Bayesian Compressive Sensing Based Optimized Node Selection Scheme in Underwater Sensor Networks

Abstract

Information acquisition in underwater sensor networks is usually limited by energy and bandwidth. Fortunately, the received signal can be represented sparsely on some basis. Therefore, a compressed sensing method can be used to collect the information by selecting a subset of the total sensor nodes. The conventional compressed sensing scheme is to select some sensor nodes randomly. The network lifetime and the correlation of sensor nodes are not considered. Therefore, it is significant to adjust the sensor node selection scheme according to these factors for the superior performance. In this paper, an optimized sensor node selection scheme is given based on Bayesian estimation theory. The advantage of Bayesian estimation is to give the closed-form expression of posterior density function and error covariance matrix. The proposed optimization problem first aims at minimizing the mean square error (MSE) of Bayesian estimation based on a given error covariance matrix. Then, the non-convex optimization problem is transformed as a convex semidefinite programming problem by relaxing the constraints. Finally, the residual energy of each sensor node is taken into account as a constraint in the optimization problem. Simulation results demonstrate that the proposed scheme has better performance than a conventional compressed sensing scheme.

Node Selection Algorithm for Underwater Acoustic Sensor Network Based on Particle Swarm Optimization

“…The propagation loss TL is expressed in the form of dB as follows: (12) Among them, the first item represents the expansion loss and the second represents the absorption loss. It can be seen from equation (12) that the propagation loss TL is not only related to the transmission distance, but also the frequency of the signal has a certain influence [20], [21].…”

“…Therefore, it is significant to adjust the sensor node selection scheme according to these factors for the superior performance. An optimized sensor node selection scheme is given based on Bayesian estimation theory in paper [12]. In [13], the error of the equilateral triangle positioning area is studied.…”

Passive Source Localization Using Compressive Sensing

“…As such, it provides a framework to obtain sparse solutions to underdetermined problems as long as the underlying signal is sparse and the replica dictionary that maps the underlying signal to the observations is sufficiently incoherent. Owing to this ability, CS is widely and successfully used in many applications, e.g., control systems [13,14], magnetic resonance image reconstruction [15,16], computer vision [17], radar detection [18,19,20], geophysics and remote sensing [21,22], speech processing [23], image processing [24,25], error correction in channel coding and estimation [26,27], oceanic engineering [28], pattern recognition and machine learning [29,30], acoustic source localization [11,31,32,33,34], etc.…”

Construction of Measurement Matrix Based on Cyclic Direct Product and QR Decomposition for Sensing and Reconstruction of Underwater Echo

“…Its main attraction is that it can overcome the shortcomings of the traditional Nyquist sampling theorem [4] in signal processing, in terms of the large amount of data to be sampled and large data storage space required, because signal sampling and compression are performed simultaneously. CS is widely applied in medical image processing [5], image compression [6,7], pattern recognition [8,9], and radar detection [10]; it can also be combined with the Bayesian algorithm [11,12], e.g., to reconstruct underwater echoes. CS involves three key components: (1) the sparse representation of the signal [13]; (2) the construction of a measurement matrix [14]; and (3) signal reconstruction [15,16].…”