Research And Implementation On Channel Estimation For OFDM Systems

In Orthogonal Frequency Division Multiplexing(OFDM) systems, the performance of the receiver highly depends on the accuracy of channel estimation. Thus improving the accuracy of channel estimation is critical in OFDM system design. This thesis researches channel estimation from both theoretical and practical perspectives. First, conventional channel estimation algorithms including Least Square(LS) and Linear Minimum Mean Square Error(LMMSE) algorithms are analyzed and compared. Then the defects and impracticality in practical systems of these algorithms are found. Finally, an enhanced LMMSE channel estimation algorithm which is more suitable for practical use, and a uniform LMMSE method which is suitable for non-ideal synchronization system are proposed. The proposed algorithms are proved of significant performance gain over conventional algorithms.In addition, a hardware platform based on picoChip is built. Different channel estimation algorithms including LS and uniform LMMSE algorithms are implemented in this platform and the performance of each is tested. The results indicate that the uniform LMMSE algorithm is a practical algorithm meeting the requirements of practical systems on performance, complexity and robustness.