Determining node duty cycle using Q-learning and linear regression for WSN
- PDF / 411,382 Bytes
- 7 Pages / 612.284 x 802.205 pts Page_size
- 63 Downloads / 198 Views
Determining node duty cycle using Q-learning and linear regression for WSN Han Yao HUANG, Kyung Tae KIM, Hee Yong YOUN College of Software, Sungkyunkwan University, Suwon 440-746, Korea c Higher Education Press 2020
Abstract Wireless sensor network (WSN) is effective for monitoring the target environment,which consists of a large number of sensor nodes of limited energy. An efficient medium access control (MAC) protocol is thus imperative to maximize the energy efficiency and performance of WSN. The most existing MAC protocols are based on the scheduling of sleep and active period of the nodes, and do not consider the relationship between the load condition and performance. In this paper a novel scheme is proposed to properly determine the duty cycle of the WSN nodes according to the load,which employs the Q-learning technique and function approximation with linear regression. This allows low-latency energy-efficient scheduling for a wide range of traffic conditions, and effectively overcomes the limitation of Q-learning with the problem of continuous state-action space. NS3 simulation reveals that the proposed scheme significantly improves the throughput, latency, and energy efficiency compared to the existing fully active scheme and S-MAC. Keywords wireless sensor network, media access control, duty-cycle scheduling, Q-learning, linear regression
1
Introduction
Wireless sensor network (WSN) has been used for a wide range of applications, including environment monitoring, smart space, medical system, and robotic exploration. WSN is comprised of a large number of autonomous electronic devices (called motes or sensors), having the capability of remote sensing, signal processing, and communication in an ad-hoc fashion [1]. The medium access control (MAC) layer in WSN is responsible for distributing the communication resources between the competing nodes, which is a main factor determining the network performance and energy efficiency. In the last two decades various MAC protocols have been developed for energy-efficient communication in WSN. The duty cycle-based MAC protocol is not effective for WSN as it may lead to increased delay. It even cannot be employed in some critical applications. A strong real-time performance is an important requirement in designing the MAC protocol for WSN. The previous approaches [2–6] adopted the idle listening Received May 2, 2019; accepted December 27, 2019 E-mail: [email protected]
technique as the main energy-saving approach, but it suffers from two problems. First, the transmission delay called sleep latency is increased as a packet needs to wait for the intermediate nodes to wake up as in S-MAC [7]. This is particularly serious in multi-hop transmission, where the transmission range of each node is smaller than the distance between the source and destination node. Second, the adaptability against variable traffic loads is not sufficient, where early sleep under high traffic load or over-listening under low traffic load may occur. Therefore, an efficient MAC protocol should be designed for WSNs, consi
Data Loading...
