Abstract:
Robust high data rate wireless communication has several challenges. Transmission of high rate information typically experiences higher delay spread in mobile environments. Furthermore, high mobility introduces time-variations which can make the transmission link less reliable.
Orthogonal Frequency Division Multiplexing (OFDM) is suitable for high delay spread applications. However, performance of OFDM systems, is affected by timing synchronization. Timing synchronization becomes challenging in high mobility applications as power delay profile of the channel can change rapidly due to the sporadic birth and death of the paths. Timing synchronization errors introduce Inter-Carrier-Interference (ICI) in OFDM systems. Furthermore, for mobile applications time-variations in one OFDM symbol introduce ICI as well, which further degrades the system performance. Therefore to have an acceptable reception quality for the applications that experience high delay and Doppler spread, design of robust timing synchronizer is essential. This thesis takes an overall look at this issue.
In this dissertation, we consider symbol timing estimation problem for OFDM systems. OFDM systems are known to be sensitive to time synchronization errors and improving the accuracy of timing offset estimation can help improve the overall system performance. We presented a Maximum-likelihood (ML) estimator method that uses the redundant information within the Cyclic Prefix (CP), which gives the coarse estimation of symbol time offset.
We also discussed the Hybrid ML estimation method, which exploits both the cyclic prefix and the pilot symbols for accurate symbol timing estimation. The performance results of the above discussed methods for AWGN and dispersive channels are simulated in the MATLAB. Simulation results shows that pilot signal design algorithms decrease the variance of the estimate, especially in a time dispersive channel and performs better as compared with the equally spaced pilot signal design algorithms.
