Soft-output Detection For Mimo Wireless Communication Systems Research

Since MIMO technique came out, it has been one of the key techniques in the next generation mobile communication. It can enhance spectral efficiency several times without increasing bandwidth. Because of that, MIMO technique is used widly. In coded MIMO system, it is a key part to find out a soft detection way with good performance and low complexity for real use on a large scale.Although MIMO detection algorithms have been researched widly and deeply, there are still many problems unsolved. This dissertation mainly studies about getting the soft decision in coded MIMO system with the combination of LLL basis reduction and similar MAP detection. A full and detailed introduction of LLL basis reduction algorithm is given in this dissertation and this technique can make the performance of hard decision better. Based on a better hard decision, three ways are given to get the soft decision using the resembling MAP detection way. Through the computer simulation tests, performance of the third way is very close to the original MAP's and the complexity is far less than that. At last, this way of detection is used in a real parameter defined MIMO system.Chapter 1 gives an introduction of MIMO system model and the detection algorithms.The most recent studies about MIMO technique and the applications of it are summarized. The significance of soft decision in coded MIMO system is also made clear.Chapter 2 introduces two types of MIMO detection: hard detection and soft detection. The formulas used in soft detection are given and the MAP detection algorithm is introduced in detail. The idear of combinating the hard decision and MAP detection way is mentioned.Chapter 3 gives the LLL basis reduction algorithm at first, three soft detection way based on hard decision and resembling MAP detection are investigated.The performance of these three ways are compared. Then the main detection algorithms are introduced as well as the mathmatic model of LLL reduction MIMO system. The simulation of the third way based on this model is given. The simulation results and theory analysis show that the performance of the third way is very close to the MAP detection but with a less complexity.Chapter 4 gives the real parameter MIMO system defined by IEEE 802.11and 802.16d criterion. The soft detection way of combinating the hard decision and similar MAP detection is tested in this system. From the results of computer simulation, it is known that the information lost in soft detection is much less than the hard one.Chapter 5 summarizes this dissertation and gives the future research direction and the next step of work for soft detection algorithms in MIMO system.In this dissertation, it is known that the performance and complexity of detection algorithm are the key factor to decide wether the MIMO technique could be used on a large scale or not. LLL basis reduction is used to optimize the BER performance of the hard decision algorithms.And the soft detection way mentioned in this dissertation could banlance the performance and complexity much better.