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

Abstract

There is a fast growing interest in exploiting Wireless Sensor Networks (WSNs) for tracking the boundaries and predicting the evolution properties of diffusive hazardous phenomena (e.g. wildfires, oil slicks etc.) often modeled as "continuous objects". We present a novel distributed algorithm for estimating and tracking the local evolution characteristics of continuous objects. The hazard's front line is approximated as a set of line segments, and the spatiotemporal evolution of each segment is modeled by a small number of parameters (orientation, direction and speed of motion). As the hazard approaches, these parameters are re-estimated using adhoc clusters (triplets) of collaborating sensor nodes. Parameters updating is based on algebraic closed-form expressions resulting from the analytical solution of a Bayesian estimation problem. Therefore, it can be implemented by microprocessors of the WSN nodes, while respecting their limited processing capabilities and strict energy constraints. Extensive computer simulations demonstrate the ability of the proposed distributed algorithm to estimate accurately the evolution characteristics of complex hazard fronts under different conditions by using reasonably dense WSNs. The proposed in-network processing scheme does not require sensor node clocks synchronization and is shown to be robust to sensor node failures and communication link failures, which are expected in harsh environments.! 2 delineate the area affected by the diffusive hazard, we present in this paper a novel decentralized algorithm which can estimate with accuracy, using dynamically formed clusters (triplets) of cooperating sensor nodes, the local evolution characteristics (orientation, direction and speed) of a continuous object. The updating of the evolution parameters is based on a Bayesian probabilistic modeling approach which relatively to our prior work ( [13], [14]) : (i) Casts the problem in a framework allowing us to account for the sensing mechanism uncertainties expected in harsh environments and also characterize the uncertainty of the estimated parameters, (ii) Improves the accuracy of the obtained local model parameter estimates, (iii) Leads to simpler algebraic expressions for updating these parameters that can be easily implemented by the commonly used processing-and power-constraint embedded microprocessors of WSN nodes, (iv) Takes into account the possibility of imperfect sensor nodes which may fail to communicate since the approaching hazard may impair their functionality.With respect to related work on predictive modeling ( [11], [12]) our approach exhibits the following advantages: It can track accurately the time varying characteristics of a local front line even if, (i) the WSN is not dense, (ii) the sensor node clocks are not synchronized, (iii) the sensing mechanism is imperfect, (iv) nodes and communication links may fail as the hazard approaches. The ability to track the spatiotemporal evolution characteristics of the local front enables making predictions about its future loca...

A Distributed Multi-Tier Emergency Alerting System Exploiting Sensors-Based Event Detection to Support Smart City Applications

“…In this scope of problem, prediction, rescuing and alerting procedures are some of the services that have been supported by sensor network initiatives. In this sense, many solutions have been proposed, addressing different challenges and particularities of critical situations in smart cities [45][46][47][48]. While promising, however, much of such solutions could be enhanced by exploiting the services provided by CityAlarm.…”

Boundary Tracking of Continuous Objects Based on Binary Tree Structured SVM for Industrial Wireless Sensor Networks

“…Recently, research efforts around the globe are focusing on exploiting the fine-grained perception of IWSNs to track the movement of continuous objects [8] [9]. Detecting the boundary of diffusing continuous objects by identifying the sensor nodes located closest to the evolving front line attracts intensive attention [10]- [12].…”

Predictive Tracking of Continuous Object Boundaries Using Sparse Local Estimates

“…The boundary (global front) of a continuous object can be approximated by a set of line segments (local fronts) and each segment's evolution can be represented by a model with a small number of parameters (e.g. orientation, direction, and speed), as shown in [27]- [29]. In our prior work we have introduced a probabilistic framework and associated distributed algorithms that can estimate the local evolution characteristics of a propagating hazard's front using WSNs of realistic density.…”

“…Moreover, we have shown how the sensor nodes can update the local front model parameters using closed-form expressions (analytical solutions of a Bayesian parameter estimation problem) that are easy to implement using their commodity microprocessors [30]. Importantly, a unique feature of our WSN-based environmental monitoring methodology is that each sensor node can also estimate the uncertainty associated with the local front estimates it produces as the hazard progresses [29].…”