112 CHAPTER 4 DESIGN OF ADAPTIVE MODULATION SYSTEM BY USING 1/3 RATE TURBO CODER (SNR Vs BER) 4.1 NECESSITY FOR SYSTEM DESIGN The improved BER was achieved by inhibiting 1/3 rated Turbo coder instead of 1/2 rated Turbo coder. It was accomplished by sending the output from 1/3 rated Turbo encoder to serial to parallel converter. The output from serial to parallel converter was inhibited to various Adaptive modulation schemes like m-arypsk, m-aryqam, m-arycpm, and m-arymhpm system. The modulated output was then fed to multiple access techniques like MC- CDMA system and OFDMA system in fast fading Rayleigh environment. The BER performance was observed for both the MC-CDMA system and OFDMA system inhibiting various modulation schemes and the corresponding simulation results were plotted in Matlab and Simulink software. 4.2 INTRODUCTION: PERFORMANCE 4.2.1 Design Procedure Multi Carrier Code Division Multiple Access (MC-CDMA) and Orthogonal Frequency Division Multiple Access (OFDMA) were highly regarded as possible candidates for implementation as a future wireless communication physical layer. It has been recognized that future enhancements in wireless communication includes the inherent ability to
113 support adhoc based wireless networks, interwork with existing broadcast Infrastructure and also adaptation to multiple transmission systems (Won Gi Jeon 1999, Samreen Amir 2009). 4.2.2 Working Principle of Turbo Coder Here Turbo coder (high performance forward error correction code) was selected to meet the dynamic channel capacity. Turbo codes find specific use where designers seek to achieve information transfer over bandwidth or latency constrained communication links in the presence of data corrupting noise. Here the encoder inhibiting two major blocks sending three sub blocks of bits was implemented. The first sub block was the m bit block of payload data. The second sub-block was n/2 parity bits for payload data computed using a RSC code and the third sub block was n/2 parity bits for a known permutation of the payload data again computed using an RSC convolution code was given to first block where as the first sub block was the 'm _bit block of interleaved payload data. The second sub-block was n/2 parity bits of interleaved payload data computed using a RSC code and the third sub block was n/2 parity bits for a known permutation of the interleaved payload data again computed using an RSC convolution code was given to the second block. The output from both the blocks was connected to dual switch and the output from dual switch was given to serial to parallel converter. The output from S/P converter was given to adaptive modulator. The output from the modulator was given MC- CDMA/OFDMA system. 4.2.2.1 Working principle of Proposed (New Design) system The combination of Turbo coder, OFDMA with adaptive modulation technique (m-aryqam, m-arypsk, m-arycpm, and marymhpm)
114 system in a free Rayleigh fading channel was compared with Turbo coder, MC-CDMA with adaptive modulation technique(m-aryqam, m-arypsk, m-arycpm and m-arymhpm) system in a free Rayleigh fading channel.the graphs were plotted between Users Vs BER for up to 32 users. 4.2.2.2 Channel Quality Estimation The impact of the channel prediction error on the bit error rate performance was analyzed by providing channel predictor based on pilot symbol assisted modulation for multiple-input multiple-output Rayleigh fading channels. Based on the prediction of the expected channel conditions for next time slot, the modulation scheme was selected by the transmitter. (Chatterjee 2003) 4.3 SIGNALING CONDITIONS The Receiver was to be intimated properly about the set of demodulator parameters to be decided by the receiver for proper receiving of the signal through proper signaling (Chatterjee 2003). 4.4 DESIGN PARAMETERS Here the MC-CDMA system was initially designed and design parameters were assigned below. Initially the 1/3 rated Turbo coder was implied for a channel bandwidth of 20MHz with 1024 subcarriers for a symbol rate of 640 kilo symbols/s in a free Rayleigh channel for up to 32 users.
115 4.4.1 System Design Parameters Table 4.1 System Design Parameters S.No Parameters Specifications 1. Channel bandwidth >20 MHz 2. Frequency 5 GHZ 3. No of subcarriers 1024 4. Subcarrier spacing 25 KHZ 5. Portion of symbol 40 µs 6. Cyclic extension duration 10 µs 7. Total symbol duration 50 µs 8. Symbol rate 640 symbols/s 9. Chip rate 20.48 Chips/s 10. Code length 32 chips 11. CDMA code Walsh hadamard Table 4.2 Channel parameters S.No Parameters Specifications 1. Mobile speed 100 km per hour 2. Number of paths Four with exponential power distribution 3. Maximum excess delay 150 µs 4. Decaying factor 10% of symbol duration
116 Table 4.3 Turbo Coder Parameters S.No Parameters Specifications 1. Rate 1/3 or 1/2 2. Constraint length 3 3. Interleaving Block interleaving with block size 1024 bits 4. Decoding MAP decoding with hard inputs Figure 4.1 Parallel concatenation of two RSC codes (1/3 code)
117 Figure 4.2 Punctured Rate R=1/2 Turbo Coder A turbo code was the parallel concatenation of a number of RSC codes. The input to the second decoder was an interleaved version of the systematic x, thus the outputs of coder 1 and coder 2 are time displaced codes generated from the same input sequence. The input sequence was only presented once at the output. The outputs of the two coders were multiplexed into the stream thus gives R=1/3 code, or might be punctured to give R=1/2 code (Charlesworth 2000, Samreen Amir 2009). 4.4.2 Implementation Procedure (Chatterjee 2003) The Developed (NEW ) system was designed for 32 users. The OFDMA/ MC-CDMA transmission system was opted for the developed system.the digital modulation techniques like m-arypsk, m-aryqam, m-arycpm and m-arymhpm were inhibited for the developed system.the Turbo coder designed for symbol rate of 640ksymbols/s and 1/3 rate was incorporated in the System.The Simulation was performed for the newly developed system using Matlab and Simulink Software.
118 4.4.3 Working Procedure of Proposed System MC-CDMA or OFDMA transmitter and the receiver configurations were given below. The input information sequence was converted in to P parallel sequences (a j0 (i), a j1 (i) a (j)p-1 (i). The S/P converter output was multiplied with spreading code of length K MC. The output N=P*K MC was modulated by using IDFT or IFFT and converted back to serial data. The guard interval was provided between the symbols to avoid inter symbol interference. The output signal was again uplifted by using RF signal and sent for transmission. In the receiver after down lifted from RF signal, The original signal was recovered by removing the m-sub carrier components (m=0, 1, 2. K MC-1 ) by using DFT or FFT and then multiplied with the gain G j (m) to combine the energy of the scattered signal. In the present modulation techniques like m-arypsk, m-aryqam, m-arycpm and m-arymhpm in OFDMA and MC-CDMA system have reached the level of maturity. Here commercial product and operational networks operate with low spectral density and facilitates coexistent with other systems. The main drawback of spread spectrum technique was failure in detection of unauthorized users. This drawback was overcome in the newly developed(new) OFDM system. Here the serial to parallel converted data streams were spread by using a given spreading code and then different subcarriers were modulated with each data stream. The use of conventional CDMA was not applicable as the data rate goes above 100 Mega bits per second due to severe intercede interference and seems to be difficulty to synchronize such a fast sequence. The solution for
119 the above problem was achieved by finding a new technique which reduces both the symbol rate and the chip rate. The common point was to change the conventional serial transmission of data/chip stream in to parallel transmission of data/chip symbols over a large no of narrowband orthogonal carriers. Hence the chip and bit duration was increased proportionally (Tom Richardson 2008, Yongzhe Xie 2003, Jagan Mohan 2010, Cheng Yang Li 2003, Frieder Sanzi and Sven Jelting 2003). 4.5 EXPLANATION OF PROPOSED(NEW DESIGN) SYSTEM In the developed (New) system designed using Matlab and Simulink, the output from 32 user Random generator with 1/2 rated Turbo Coder inhibited adaptive modulator data source was connected to MC- CDMA/OFDMA transmitter. The Output was connected to the receiver by passing it through a Rayleigh channel i.e. through a Random noise generator. In the receiver, the output from MC-CDMA/OFDMA receiver was connected to the demodulator. The output from the demodulator was connected to the display via error calibrator. The output display was Spectrum analyzer which could be used for analysation of signals in frequency domain (Lei Yei and Alister Burr 2009, Chatterji 2003). In the developed (NEW) System shown in Figure 4.3, data was provided to 32 user channel with Turbo encoder and Adaptive modulator consisting of m-ary PSK, m-aryqam, m-arycpm and m-arymhpm. The modulator output was connected to MC-CDMA/OFDMA transmitter. The spread signal from the MC-CDMA transmitter was transmitted through the Rayleigh fading channel.
120 Figure 4.3 Proposed (NEW DESIGN) System In the Receiver, the received signal was passed through the MC-CDMA /OFDMA Receiver. The output signal from MC-CDMA receiver was taken to Adaptive demodulator. The demodulator output was given to the decoder. The decoded data was finally taken as the output The Figure 4.4 represents the simple transmitter with data source, transmitter and channel. Here the Random integer generator was chosen as source to provide necessary data to Adaptive modulator. The modulated data was spread using MC-CDMA transmitter and passed through the Rayleigh fading channel. Rayleigh fading was used in heavily built-up city centers where there was no line of sight between the transmitter and receiver. The buildings and other objects attenuate, reflect, refract and diffract the signal. In
121 troposphere and ionosphere signal propagation, the particles in the atmospheric layer act as scatters. The Output from Rayleigh fading channel was taken to scope for display. Figure 4.5 Simple modulator in Rayleigh fading environment The Figure 4.5 represents the simple Transmitter with data source, transmitter and channel. Here the Random integer generator was chosen as source to provide necessary data to adaptive modulator for modulation. The modulated data was spread using MC-CDMA transmitter and passed through the Raleigh fading channel. Raleigh fading channel was taken to scope for display. The user circuit was designed for up to 32 users. The user rate was increased from one up to 32. 4.5.1 Mc-CDMA/OFDMA System Simulation Result The simulation was carried out by using Matlab and Simulink software for various users of up to 32 and the graph were plotted between SNR Vs BER for digital modulation techniques like m-arypsk, m-aryqam, m-arycpm and m-arymhpm for both OFDMA and MC-CDMA system and the graphs were plotted. m-arypsk, m-aryqam, m-arycpm, m- arymhpm performance were plotted using Matlab and Simulink software for up to 32 users and the graphs were plotted.
122 Figure 4.6 SNR Vs BER for PSK Figure 4.6 indicates how the BER performance varies under different channel conditions or channel SNR for given number of 4users, 8users16users and 32 users. It can accommodate the highest number of users while maintaining a particular BER performance than other methods. The system was simulated for a fixed channel condition (CSNR = 15 db) with varying number of up to 32 users. The system first assigns the highest order modulation format and as the number of users increases, the system changes the modulation scheme appropriately according to the estimated channel conditions (CNR on pilot symbol). The Process begins with 16 PSK as the modulation scheme When the number of users increases to four, the BER degrades below the threshold level and hence the system switches to 8 PSK. Similarly, when the number of users increases to eight the system switches to 4 PSK and when the number of users is increases to sixteen, the system switches to 2 PSK.
123 Figure 4.7 SNR Vs BER for QAM Figure 4.7 indicates the BER performance varies under different channel conditions or channel SNR for given number of 4users, 8users, 16users and 32 users for QAM. The system was simulated for a fixed channel condition (CSNR = 15 db) with varying number of up to 32 users. The system first assigns the lowest order modulation format and as the number of users increases, the system changes the modulation scheme appropriately according to the estimated channel conditions (CNR on pilot symbol). The Process begins with 2 QAM as the modulation scheme When the number of users increases to four, the BER degrades below the threshold level and hence the system switches to 16 QAM. Similarly, when the number of users increases to eight the system switches to 8 QAM and when the number of users is increases to sixteen, the system switches to 4QAM and so on.
124 Figure 4.8 SNR Vs BER for CPM Figure 4.8 indicates the BER performance varies under different channel conditions or channel SNR for given number of 4users, 8users, 16users and 32 users for CPM. The system was simulated for a fixed channel condition (CSNR = 15 db) with varying number of up to 32 users. The system first assigns the lowest order modulation format and as the number of users increases, the system changes the modulation scheme appropriately according to the estimated channel conditions (CNR on pilot symbol). The Process begins with 4 CPM as the modulation scheme When the number of users increases to four, the BER degrades below the threshold level and hence the system switches to 2CPM. Similarly, when the number of users increases to eight the system switches to 16 CPM and when the number of users is increases to sixteen, the system switches to 8 CPM and so on.
125 Figure 4.9 SNR Vs BER for MHPM Figure 4.9 indicates the BER performance varies under different channel conditions or channel SNR for given number of 4users, 8users, 16users and 32 users for MHPM. The system was simulated for a fixed channel condition (CSNR = 15 db) with varying number of up to 32 users. The system first assigns the lowest order modulation format and as the number of users increases, the system changes the modulation scheme appropriately according to the estimated channel conditions (CNR on pilot symbol). The Process begins with 4 MHPM as the modulation scheme When the number of users increases to four, the BER degrades below the threshold level and hence the system switches to 2MHPM. Similarly, when the number of users increases to eight the system switches to 16 MHPM and when the number of users is increases to sixteen, the system switches to 8 MHPM and so on.
126 4.6 RESULT The simulation was carried out by using Matlab and Simulink software for various users of up to 32 and the graph were plotted between SNR Vs BER for digital modulation techniques like m-arypsk, m-aryqam, m-arycpm and m-arymhpm for both OFDMA and MC-CDMA system and the graphs were plotted. The m-arypsk, m-aryqam, m-arycpm, m- arymhpm performance were plotted using Matlab and Simulink software for up to 32 users. From the graphical analysis it is clear that the m-ary PSK, m-ary QAM, m-ary CPM and m-ary MHPM varies its array accordingly as the user rate increases from 1 user to 32 users and hence maintaining constant BER..