Abstract:
Equalization techniques compensate for the time dispersion introduced by
communication channels and combat the resulting inter-symbol interference (ISI) effect.
In blind channel equalization, desired signal or training sequence is not available, so
Higher Order Statistics (HOS) of signal are used to determine the instantaneous error
e(n). This instantaneous error is then used to update the filter taps. This thesis thoroughly
investigates the performance of Improved Rectangular Contour Algorithm (IRCA) for
odd bit M-ary Rectangular QAM (RQAM). IRCA is an extension of Rectangular Contour
Algorithm (RRECTCA), designed for RQAM. Multiple zero error contours are introduced
using the concept of slicing in order to achieve enhanced performance and faster
convergence rate. Simulation results are provided to validate the claim that the
performance of IRCA is better than that of its parent algorithm. The computational
complexity of both algorithms i.e. IRCA and RRECTCA is almost same.
