This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. IEEE SYSTEMS JOURNAL 1 Demand-Based Coverage and Connectivity-Preserving Routing in Wireless Sensor Networks Hari Prabhat Gupta and S. V. Rao Abstract—An important issue of research in wireless sensor networks (WSNs) with dense and random deployment of sensors is to minimize the energy consumption while ensuring the desired coverage of the field of interest and connectivity of the network. In this paper, we present a demand-based coverage and connectivity- preserving routing protocol to provide desired coverage and con- nectivity requirements in WSNs. The protocol reduces the energy consumption by assigning the minimum required sensing range to the sensors and using a scheduling protocol to periodically turn off the communication radios of the sensors in a coordinated manner and a local route optimization with a power control technique. The proposed protocol is fully distributed and does not use any geographical information. Our simulations show that the proposed protocol effectively maintains the desired coverage and connectivity of the network and prolongs the network lifetime. Index Terms—Coverage, energy efficiency, routing, scheduling, wireless sensor network (WSN). I. I NTRODUCTION A TYPICAL wireless sensor network (WSN) consists of several tiny and low-power sensors that communicate with each other while they monitor a field of interest (FoI). WSNs find their applications in many areas that include bat- tlefield surveillance, environment monitoring, and forest fire detection [1]. Most of these applications use a random and large-scale deployment of low-cost sensors, where the sensors are spread across the FoI. In any application of a WSN that involves monitoring an FoI, sensing coverage or simply cov- erage is acknowledged as an important metric to measure the quality of service of the network. It specifies how well an FoI is monitored by the WSN. Any event that occurs in the FoI can be detected by a sensor if the location of the event is within its sensing region. A point in an FoI is said to be covered if it falls within the sensing region of at least one sensor. If every point in the FoI is covered, such coverage is termed as complete coverage in the literature [2]. Complete coverage indicates a high level of reliability yet is often difficult to be realized in practice [3]. Manuscript received February 4, 2014; revised April 25, 2014; accepted June 1, 2014. The authors are with the Department of Computer Science and Engineering, Indian Institute of Technology Guwahati, Guwahati 781 039, India (e-mail: g.hari@iitg.ernet.in; svrao@iitg.ernet.in). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/JSYST.2014.2333656 In certain applications such as forest fire detection and weather forecasting, the requirement of complete coverage may be too expensive or unnecessary. For example, in the summer season, the entire forest area may need to be covered, whereas in the other seasons, only subregions may need to be covered [4], [5]. When complete coverage is not necessary, WSNs can be made energy efficient by relaxing the quality of coverage (QoC). The problem of partial coverage refers to the relaxation in the QoC that requires only subregions of the FoI to be covered. Coverage ratio is the area covered by the sensors to the total area of the FoI [1]. In terms of the coverage ratio, the partial coverage problem requires that the coverage ratio be no less than a predefined threshold (smaller than unity). If the coverage ratio is desired to be unity, then it degenerates to the complete coverage problem [4]. In this paper, we address the partial coverage problem in WSNs, which has been also recently addressed in [6]–[8]. Connectivity is a complementary problem of coverage. Cov- erage quantifies the quality of monitoring an FoI, and connec- tivity indicates accessibility to the sensory data by the sink. Minimizing the energy consumed is an important issue to be addressed in WSNs because the batteries powering the sensors may not be accessible for recharging often [9]–[12]. There are various ways of reducing the energy consumption to prolong the network lifetime. One mechanism used to reduce energy expenditure is to periodically turn off the communication radios of the sensors in a coordinated manner. The sensors with the communication radio turned off are said to be in the redundant mode [3], [13]. However, the sensing units of the sensors remain active. Another approach used to reduce energy consumption is to minimize the transmission power needed to forward sensory data in WSNs [14]. In this paper, we study the coverage and connectivity prob- lem in WSNs. We assume that the sensors are deployed in the FoI at random and independent of each other. Each sensor can separately control the states of communication and sensing units, i.e., the state (ON or OFF) of the communication radio is independent from the sensing unit. This is an assumption widely used in the literature on the coverage and connectivity problem [13]. Major Contributions: The major contributions of our work in this area or research are as follows. 1) For the desired coverage maintenance, we propose an energy-efficient approach in WSNs. Currently, some commercially available sensors are capable of adjusting their sensing ranges to control the cost associated with 1932-8184 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.