Adaptive filtering algorithms for channel equalization in wireless communication

The major problems in today's wireless communications is time dispersion and inter symbol interference. To combat these issues various adaptive equalization techniques are used. In literature state of art algorithms of adaptive equalization will be studied in detail, however the utmost goal of these algorithms is to attain high convergence rate, less complexity and least error. In this paper a comparison between least mean square algorithm and fractional least mean square algorithm is presented and experimentally proved that the rate of convergence is high in case of fractional least mean square algorithm. Also it is observed that LMS algorithm has better performance for random signals and fractional LMS has proved very efficiently in case of deterministic signals. These both algorithms are implemented in MATLAB and eight channels with different tap weights are used but in this paper only selected are presented.