The Research On Spatial Modulation Detection Algorithm

High-speed developed mobile communication technology makes people pursue the communication system which has higher data transmission rates and reliability. By setting up multi-antennas in transmitting and receiving terminals, MIMO (Multiple-Input Multiple-Output) technology dramatically increases data transmission and spectrum efficiency. SM (Spatial Modulation) technology has received extensive attention recently as a potential novel multi-antennas transmission scheme. In this background, for the high computational complexity of the optimal ML (Maximum Likelihood) detection algorithm, the paper focuses on studying low-complexity SM detection algorithms, and the main work is as follows:Firstly, for the SM systems, the paper studies the property comparison among some conventional detection algorithms by computer simulation. These algorithms include ML detection, MRC (Maximum-Ratio Combining) detection, MMRC (Modified MRC) detection, MS (Multistage) detection and DBD (Distance-based ordered) detection. Meanwhile, some important inferences are obtained. These researches lay a good foundation for future study.Secondly, for the SM signals with QPSK constellation modulation, the paper proposes a novel optimal simplified ML algorithm based on the global optimal conditions of a binary quadratic programming. In the case of keeping the optimal detection performance, the new algorithm significantly reduces the computational complexity of the conventional ML algorithm and has a wider application value in large antennas SM systems.Finally, for the SM signals with any constellation modulation, the paper proposes a novel low-complexity suboptimal detection algorithm based on the optimization criterion of conventional ML algorithm by improving the active antenna index's reliability criterion of MMRC algorithm. Property simulation and complexity analysis show that the detection property of the active antenna index's reliability criterion of new algorithm is better than MMRC criterion and DBD criterion with a little low computation cost.