Movement-Efficient Sensor Deployment in Wireless Sensor Networks

Abstract

We study a mobile wireless sensor network (MWSN) consisting of multiple mobile sensors or robots. Two key issues in MWSNs energy consumption, which is dominated by sensor movement, and sensing coverage have attracted plenty of attention, but the interaction of these issues is not well studied. To take both sensing coverage and movement energy consumption into consideration, we model the sensor deployment problem as a constrained source coding problem. Our goal is to find an optimal sensor deployment to maximize the sensing coverage with specific energy constraints. We derive necessary conditions to the optimal sensor deployment with (i) total energy constraint and (ii) network lifetime constraint. Using these necessary conditions, we design Lloyd-like algorithms to provide a trade-off between sensing coverage and energy consumption. Simulation results show that our algorithms outperform the existing relocation algorithms.

Movement-Efficient Sensor Deployment in Wireless Sensor Networks With Limited Communication Range

“…June 20,2018 To evaluate the sensing uncertainty in heterogeneous MWSNs, we consider the Centroidal Vonoroi Tessellation function [1], [6], [7], [11], [12] defined as…”

“…In this section, we extend the proposed algorithms, CCML and DCML, to other kinds of sensing tasks: area coverage and target coverage. We employ the binary coverage model [1]- [15] in which Sensor n can only detect the points within its sensing range r n . Intuitively, in order to decrease the sensing uncertainty, CCML and DCML deploy sensors into high-density regions, and thus the points with high density are more likely to be covered.…”

“…It is mainly used when there is no specific target. The widely used CVT model (see more details in Section II) can be presented as a quantizer with the sensing uncertainty as its distortion [1]- [12], [16]- [20].…”

Quantizers with Parameterized Distortion Measures

“…Since our goal is to minimize the average transmission power (6) we define the nth parameter distortion function as D(ω, q n , a n , b n ) = β · a n q n − ω 2 2 + b n γ (8) where a n = h −1/γ n and b n = h 2−1/γ n . As can be seen from (8), the distortion is a function of the parameter h n in addition to the distance between the reproduction point q n and the represented point ω. From a quantization point of view, one can start with the distortion function (8) without knowing the UAV power consumption formulas in this section.…”

“…where the generalized Voronoi regions V n (Q, h) are defined as the set of sample points ω with smallest distortion to the nth quantization point q n with parameter h n . Minimizing the average distortionD(Q, h, V) over all parameter and quantization points can be seen as an N −facility locational-parameter optimization problem [6]- [8], [20]. By the definition of the Voronoi regions (10), this is equivalent to the minimum average distortion over all N −level parameter quantizers…”