Probing techniques and estimation processes for fine-time synchronization of FH systems

For frequency-hopped communication systems, the hop patterns of the transmitter and receiver must be closely aligned in time to prevent performance degradation. Coarse-time synchronization is achieved when this misalignment is reduced to less than one-half hop period. Fine-time synchronization reduces this error to typically a few percent of a hop period. For satellite uplinks, where several noncoherent MFSK user signals are combined by FDMA, each user's transmitter must be synchronized with the satellite receiver. Because of the long round-trip delay between the transmitter and receiver, accurate estimates of the time error are made at the receiver and sent to the user, who then uses the estimate to adjust the phase of the transmitter hop period. Several methods are described which lead to biased, consistent estimators for the fine-time synchronization error. These methods are based upon estimation of the second-order moments of the envelope-detected, match-filtered, received signal. The performance of these estimators is determined as a function of the Gaussian noise level and of the number of samples L used to form the estimate. The estimator mean and standard deviation are approximated by means of the central limit theorem. Solutions compare favorably with simulation results. Results for signal-to-noise ratios of 0, 5, and 10 dB and with L between 1 and 512 show that these estimators are superior to previously known methods.