Skip to Main Content
In this paper, frequency and time correlation of the underwater channel are exploited to obtain a low-complexity adaptive channel estimation algorithm for multiple-input-multiple-output (MIMO) spatial multiplexing of independent data streams. The algorithm is coupled with nonuniform Doppler prediction and tracking, which enable decision-directed operation and reduces the overhead. Performance is demonstrated on experimental data recorded in several shallow-water channels over distances on the order of 1 km. Nearly error-free performance is observed for two and four transmitters with BCH(64,10) encoded quadrature phase-shift keying (QPSK) signals. With a 24-kHz bandwidth, overall data rates of up to 23 kb/s after coding were achieved with 2048 carriers. Good results have also been observed in two other experiments with varying MIMO-OFDM (orthogonal frequency-division multiplexing) configurations.