Title: DSP: a deep learning based approach to extend the lifetime of wireless sensor networks

Authors: Jack Press; Suzan Arslanturk

Addresses: Department of Computer Science, Wayne State University, Detroit, MI, 48202, USA ' Department of Computer Science, Wayne State University, Detroit, MI, 48202, USA

Abstract: Wireless sensor networks (WSNs) equipped with batteries and solar panels enabled applications in areas such as environmental monitoring, agricultural, military, and medical systems. Research has shown that batteries often fail earlier than their projected lifetime. Sensor-nodes with solar panels placed in areas with sufficient sunlight can have their batteries recharged and can stay online for longer periods. However, areas with insufficient sunlight may need further adjustments. In this study, we present a dynamic sleep protocol (DSP) to forecast the lifetime of a sensor-node by dynamically adjusting the sleep period between transmissions. We have used a deep recurrent neural network (RNN) with long short term memory (LSTM) units to forecast the lifetime of the batteries and have discussed potential optimisation functions to adjust the sleep period. Our results have shown that an accurate identification of the battery lifetime with accurate adjustments would lead to longer operating hours without sacrificing the system performance.

Keywords: WSN; wireless sensor network; deep learning; machine learning; solar; battery; optimisation; scheduling.

DOI: 10.1504/IJSNET.2020.109718

International Journal of Sensor Networks, 2020 Vol.34 No.1, pp.38 - 48

Received: 24 Feb 2020
Accepted: 05 Mar 2020
Published online: 21 Sep 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article