A Novel Continuous Object Tracking Scheme for Energy-Constrained Wireless Sensor Networks

Abstract

In order to monitor continuously moving phenomena such as wile fire and hazardous bio-chemical material in wireless sensor network, boundary tracking approach has been widely used by reason of its huge scale and extensive diffusion property. With the boundary tracking scheme, the energy efficiency is expected to improve if only sensor nodes near the boundary of continuous object actively participate in boundary tracking process, while other sensor nodes stay in sleep mode for energy saving. In this paper, we propose a predictive continuous object tracking scheme, which uses minimum set of active sensing nodes to reduce energy consumption. The proposed scheme predicts the future boundary line, which provides the knowledge for a wakeup mechanism to decide which sleeping nodes need to be activated for future tracking. The proposed algorithm is verified with simulation results that total energy consumption can be dramatically reduced under acceptable boundary detection accuracy. Index Terms-Continuous object and boundary tracking

Estimating the Spatiotemporal Evolution Characteristics of Diffusive Hazards Using Wireless Sensor Networks

“…Having computed the mixture weights using (8), Master S M i calculates the Normal distribution parametersû i and s 2 i using equations (11) and (12). By applying simple manipulations on the Bayes theorem it can be proved [26] that since the prior N (u i , s 2 i ) and the likelihood N (û i ,ŝ 2 i ) are both Gaussian, the posterior will also be a Gaussian N (u * i , s * 2 i ) (conjugate distributions, see Figure 4b) with parameters provided by the following easy to compute closed form expressions:…”

“…These schemes do not attempt to estimate the local front line's evolution characteristics or predict their spatiotemporal evolution. One notable exception is the work in [11], [12] where the authors introduced a simple way to estimate, as we do, the speed and direction of the local front. They use them to implement a "wake up" mechanism to decide which "sleeping" nodes to activate selectively for near term front tracking in order to reduce the network's overall energy consumption.…”

“…WSNs have also been used for single and multiple target tracking [1]- [3], and due to their rapidly dropping cost they are also gaining popularity in environmental monitoring applications [4]. Recently, sensor network-based methods have been proposed for detecting the boundaries of diffusive hazardous phenomena [5]- [12], modeled as "continuous objects". However, traditional target tracking algorithms cannot be applied for continuous object tracking, since the two problems are fundamentally different.…”

“…Although these methods can estimate implicitly the boundaries of an evolving hazard, they require unrealistic network densities (thousands of deployed sensor nodes per km 2 ) which renders them impractical even for small scale environmental monitoring applications. Furthermore, the reported algorithms (with few exceptions as those in [11], [12]) do not provide information about the spatiotemporal evolution characteristics (e.g. direction and speed) of the diffusing phenomenon and therefore cannot be exploited directly to make valuable predictions.…”

Continuous Object Monitoring Based on Wireless Sensor Network and RBF

“…When these states are studied, these objects are generally looked at continuous objects. Continuous objects [4][5][6] generally refers to the volume and shape can not be ignored, cannot be regarded as the object of particles. The main task of monitoring the continuous object is to monitor the boundary of the continuous object.…”