The Design And Study Of Signal Detection Of Massive MIMO

Massive MIMO is seen as the main technology in 5G network. Thanks to the contribution of its pioneer technology, Massive MIMO is praised by many scientists and engineers. In the eyes of many people, Massive MIMO is hopefully to be the core of 5G.The signal detection plays an important role in MIMO. The research of MIMO signal detection is becoming more and more important. A good signal detection algorithm can greatly improve the performance of MIMO system. During the improvement of MIMO, it is important to make sure the relationship between the numbers of antennas and performance of the system.This paper compared two signal detection algorithms, ZF and MMSE, by testing their performance of BER in different conditions. Do the research about the influence between sharply increasing numbers of antennas and the performance of MIMO system. The conclusion of the match of signal detection algorithm and the best performance of Massive MIMO will be drawn.The technology in hardware plays a very important role in many technologies in 5G. It is easy to realize that the operating speed of real number field is faster than the complex number field’s, while most algorithms of signal detection are in complex number field. A kind of Sphere Decoding algorithm in real number field solves this problem in some way. This SD algorithm has some disadvantages in the same time, such as the choosing of original radius and the way of decreasing the radius. This paper raises a kind of Variable Radius-Sphere Decoding to solve the problems about SD algorithm.To decrease the computation complexity of SD algorithm and improve the method of decreasing radius, the VR-SD algorithm raises a kind of method. Instead of decreasing the radius little by little, the method is about to find the target signal point in a variable way in order to decrease the cost in time of the system.