Sharpening timing-metrics for auto-correlation based coarse symbol synchronization in OFDM systems

The auto-correlation based coarse symbol timing scheme yields good correlation timing-metric for wireless orthogonal frequency division multiplexing (OFDM) systems under independent Rayleigh fading channels. However, it generates a plateau in its timing-metric due to the structure of the preamble added with cyclic prefix (CP). Inherently, the mean square error (MSE) performance of a coarse symbol timing scheme is closely related to the sharpness of its main-lobe timing-metric. Some of the literatures focus on polishing the auto-correlation timing-metrics to find the frame start point. One of the well-known methods uses arithmetic mean of the auto-correlation timing-metrics to generate a peak after post-processing. We propose a post-processing scheme using weighted arithmetic mean of the auto-correlation timing-metrics, and provide a different viewpoint of analyzing the performance of post-processing schemes in additive white Gaussian noise (AWGN) channels, which may be helpful in timing-metric design. By using the knowledge of root mean square (rms) delay spread, we get a general coarse symbol timing scheme for both AWGN and Rayleigh fading channels. The obtained simulation results indicate that the MSE performance of the post-processing scheme is a function of the main-lobe sharpness.