Deployment strategies in the wireless sensor network: A comprehensive review

Abstract

Wireless Sensor Networks (WSNs) have come across several challenges such as node deployment, the reduction of power consumption, secure routing, bandwidth allocation, Quality of Service (QoS), and so forth. Since sensor deployment is an important matter due to its influence on cost and the network capability of WSN, the focus of this study is the deployment issue and related concerns such as coverage, connectivity, and energy efficiency, which have a great impact on the performance of WSNs. To the best of our knowledge, there are no studies that analyze and review the current scope completely. In this paper, some important research in the scope of sensor deployment will be investigated and analyzed as well as identifying their main specification. The deployment problem is classified based on few important factors and four deployment strategies and their related results are studied in each class. Also, the advantages and disadvantages along with important challenges of several strategies have been discussed so that more efficient deployment strategies can be developed in future.

Introduction

With the development of distribution environments such as the grid computing [30], [44], [70], cloud computing [12], [43], [47], Peer-to-Peer networks [46], Expert Cloud [8], [26], [28], [43], [50], [54], electronic management [52], [53], [68], [84], [10], knowledge management [85], [24] and MapReduce [45], nowadays, Wireless Sensor Network (WSN) becomes more popular than before. It is an emerging paradigm of computing and networking, which can be defined as a network of minuscule, diminutive, inexpensive, and keenly intellective devices, called sensor nodes. Sensor nodes are spatially distributed and they work cooperatively to communicate information gathered from the monitored field through wireless links and send them to a sink, which either uses the data locally or communicates it to other networks [59]. The development of WSNs was originally motivated by military applications [78], for example in battlefield surveillance they could be used to detect, locate, or track enemy movements. WSNs are currently employed in many industrial and civilian application areas including industrial process monitoring and control [57], environment and habitat monitoring [5], [15], [79], healthcare applications [4], home automation [66], and traffic control [42]. In the case of natural disasters [89], sensor nodes can sense and detect the environment to forecast disasters in advance. The wide range of potential WSN applications call for a rapidly growing multi-billion dollar market, but this would require further major progress in WSN standards and technologies to support new applications [21]. Despite the continuous development of WSNs, there are still several research challenges related to wireless sensor communication due to the restricted features of low priced sensor node hardware and the common necessity for the nodes to work for long time periods. Besides, there are challenges originating from close interaction between the WSN and the environment that need to be investigated, namely uncertainty related to sensors readings, harsh deployment environments and combining sensory data from multiple sensors [64].

For most WSNs, a major design step is to selectively decide the locations of the sensors in order to maximize the covered area of the targeted region. This particular problem has different appellations in the literature, e.g. placement, coverage or the deployment problem in WSNs [32]. The deployment of sensors can be random (e.g. dropping sensors in a hostile terrain or a disaster area) or deterministic (e.g. placing sensors along a pipeline to monitor pressure and/or temperature, and boundary surveillance) [27], and it depends mainly on the type of application, the environment, and the sensors themselves. The planning strategy of the deployment problem affects transmission rate of the sensors as well as the coverage and lifetime of the whole system, making the deployment a very critical issue in WSNs [74].

In general, poor deployment of sensor nodes leads to inefficient network connectivity or redundancy of coverage. A well-chosen deployment strategy will not only reduce cost but also extend the network lifetime; therefore, the deployment of a WSN is a critical problem. Deployment planning requires consideration of several objectives such as energy consumption, sensing coverage, network lifetime, network connectivity, and so forth. Often these objectives conflict with one another, and operational trade-offs must be established during network design [76].

In this paper, we classify the deployment methods and algorithms proposed in the literature for both predetermined and random deployments. This classification is based on the most important objectives used for modeling and solving the deployment problem. The deployment strategies are classified into four main categories: increasing the coverage, enhancing the connectivity, improving energy efficiency and optimizing the lifetime, and finally multi-objective deployments. To the best of our knowledge, a comparative study on the deployment issue considering these categories has not been conducted as of yet. [81] categorizes different approaches based on their node positioning techniques (static vs. dynamic [41]), and compares them based on their objectives and methodologies. However, it does not consider multi-objectivity of the strategies. [88] investigate the coverage and connectivity issues from different aspects of deployment strategy, sleep schedule mechanism and coverage radius. We have investigated state-of-the-art strategies in each category and depicted their advantages and disadvantages. We have compared all presented strategies based on some important factors regarding deploying sensors, such as load balancing, energy distribution, scalability, sensor's sensing range, a region of interest, network cost and so on. As well, a side-by-side comparison of all discussed strategies is presented, and few open issues are addressed.

The basic concepts and preliminaries are provided in the next section. Section 3 discusses related papers and research in the scope of sensor deployment under four main categories. The taxonomy and comparison of analyzed strategies are presented in Section 4. Section 5 maps out open-ended issues and finally Section 6 concludes the paper.