Channel Estimation And Capacity Analysis Methods Of MIMO-OFDM System

MIMO-OFDM system is one of the hottest technologies of next generation communications, which has been widely used in broadband access, digital broadcasting and many other aspects. It combines the technical advantages of MIMO and OFDM to meet the requirements of higher performance. Many researchers have put forward related schemes from different aspects. One of the most important problems is channel estimation. Channel estimation is the key part in the design of receivers. For MIMO-OFDM system, users and antennas are much more than that in traditional systems, which leads to the increasing of channel number and complexity. Therefore, the channel estimation for MIMO-OFDM system which has strict performance requirements is more important than ever.In this thesis, MIMO-OFDM system model and its channel estimation methods are widely studied. Common channel estimation methods are generally divided into non-blind and blind channel estimation. To make use of their advantages and to avoid their defects, a new method named semi-blind estimation is applied. Combined with pilots and blind operation, it can improve the system’s real-time performance and reduce implementation difficulty.Although related algorithms are frequently found in various researches, but most of them are only available in limited field. Considering MIMO-OFDM system is a nonlinear system, a new method combined with radial basis function (RBF) neural network is constructed in this thesis. Researches show that intelligent signal processing methods can effectively deal with channel estimation in nonlinear system. Neural network, which is one of intelligent signal processing methods, can simulate nonlinear structure with high compatibility. Based on semi-blind estimation method, this thesis proposes the RBF neural network algorithm. It greatly improves the estimation accuracy. Then genetic algorithm is also been inserted into the proposed estimation algorithm. Results show that it significantly improves the estimation accuracy by weakening its real-time aspects.At last, this thesis studies on the application of the proposed estimation method in a variety of actual channel environments including cooperative communication environment, and the advantages of proposed estimation method is proved. It also made research on capacity analysis problem by means of simulation.