Optimum Equalizer and PSK Modulation Technique for MIMO Wireless Communication System

Abstract

Most recently, there have been meaningful and significant advancements in the area of wireless communications, specifically with the introduction of multiple-input multiple-output (MIMO) and the ever-increasing use of equalizers in MIMO wireless communication. Based on phase shift keying (PSK) modulation and from the knowledge of zero-forcing (ZF), maximum likelihood (ML) and minimum mean square error (MMSE) equalizers, this thesis proposes a smart equalizer alternative to existing optimum equalizer termed as Modified MMSE equalizer that reduces bit error rate (BER) via spatial multiplexing (SM), diversity combining and splits up each user's MIMO channel into parallel sub-channels with identical SNRs. ZF equalizer amplifies noise at the time of matrix inversion and the complexity of ML equalizer increases exponentially with increasing number of transmitting antennas and higher order modulation. On the contrary, MMSE and MMSE-SIC cannot exhibit superior performance than ML and ISI is not completely removed, as a result a new optimal equalizer determination is crucial. This thesis sets up a novel technique using MMSE equalizer that eliminates the limitations occurred by the ML, MMSE and ZF equalizers while offering / optimal performance. The new equalizer is the upgraded version of MMSE equalizer that incorporates channel equalization and noise addition with successive interference cancellation by means of matrix ordering and modified vector estimation. Channel matrix with noise is first inverted to remove the effects of intersymbol interference (ISI). The channel is then multiplied with the transmitted signal and its interference is cancelled consecutively in accordance to higher signal power forming a proposed matrix order and modified vector. In this way, noise as well as ISI is eliminated in order by subtracting its effect from the subsequent stages and the process continues in an iterative way which optimizes the performance criterion. BER vs. signal-noise ratio (SNR) curve of Modified MMSE equalizer exceeds all of the equalizers considered in this research even than that of the existing optimum ML equalizer i.e. performance characteristics of Modified MMSE is optimal with respect to all of the equalizers. Besides, Modified MMSE equalizer eliminates noise as well as ISI while keeping complexity low as it does not involve any exponential terms akin to ML equalizer. The complexity is reduced from exponential to linear. The new equalizer technique is compared with other linear and non-linear techniques comprising ZF, ML and MMSE equalizers. MatLab toolbox is exploited for simulation that shows superior performance of the proposed equalizer over the conventional ones. Rayleigh fading channel is used in the simulation as it enables the computation of the expected complexity up to simpler expression in the paradigm. So far, for a realistic perspective, the optimal performance as well as simpler complexity is obtained at the same time and the optimal equalizer can be evaluated efficiently for MIMO wireless communication.