A Tabu Based Neural Network Training Algorithm for Equalization of Communication Channels

This paper presents a new approach to equalization of communication channels using Artificial Neural Networks (ANNs). A novel method of training the ANNs using Tabu based Back Propagation (TBBP) Algorithm is described. The algorithm uses the Tabu Search (TS) to improve the performance of the equalizer as it searches for global minima which is many a time escaped while Back Propagation (BP) algorithm is applied for this purpose. From the results it can be noted that the proposed algorithm improves the classification capability of the ANNs in differentiating the received data.