Compressed sensing-based channel estimation and prediction for underwater acoustic communications

Zhang, Yi (2017) Compressed sensing-based channel estimation and prediction for underwater acoustic communications. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (12MB)


This thesis develops approaches for estimating and predicting sparse shallow-water acoustic communication channels. The broadband shallow-water channel has three characterizations: a large dimension of channel impulse response caused by excessively long delay spread, fast temporal variability induced by scattering from the moving sea surface, and a sparse channel structure due to the resolvable paths. Traditional least square estimation techniques fail to utilize the sparse channel structure, and suffer from the limitations on the capability of estimating large-dimensional channels with rapid fluctuations. Compressed sensing also known as compressive sensing (CS), has been intensively studied recently. It has been applied in various areas such as imaging, radar, speech recognition, and data acquisition. Recently, applying CS to sparse channel estimation has been largely accepted. This thesis details the application of CS to sparse estimation of both time-invariant and time-varying shallow-water acoustic channels. Specifically, various reconstruction algorithms are used to find the sparse channel coefficients. However, a priori knowledge of channel sparsity is often not available in practice. The first part of the thesis proposes an improved greedy pursuit algorithm which iteratively identifies the sparse channel coefficients without requiring a priori knowledge of channel sparsity. Then, the proposed algorithm is employed to estimate both time-invariant and time-varying sparse channels. In addition, a comparative study of the state-of-the-art of various CS-based signal reconstruction algorithms is performed to gain better understanding of the mathematical insights. Furthermore, based on CS theory, different pilot placement choices will directly affect the performance of the channel estimation algorithm. The second part of the thesis investigates the pilot pattern design in sparse channel estimation. Unlike the equally spaced pilots for conventional channel estimation, randomly placed pilot tones are most used in existing CS-based channel estimation methods. In order to improve the efficiency of the optimal pilot pattern searching, a novel pilot pattern selection scheme is proposed based on the concatenated cyclic difference set. The performance of the proposed design is also compared with the existing search-based pilot placement methods. It should be noted that the proposed reconstruction algorithm and the pilot placement scheme are not restricted to underwater acoustic communication systems, but they can be applied so sparse channel estimation in other communication systems. Finally, an outdated channel estimation will lead to severe performance degradation when the channel varies rapidly. Hence, to predict future channel state information, an efficient sparse channel prediction scheme is proposed which does not require any statistical a priori knowledge of channels and noise. A receiver structure which combines a sparse channel estimator and a decision feedback based adaptive channel predictor is developed to further improve the prediction accuracy.Simulation results are shown to demonstrate the performance of the proposed algorithms and schemes. The study of this thesis contributes to a better understanding of the channel physical constraints on algorithm design and potential performance improvement.

Item Type: Thesis (Doctoral (PhD))
Item ID: 12818
Additional Information: Includes bibliographical references (pages 132-149).
Keywords: Underwater Acoustic Communications, Sparse Channel Estimation, Adaptive Channel Prediction, Compressed Sensing, Sparse Signal Reconstruction
Department(s): Engineering and Applied Science, Faculty of
Date: 13 July 2017
Date Type: Submission
Library of Congress Subject Heading: Underwater acoustic telemetry; Compressed sensing (Telecommunication)

Actions (login required)

View Item View Item


Downloads per month over the past year

View more statistics