Modifying SNR-independent velocity estimation method to make it suitable for SNR estimation in shallow water acoustic communication

Ref. 3 gives SNR-independent velocity estimation method. Subsection 1.2 of the full paper discusses the signal model for shallow-water multi-path acoustic channel. Taking full advantage of the fact that carrier frequency is very much lower in water than that in air, subsection 2.2 simplifies greatly Ref. 3 and derives eqs. (15) and (16) needed for implementing the algorithm of SNR estimation in shallow-water acoustic communication. Subsection 3.2 gives shallow-water lake test results in Figs. 3, 4 and 7. Using the lake test results, subsection 3.2 gives computer simulation results in Figs. 5, 6, 8 and 9. The computer simulation results show preliminarily that: (1) the mean estimation error is less than 1 dB for SNR ranging from -10 dB to 10 dB; (2) the estimation results in shallow-water acoustic channel are independent of the number of paths.