.0498 16.5039 x x x x x x 75.471 x 37.471 x 38 x 74.248 x 38.8965 35.3516 x x 40.1124 20.0929 20.0195 68.7383 35.1445 33.5938 101.861 49.9612 x x x x x x 2 x x 54.0772 27.1241 26.9531 93.7998 46.4365 47.3633 100.120 49.9208 x x x x x x 8 x x 51.9 x 101.158 49.8887 51.2695 2 x x x 50.2 x 96.8125 49.5469 47.2656 x x x 8.6348 x 4.1101 x 4.5247 x 14.2568 7.1279 x x 7.1289 x 5.6026 x 2.8026 x 2.8 x 6.3613 x 2.748 x 3.6133 x 6. References [1] Danijela Čabrić, Robert W. Brodersen Berkeley, Physical Layer Design Issues Unique to Cognitive Radio Systems, Wireless Research Center, University of California at Berkeley, USA. [2] Wang Weifang, Denial of Service Attacks in Cognitive Radio Networks, 2nd Conference on Environmental Science and Information Application Technology,2010 IEEE. [3] Xueying Zhang Cheng Li, The security in cognitive radio networks: A survey, 2009. [4] Peha J. M.: Approaches to spectrum sharing, IEEE Communications Mag., 2005, 43, (2), pp. 10 12. [5] Mitola J., Maguire G.: Cognitive radio: making software radios more personal, IEEE Pers. Communications, 1999, 6, (4), pp. 13 18. [6] www.wirelesscommunications.nl [7] Syed Asif, Abdullah-al-muraf, S.M. Anisul Islam, Amitavo Tikader, Abdul Alim, Comparison of BER between uncoded signal and coded signal over slow Rayleigh fading channel,journal of theoretical and applied information technology, 2005-2009. [8] www.cs.emu.edu [9] Chip Fleming, A tutorial on convolution encoding with viterbi decoding. 1702