LOW COMPLEXITY DETECTION ALGORITHMS IN MIMO SYSTEMS

Abstract

Wireless communication is one of the most promising technologies in recent years. As the technology advances, researchers are trying to discover more suitable algorithms/ techniques in this field. One of emerging technologies now a days is MIMO, which stands for Multiple Input and Multiple Output. One of the greatest advantages of using such a technology is that a high data rate of the order of a few Gb/s is a reality. MIMO systems with multiple antennas at both the transmitter and receiver sides provide several promising advantages. Some of them are high data rate, reduced signal distortion, higher accuracy and better spectral efficiency. Other than these significant advantages, there are added advantages of transmit diversity. The main segment of a MIMO system is the detection of the received signal. Multiple data streams are transmitted concurrently from multiple antennas and received at the receiver side. The reception technique at the receiver concludes the complexity and BER performance at the receiver side. ISI is a vital issue in broadband applications. Therefore, detection methods are used in such a way that the problem of the interference is reduced. There are many multiuser detection algorithms and each one is having some performance complexity tradeoff. Researchers are still aiming for more such algorithms which are easy to implement with reduced complexity and have near optimal performance. In this thesis, different MIMO detection algorithms like Zero Forcing (ZF) , Maximum iii Likelihood (ML) and Minimum Mean Square Error (MMSE) detection algorithms have been mentioned with their variants. Detection algorithm like ML (Maximum Likelihood) shows the optimal BER performance, but has some complexity issues. As the number of antennas increases, Maximum Likelihood (ML) detectors complexity increases many fold. Zero Forcing and Minimum Mean Square Error detection algorithms are Linear Detectors that show suboptimal BER performance with an added advantage of lower complexity. Further, some low complexity algorithms like Successive Interference Cancellation (SIC) are mentioned. In SIC, firstly, the signals from various transmitters are detected and then the effect of interference are cancelled. One more modified approach mentioned later in this thesis is Optimal Ordering. In this thesis, the BER performance of various MIMO detection algorithms have been compared. The channel considered here is a Rayleigh multipath channel and BPSK and QAM have been selected as modulation techniques. Simulations of all detection algorithms are done in MATLAB and their Bit Error Rate performance has been observed.