Markovian approach to model Underwater Acoustic channel: Techniques comparison

Abstract

In the last years, Underwater Acoustic (UWA) sensor networks have exponentially grown in many scientific, industrial and research areas. Wireless underwater communications are required in many application fields, such as real time remote control of seabed and oil rigs, monitoring of underwater environments, collecting of scientific data recorded by stations on the seabed, conversation between divers, mapping of the seabed (in order either to detect objects or to discover new resources), prevention of disasters, and many others. In order to allow these applications, the aspect of physical phenomena affecting acoustic communications cannot be neglected. The shallow-water acoustic channel is different from the radio channels in many aspects. The available bandwidth of the UWA channel is limited and it depends on both range and frequency. Within this limited bandwidth, the acoustic signals are affected by time-varying multipath, which may creates severe inter symbol interference (ISI) and large Doppler shifts and spreads. These characteristics restrict the range and bandwidth for the reliable communications. Many works have already treated underwater acoustic channel modeling problem, however, at the best of our knowledge, they work only at the bit level and they are not suitable for those contexts in which a high level model is required. For this purpose, our paper discusses about a high level channel model based on Markov Chain approach for the underwater environment. Finite State Markov Model is developed for Packet Error Rate (PER) evaluation in an underwater channel, using the concept of error trace analysis. Some high level models well known in literature are compared to obtain statistical evaluations in order to find the model best fitting the underwater channel dynamics. Simulation and analysis are made in Matlab.