Design and Implementation of Adaptive Modulation Modem Based on Software Defined Radio(SDR) for WiMAX System

This paper presents design and imp lementation of adaptive modulation modem for WiMAX system. (BPSK, QPSK, 8QAM, 16QAM, 32QAM and 64QAM ) are used in this work . Software Defined Radio(SDR) is used for implementing this modem. This work examines the benefits of using adaptive modulation in terms of probability of bit error and spectral efficiency. It s pecifically e xamines t he perform ance enhancement m ade possible by using linear predicti on a long with channel estimation in conjunction wit h adaptive modulation. Simulation results proved that t he adaptive system performance w ith e stimator an d predictor is better than other modulation a lone. The s imulation results for adaptive modulation in compared to each modulation technique alone show that for BER=10 -3 with (f d =50Hz -200Hz)system the improvement occurs by decreasing S/N by 2-3dB. As for BE R=10 -4 with same Doppler freque ncy, the system improvement tak es pl ace by de creasing S/N by 1.3dB-4dB. Recarding BER=10 -5 with same Doppler frequency, the improvement is by decreasing S/N by 1.5dB-5 dB. Simulation results als o s how the flexibility of t he adaptive system to operate w ith di fferent level of modulation based on switching of S/N. Matlab7.8(R2009a) used for simula tion of ada ptive modulation system with AWGN and fading channel.


Introduction
Wireless communications has been one of the fastest growing segments in the telecommunications industry.The various advantages of 3G over 2G such as higher data rate as well as increased system capacity have been major motivations to move to 3G [1].It is well known that the fundamental limitation of wireless systems is constituted by their time-variant channel fading, which results in dramatic fluctuations in signal to noise ratio SNR The traditional wireless systems are designed to provide good transmission quality for the worst channel conditions.As a result, signal to noise ratio that are much larger than the target are achieved over a large portion of the coverage area and transmission time, which leads to inefficient utilization of the full channel capacity.In addition, the integration of the

Adaptation Boundaries
The first step in adaptive modulation modem is to define a way to select which modulation scheme is best suited for the present (or future in the case of delayed feedback).SNR at receiver is used [7] as a good channel metric to decide the selection(or switching) of the modulation scheme.The ranges of S/N (in the receiver) will be used to select modulation scheme (in the transmitter) based on AWGN performance for each modulation scheme.Figure ( where: B is spectral efficiency of adaptive modulation. The BER performance of equation( 7) is simulated with switching levels shown in Table (1) for .

3.SNR Estimation
There are several methods to estimate signal to noise power ratio [10].The purpose of measuring SNR is to get a more accurate view of the channel state.The received and demodulated signal will be corrupted by both the Rayleigh channel and receiver noise (whose statistics do not change over short intervals).In general the Rayleigh fluctuations are too quick to use for the adaptation, so the goal is to track (and adapt to) shadow fading while averaging out the Rayleigh fading [11].The SNR measurement approach is shown in Figure (2).The received signal is passed through a square-law envelope detector and then amplified using a linear or log amplifier [10].Consider both types of amplifiers, since the statistics of the Rayleigh fading are simpler for a linear amplifier, while the statistics of the log-normal fading are simpler for a log amplifier.

4.Channel Prediction
Adaptive modulation methods depend on accurate channel state information (CSI) that can be estimated at the receiver and sent to the transmitter via a feedback channel [12].This information would allow the transmitter to choose the appropriate transmitted signal.The feedback delay and overhead, processing delay and practical constraints on modulation switching rates have to be taken into account in the performance analysis of adaptive modulation methods.For very slowly fading channel, CSI is sufficient for reliable adaptive system design.However, for rapidly time variant fading that corresponds to realistic mobile speeds, even small delay will cause significant degradation of performance since channel variation due to large Doppler shifts usually results in a different channel at the time of transmission than at the time of channel estimation [2].To realize the potential of adaptive transmission methods, the channel state information (CSI) is obtained by channel prediction [13].
The idea behind channel prediction is to use past and present channel samples to predict future samples.The implementation of prediction scheme is for specific purpose to expect the future power level of the Rayleigh channel [14]  3. Receiver which is responsible for data reception and demodulation of the received data.Once, the data has been demodulated, an estimate and prediction of the received S/N then feedback to the transmitter via the feedback loop.

Description of system model
The design parameters of the system are shown in table(2).
6. Description of the Designed system components

Design of Digital Arm Filter
The digital arm filter is designed by using the hamming window for simplicity in designing filters for Iand Qchannel.Lowpass filter is used in order to remove the high frequency components from the output of multiplier.The two digital arm filters in receiver must have the same design to avoid jitter and the order is not too high because that means delay.This filter will be FIR filter because it has linear frequency response.This means no phase distortion is introduced into the signal by the filter [15].Table (3) shows the parameters selected for the filter.
The transfer function is given by: ……..( 9) The hamming window is given by[16]: PDF created with pdfFactory Pro trial version www.pdffactory.comwhere: ) is the window function, is the desired impulse response of the filter, is chosen proportional to sampling rate, M is the filter order.

Design of Estimation Circuit
The first stage of estimation circuit is square low envelop detector as shown in Figure (2).Squaring the signal effectively demodulates the input signal by using itself as the carrier wave.The envelope can then be extracted by keeping all the DC low-frequency energy and eliminating the high-frequency energy.The second stage of estimation circuit is amplifier(linear or non linear).
The third stage is RC low pass filter whose cutoff frequency will be very small in order to produce the DC level.The type of selected modulation is based on the value of SNR estimated by this estimator.The output of estimation unit is given by: … (11) where: is the output voltage level, is the square input signal, is the gain of amplifier and is the transfer function of the feedback loop digital filter.Table(4) shows the parameters selected for the estimation circuit.

Linear predictor
Linear predictor is used to predict the current value of the signal based on the past samples.Linear predictor is selected to reduce the delay time taken for estimation and time required to return the estimation value to the transmitter [9]

Figures(6,7)
show the performance of BPSK system over AWGN and Rayleigh fading evaluated by plotting the Bit error rate(BER) versus the (S/N) .

QPSK system simulation results
The performance of QPSK system will be evaluated by plotting the Bit error rate(BER) versus the (SNR).

M-QAM system simulation results
The performance of M-QAM system will be evaluated by plotting the Bit error rate(BER) versus the (SNR) in the presence of AWGN and Rayleigh fading for different values of Doppler frequency.Figures (10,11,12,13,14,15,16,17) show the performance of 8QAM, 16QAM, 32QAM and 64QAM over AWGN and Rayleigh fading.

Simulation results of Adaptive modulation with estimation and prediction
Figures (18,19,20,21,22,23) show the results of bit error rate versus SNR for (BPSK, QPSK, 8QAM, 16QAM, 32QAM, 64QAM) system respectively where estimation and prediction are used.These results show the effectiveness of estimator and predictor to improve the results by degrading the required SNR for a given BER.Figures(24,25,26)show the variation of spectral efficiency of adaptive modulation with estimation and prediction for different values of BER ( 10 -3 ,10 -4 ,10 -5 ).Simulation results concide with the theoretical results since the required BER is a fixed value in channel and ideal estimation in order to control switch level of SNR.

Discussion of simulation results
1. Adaptive modulation improved the power spectral efficiency if compared with uniform modulation for specified parameters.The simulated curves coincide well with the theoretical curves for different Doppler rates attributed to the presence of estimator and predictor so good spectral efficiency curves will be obtained.2. Doppler frequency reduces the system performance especially at high values either with modulations alone or with adaptive modulation and suffer losses in BER.This leads to use prediction.3. Prediction improves the system performance as compared with system without prediction but prediction is not affected too much at very high Doppler frequency but system PDF created with pdfFactory Pro trial version www.pdffactory.comperformance with prediction is better than without prediction.4. Finally, the results obtained from the simulation prove that adaptive modulation system was operated with high accuracy and stable performance

8.Conclusions
The following points represent the main conclusions obtained from this work:-1.The use of adaptive modulation allows a wireless system to choose the highest order modulation depending on the SNR.Different order modulations allow to send more bit per symbol and thus achieve higher data rate and better spectral efficiency.2. Adaptive modulation is suggested for WiMAX system due to its capability to satisfy the WiMAX requirements.
3. Simulation results show that in the presence of fading, the system degrades by 20dB for given BER if compared to presence of AWGN.However, this degradation in system efficiency can be improved by using adaptive modulation and keep BER in the same level by selecting different modulation (based on SNR estimation for pre-defined target BER).4.Channel estimation with prediction will improve the performance of adaptive modulation.The system for BER 10 -4 will be improved by 2dB if compared with system without estimation.5.The system performance can be improved by adding linear prediction to avoid delay which is caused by time to estimate the SNR of the received signal and the time required to return the estimated value to transmitter.
6. Simulation results show that as Doppler frequency increased up to 200Hz system, performance decreased the required SNR by 1.5dB for BER 10 -3 and 1dB for BER 10 -4 and 1.5dB for BER 10 -5 if compared with system performance without estimator and predictor.
7. Implementing this system by SDR provides more flexibility with lower cost and time.The work can be extended by using FPGA with VHDL language to store the software used in this system.

Order of predictor 3
The

Figure( 4
Figure(4) shows the general layout of the proposed system.Figure(5)represents the flow chart of the designed system.The main parts of the implemented proposed system are:-1.Transmitter: The transmitter is responsible for: a. Generating the symbols of the transmitted data which is transmitted over a wireless channel.b.Selecting modulation schemes based on S/N(which is estimated in the receiver).The modulation schemes used in the proposed system are BPSK,QPSK, 4QAM,8QAM,16QAM,32QAM,64Q AM. 2. Channel: Mobile wireless channel with AWGN and Rayleigh fading are used in simulation.

Figure
pdfFactory Pro trial version www.pdffactory.com

Journal, Vol.28, No.14, 2010 Design and Implementation of Adaptive Modulation Modem Based on Software Defined Radio(SDR) for WiMAX System 4732
PDF created with pdfFactory Pro trial version www.pdffactory.comEng.& Tech.

com Eng. & Tech. Journal, Vol.28, No.14, 2010 Design and Implementation of Adaptive Modulation Modem Based on Software Defined Radio(SDR) for WiMAX System 4733
PDF created with pdfFactory Pro trial version www.pdffactory.

& Tech. Journal, Vol.28, No.14, 2010 Design and Implementation of Adaptive Modulation Modem Based on Software Defined Radio(SDR) for WiMAX System 4734
PDF created with pdfFactory Pro trial version www.pdffactory.comEng.