Comparison of channel equalization filtering techniques in underwater acoustic communications

Abstract

In this thesis, underwater acoustic communications signal processing techniques, which are used to equalize the distortional effects associated with the ocean as a communications channel, are investigated for a shallow water ocean environment. The majority of current signal processing techniques employ a Finite Impulse Response (FIR) filter. Three equalization filters were investigated and presented as alternatives; they were the passive time-reversed filter, the inverse filter, and the Infinite Impulse Response (IIR) filter. The main advantage of the passive time-reversed filter and the inverse filter is simplicity of design. Bit error rates for the time-reversed filter were consistently around 10-1 and those for the inverse filter were greater than 10-1. However, inability of the passive time-reversed filter to completely eliminate multipath components and the ill-conditioned nature of the inverse filter made it difficult to achieve Probability of Error results below 10-1. Research into the development of an array receiver using a time-reversed filter should improve calculated bit error rates. Simulations of the IIR filter were conducted with limited success. The main advantage of an IIR filter is that fewer parameters are required in the design of the filter. However, the potential for instability in the filter is a significant limitation. Probability of Error results were found to be on the order of those for current FIR filters at close ranges. Unfortunately, instability issues arose for receivers as range from the source increased. This research on the IIR filter is still in the embryonic stage, whereas research using FIR filters is relatively highly developed. Further research is needed to address the issue of instability in IIR filters in order to make them an effective signal processing technique employable in underwater acoustic communications.