Energy Management and Power Allocation for Underwater Acoustic Sensor Network

This paper investigates energy allocation in underwater acoustic nodes powered by energy harvesting. Our goal is to maximize the expected total amount of delivered data over a finite time slots. Two scenarios are considered for different knowledge levels of channel state information (CSI). In one sc...

Full description

Saved in:
Bibliographic Details
Published inIEEE sensors journal Vol. 17; no. 19; pp. 6451 - 6462
Main Authors Jing, Lianyou, He, Chengbing, Huang, Jianguo, Ding, Zhi
Format Journal Article
LanguageEnglish
Published New York IEEE 01.10.2017
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Acoustics
Batteries
Complexity
Data transmission
Dynamic programming
energy allocation
Energy harvesting
Energy management
Energy transmission
Markov chains
Markov processes
Resource management
Sensors
Transmitters
underwater acoustic sensor network
Underwater acoustics
Online AccessGet full text

Cover

More Information
Summary:This paper investigates energy allocation in underwater acoustic nodes powered by energy harvesting. Our goal is to maximize the expected total amount of delivered data over a finite time slots. Two scenarios are considered for different knowledge levels of channel state information (CSI). In one scenario, the transmitter receives CSI at the end of each data transmission epoch and we consider the energy allocation problem for sensing and transmission. In the second scenario, the transmitter receives delayed CSI after multiple time slots, while only the energy allocation for transmission is considered. The underwater acoustic channel is modeled as a finite-state Markov chain to characterize its time varying nature. We employ a stochastic dynamic programming (DP) approach to derive the optimal allocation policy for both scenarios. To reduce the inherent computation complexity of DP approach, we also present a suboptimal algorithm by analyzing the structure of DP solution. Our results show the DP approach achieves substantial performance improvement that is preserved substantially by the suboptimal algorithm to provide a good performance-complexity tradeoff.
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2017.2737229