Authors

Lei WanFollow

Date of Completion

5-5-2014

Embargo Period

5-2-2014

Keywords

Underwater acoustic communication, OFDM, AMC, Doppler estimation, DSP implementation, DCC, field performance

Major Advisor

Shengli Zhou

Associate Advisor

Zhijie Shi

Associate Advisor

Peter Willett

Field of Study

Electrical Engineering

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

Multicarrier modulation in the form of orthogonal frequency division multiplexing (OFDM) has now been recognized as an appealing solution for high data rate communications over underwater acoustic channels with large delay spread. This dissertation covers three research topics: (i) transceiver algorithm design, (ii) real- time transceiver implementation and optimization on DSP platforms, and (iii) analysis of environmental impact on communication performance, unified under a common objective of bringing the underwater acoustic OFDM technology into practical systems.

On algorithm design for underwater acoustic OFDM, this thesis first investigates a key receiver module of Doppler scale estimation. We compare various Doppler scale estimators exploiting different OFDM signal structures. Second, this thesis investigates adaptive modulation and coding (AMC) for underwater acoustic OFDM, where the transmitter parameters adapt to time-varying channel conditions.

For DSP-based implementation, we first optimize the receiver algorithms to achieve real-time receiver processing. In this thesis, we consider two setups: a single transmitter and a single receiver, and two transmitters and two receivers. And we pursue both floating- and fixed-point implementations. Second, we implement an OFDM- modulated dynamic coded cooperation (DCC) in a three-node network with a source, a destination, and a relay.

Finally, this thesis analyzes the performance of underwater OFDM modems in a recent two-month deployment in the Chesapeake Bay. We correlate the receiver performance with environmental parameters, and also explore advanced offline receiver algorithms to process data sets that failed decoding during online operations.

COinS