Improving the performance of spectrally efficient MIMO OFDM system through channel estimation using optimization techniques

Abstract

Multiple Input and Multiple Output - Orthogonal Frequency Division Multiplexing (MIMO-OFDM) has fascinated trivial attention, and has become a potential technique for high data rate with high robustness, in wireless communication systems. It is difficult to achieve high spectral efficiency and performance simultaneously, over fast fading channels. To attend to the above, Time Frequency Training - Orthogonal Frequency Division Multiplexing (TFT-OFDM)is suggested and its performance is obtained. The group pilots address the spectral efficiency and the joint channel estimation maintains the performance of the system. The Least Square (LS) estimation exhibits simple estimation procedure, but its performance is limited and Minimum Mean Square Error (MMSE) estimation method considerably reduces the mean squareerror, but computational complexity is high. In TFT-OFDM, the channel is estimated by utilizingfrequency domain and time domain pilots, named joint time and frequency channel estimation. This thesis presented the performance of the TFT-OFDM with the conventional CP-OFDM. Further improvement of the performances of such systems is critical to practical applications and is achieved by the use of some intelligent algorithms proposed based on local optimization, defining fitness function and object function. Such intelligent algorithms can solve the optimization problems, such as calculating the maximum value or the minimum value of a function in a solution space. In signal detection, anintelligent algorithm takes the maximum likelihood function as the objective function, and the solution space are all possible transmit signal vectors. newline newline newline