Agent-based decentralised coordination for sensor networks using the max-sum algorithm

A. Farinelli, A. Rogers, N. R. Jennings
2013 Autonomous Agents and Multi-Agent Systems  
In this paper, we consider the generic problem of how a network of physically distributed, computationally constrained devices can make coordinated decisions to maximise the effectiveness of the whole sensor network. In particular, we propose a new agent-based representation of the problem, based on the factor graph, and use state-of-the-art DCOP heuristics (i.e., DSA and the max-sum algorithm) to generate sub-optimal solutions. In more detail, we formally model a specific real-world problem
more » ... re energy-harvesting sensors are deployed within an urban environment to detect vehicle movements. The sensors coordinate their sense/sleep schedules, maintaining energy neutral operation while maximising vehicle detection probability. We theoretically analyse the performance of the sensor network for various coordination strategies and show that by appropriately coordinating their schedules the sensors can achieve significantly improved system-wide performance, detecting up to 50% of the events that a randomly coordinated network fails to detect. Finally, we deploy our coordination approach in a realistic simulation of our wide area surveillance problem, comparing its performance to a number of benchmarking coordination strategies. In this setting, our approach achieves up to a 57% reduction in the number of missed vehicles (compared to an uncoordinated network). This performance is close to that achieved by a benchmark centralised algorithm (simulated annealing) and to a continuously powered network (which is an unreachable upper bound for any coordination approach). phenomena in remote locations [19] . A fundamental challenge within all such applications arises due to the fact that the sensors within these networks are often deployed in an ad hoc manner (e.g. dropped from an aircraft or ground vehicle within a military surveillance application), and thus, the local environment of each sensor, and hence the exact configuration of the network, can not be determined prior to deployment. Rather, the sensors themselves must be equipped with the capability to autonomously adapt, sometime after deployment, once the local environment in which they (and their neighbours) find themselves has been determined. Examples of such adaptation include determining the most energyefficient communication paths within the network once the actual reliability of communication links between individual sensors has been measured in situ [39], dynamically determining the optimal orientation of range and bearing sensors to track multiple moving targets as they move through the sensor network [12] , and in the application that we consider in detail in this paper, coordinating the sense/sleep schedules (or duty cycles) of power constrained sensors deployed in a wide-area surveillance task, once the degree of overlap of the sensing fields of nearby sensors has been determined.
doi:10.1007/s10458-013-9225-1 fatcat:tqqmbvkidfe2fdahfnd753mqjq