دانلود رایگان ترجمه مقاله زمانبندی آگاهی از پوشش در شبکه های حسگر بی سیم (نشریه الزویر 2015)

زمانبندی آگاهی از پوشش در شبکه های حسگر بی سیم: یک رویکرد قرار گیری بهینه

Coverage Aware Scheduling in Wireless Sensor Networks: An Optimal Placement Approach

Abstract

Area coverage is an important issue in wireless sensor networks, which reflects how well an area is monitored or tracked by sensors. But, since a sensor network has restricted energy resources, energy efficiency is vital for this area coverage. One of the most efficient solutions to provide energy-aware area coverage is scheduling. That is, without any assumption about sensors’ locations, only a distributed and parallel scheduling method determines which sensors should be on and which ones should be off in each decision period. The ultimate objective is to maximize network lifetime and keeping a target level of area coverage. A major part of the algorithms proposed in this field, schedule a sensor node activity based on its neighbors’ information. Such information includes the distances of a node from its neighbors, the number of its active neighbors, etc. Indeed, message exchange is essential in the implementation of these algorithms which causes to increased energy consumption. In this paper, we propose a distributed scheduling algorithm, at which, each node itself decides to make its sensor on or off based on its location information and the node density over the target area. For this purpose, we first compute the minimum number of nodes that are enough to cover the target area. Then we obtain the best locations for theses nodes. Based on these computed location the area is partitioned into some sub-area, each one coverable by only one sensor. Then in each sub-area, a local scheduling procedure schedules the activation order of sensor. Simulation results show that the proposed algorithm, called CAOP, can maximize the network lifetime while maintaining complete area coverage.

1 Introduction

Wireless sensor networks (WSNs) consist of small-sized sensor nodes [1] and have been used in many applications such as vehicle tracking, danger alert and battlefield surveillance. Generally the wireless sensor nodes are equipped with finite power batteries that can be deployed cheaply and rapidly for monitoring the environment events. In such a network replacing node battery is impossible, hence, in order to prolong the network lifetime and to keep a high degree of reliability, a type of redundancy is provides by deploying a high density of sensors [2]. But, if all the sensor nodes are active at the same time, a major part of nodes’ energy will be consumed due to an extra amount of message exchanges. Obviously, this can lead to fast depletion of nodes’ energy and ending the network life. To prevent the nodes from unnecessary activities, we can use some scheduling algorithms. Scheduling algorithms are applied to schedule nodes to be activated alternatively to save the overall energy of the system. These scheduling algorithms are often aware of some design goals such as connectivity, coverage, energy saving and so on. That is, since the scheduling policy makes some sensors on and makes others off, naturally it may cause to reduced connectivity and coverage. Hence, a subset of nodes that are kept active for sensing and communication must ensure coverage, connectivity and other design objectives.

The coverage concept is one of the most fundamental issues in WSNs, which directly affects the quality of service (QoS). The coverage indicates how well the deployed sensor nodes can track a set of targets. Sensor activation scheduling under constraint on covering of targets is called the coverage problem in the literature. Generally, there exist three types of coverage in a WSN: point coverage, barrier coverage and area coverage [3]. Point coverage refers that the separate target points can be covered at any time [4]. The objective of barrier coverage is to minimize the probability of undetected penetration through the sensor network [3]. In the area coverage problem, the goal is to cover some areas. In all the proposed solutions the goal is to find a minimum set of active nodes that can maintain coverage and connectivity of the network in a suitable level while the consumption of the energy by the sensor nodes is as balanced as possible.

By far, many coverage scheduling algorithms have been proposed. Some of these algorithms [5] assume that the nodes know their locations by using some devices or techniques. Others [6] adopt that the distances between nodes can be achieved by using the received signal strengths. Some others [7] consider some mobile nodes with controllable mobility. Some other researchers have also suggested coverage control algorithms without usage of any location, distance or angle information of the sensor nodes.

From other point of view, a sensor activity scheduling algorithm can be either centralized or distributed. A distributed scheduling protocol can be easily expanded to largescale sensor networks, so it is more desirable. It is often assumed in these protocols that the time is divided into rounds. At the beginning of each round, all sensor nodes wake up and must make their activity decisions the current round. This decision stage is based on the message exchanges. Moreover, at the end of a round, all sensors must be activated and decision stage is required to be repeated in the next round. Athwart, the power consumption rate of the message exchanges may be very high, if the network scale is great.

This paper assumes that any sensor node knows its location and then tries to design an energy-aware area coverage scheduling algorithm. The proposed algorithm follows a distributed approach and is called Coverage Aware scheduling for Optimal Placement of sensors (CAOP). Its design goal is to maintain area coverage and to prolong the network lifetime by efficient energy consumption. Rest of this paper is organized as follows. In the next section, some existing schedulingbased algorithms in WSN coverage problem are reviewed. Section 3 brings the required preliminaries for this work. In Sect. 3, the proposed algorithm (CAOP) is given. Section 4 evaluates the performance of the proposed algorithm and compares its outcomes with other related works. Finally, in Sect. 5, this paper is concluded.