The breadth first signal decoder (BSIDE) is well known for its optimal maximum likelihood (ML) performance with lesser complexity. In this paper, we analyze a multiple-input multiple-output (MIMO) detection scheme that combines; column norm based ordering minimum mean square error (MMSE) and BSIDE detection methods. The investigation is carried out with a breadth first tree traversal technique, where the computational complexity encountered at the lower layers of the tree is high. This can be eliminated by carrying detection in the lower half of the tree structure using MMSE and upper half using BSIDE, after rearranging the column of the channel using norm calculation. The simulation results show that this approach achieves 22% of complexity reduction for 2x2 and 50% for 4x4 MIMO systems without any degradation in the performance.

In a massive multiple-input multiple-output (MIMO) system, a large number of receiving antennas at the base station can simultaneously serve multiple users. Linear detectors can achieve optimal performance but require large dimensional matrix inversion, which requires a large number of arithmetic operations. Several low complexity solutions are reported in the literature. In this work, we have presented an improved two-dimensional double successive projection (I2D-DSP) algorithm for massive MIMO detection. Simulation results show that the proposed detector performs better than the conventional 2D-DSP algorithm at a lower complexity. The performance under channel correlation also improves with the I2D-DSP scheme. We further developed a soft information generation algorithm to reduce the number of magnitude comparisons. The proposed soft symbol generation method uses real domain operation and can reduce almost 90% flops and magnitude comparisons.

Multiple Input Multiple Output (MIMO (techniques use multiple antennas at both transmitter and receiver for increasing the channel reliability and enhancing the spectral efficiency of wireless communication system.MIMO Spatial Multiplexing (SM) is a technology that can increase the channel capacity without additional spectral resources. The implementation of MIMO detection techniques become a difficult mission as the computational complexity increases with the number of transmitting antenna and constellation size. So designing detection techniques that can recover transmitted signals from Spatial Multiplexing (SM) MIMO with reduced complexity and high performance is challenging. In this survey, the general model of MIMO communication system is presented in addition to multiple MIMO Spatial Multiplexing (SM) detection techniques. These detection techniques are divided into different categories, such as linear detection, Non-linear detection and tree-search detection. Detailed discussions on the advantages and disadvantages of each detection algorithm are introduced. Hardware implementation of Sphere Decoder (SD) algorithm using VHDL/FPGA is also presented.