Adaptive Proportionate Normalized Least Mean Squares channel equalizer for MIMO-OFDM systems

The rise in data-centric user applications in recent times demands a propagation-friendly channel for data transmission. Multiple Input Multiple Output - Orthogonal Frequency Division Multiplexing (MIMO-OFDM) systems are employed for wireless broadband communications and considerable performance with regards to data rates is observed. Over the years investigations for the use of adaptive algorithms in channel equalization has been carried out. The Least Mean Squares equalizer (LMS) is a very popular channel equalizer used in MIMO-OFDM systems because of its faster convergence and low computational complexity. Keeping the performance constraints in view, we propose an equalizer employing the Improved Proportionate Normalized Least Mean Squares (IPNLMS) algorithm. The proposed method effectively deals with the issues regarding channel impairments and an increase in performance over the conventional method is achieved. Simulation results over a range of Signal to Noise Ratios (SNRs) with use of a training sequence for both LMS and IPNLMS equalizer is shown. The IPNLMS adaptive algorithm performs much better than the LMS algorithm and a significant improvement in Bit Error Rate (BER) is obtained. A comparative analysis based on optimal BER is made and the effectiveness of the adaptive equalization technique in general is shown. The proposed algorithm is a simple channel equalization technique and effectively removes the effect caused by the hostilities of the propagation channel by removing inter symbol interference (ISI).