In the signal combining system of deep space network, the estimation error of time-delay between signals will reduce the effectiveness. The time-delay alignment technique based on combined output signal as the reference (CC-SUMPLE algorithm) makes use of the mutual information offered by multi-antenna and improves the alignment performance. However, it only takes ordinary cross-correlation into consideration rather than the cyclostationary of digital communication signal during calculating time-delay in the iterative process. As to this problem, this paper proposes multi-antenna signal time-delay alignment algorithm based on cyclostationary of communication signal (MCCC-SUMPLE algorithm) which reconstructs the combined reference signal and takes advantage of multi-cycle frequencies. The simulation results show that the proposed algorithm will improve the estimation accuracy and time-delay alignment performance compared with CC-SUMPLE algorithm.

In the wireless sensor network, every node is energy and resource limited and the received signal is weak, which requires less complexity and low computation signal combination algorithm to improve the signal-to-noise ratio (SNR) of the combined output signal. Considering the computation and time-varying delay, this paper proposes an adaptive feedback time-delay alignment method based on equalizer taps to decrease the SNR loss caused by time-delay difference. The relationship between the equalizer taps and time-delay difference is derived according to the Nyquist first criterion and the gradient descent algorithm is adopted to iteratively adjust the time delay of one antenna signal. The numerical results show that the algorithm can almost eliminate the time difference and the SNR loss decreases. The bit error rate (BER) performance is obviously improved.