ON THE USE OF A JANUS-INSPIRED DATA FRAME TO ESTIMATE CHANNEL CHARACTERISTICS FOR ADAPTIVE COMMUNICATION SYSTEMS

Abstract

The JANUS protocol is being used widely in underwater acoustic sensor networks for various applications including but not limited to mammal habitat monitoring, event detection, submarine communication and surveillance due to their reliability. JANUS was designed with error correction techniques and robust modulation schemes to make them immune to the harsh channel conditions. In this thesis, a broadcast message that incorporates various features of the JANUS standard is designed to obtain an average estimate of the channel impairments in an underwater communication link. The discovered channel parameters such as signal-to-noise ratio (SNR), channel impulse response (CIR), Doppler shift, Doppler spread, and RMS delay spread can be effectively utilized to define the transmission parameters for a reconfigurable physical layer for an underwater communication network. The thesis describes the algorithms used to estimate the gain, multipath properties and Doppler spread within the channel. These algorithms are then demonstrated over a simulated data as well as on a measured data-set for a communication link with a bandwidth of 5 kHz at 27.5 kHz. The estimated channel data provides valuable information for estimating the achievable throughput in a channel, considering factors such as the number of subcarriers and modulation ratio. Conversely, for a desired throughput, the estimated channel data can also be used to determine the optimal number of subcarriers for a multi-carrier modulation system.