ADAPTIVE MODULATION AND CODING FOR WIMAX SYSTEMS

This project is on design and implementation of adaptive modulation in Wimax
systems for broadband wireless networks. The majority of the work done is on
acquiring the cannel signal to noise ratio (SNR) and comparing two pilot aided
SNR estimation techniques with a blind estimator. Reddy Noise power estimator
and Subspace based estimator are pilot aided, both of which are a combination of
channel estimation and noise estimation. The third is a blind autocorrelation
based Linear Predictor Interpolator. Fixed transmission systems make use of
average modulation and coding which has reasonable performance in both high
and low SNR conditions in the communication channel. This results in channel
bandwidth under-utilization in high SNR conditions. The use of adaptive
modulation improves the spectral efficiency by making use of variable bit
loading, depending on the prevailing SNR.A higher modulation and coding rate
is used in high SNR conditions, while a lower and noise robust is used in low
SNR conditions. The main task of the project is to do a performance analysis of
SNR estimation algorithms by implementing each on the IEEE 802.16d standard
for broadband wireless local and metropolitan area networks (WMAN) working
model. The SNR estimation technique which results in optimal performance will
later be used to in a feedback system which inputs the channel SNR into a control
mechanism that makes the decision to switch modulation and coding rate based
on a predetermined SNR threshold value for each rate that ensures acceptable bit
error rate (BER). First the SNR estimation using the three algorithms is
implemented in Wimax system, based on the transmission protocols specified in
Wimax. The performance of the estimators is then compared by determining the
mean square error and computation time for each as the performance criteria and
to compare the performance of the unaided SNR estimator in comparison to the
conventional pilot aided SNR estimators.