Short Data Record Filtering for Adaptive Underwater Acoustic Communications

Spread-spectrum signaling offers robust communication over the multipath underwater acoustic channel, while at the same time allows multiple users to simultaneously occupy the entire continuum of the device-accessible bandwidth. However, limitations in temporal and spatial coherence of the underwater acoustic channel affect the maximum spreading code length and therefore the effective link data rate. In this work, we propose adaptive code length optimization toward high-rate spread-spectrum underwater acoustic communications. We first evaluate the bit error rate (BER) performance of the estimated minimum variance distortionless response (MVDR) and auxiliary-vector (AV) linear filter receivers. We observe that the AV-filter based receiver significantly outperforms the estimated MVDR solution under limited data-record availability. We then investigate data rate trade-offs for optimized code lengths under pre-defined BER constraints in time-varying underwater acoustic channels. Simulation results verify that the proposed adaptive packetization scheme offers superior data rate performance under short data-record constraints imposed by the non-stationary nature of the underwater environment.