Mobile Sensor Networks Self-Organization for System Utility Maximization: Work in Progress

This paper reports on work in progress on mobile sensor network self-organization for system utility maximization with system utility being the difference between the aggregate utility of sensor information and penalty associated with sensor battery energy expenditure. We assume that location-dependent marginal utility of each sensor is determined and communicated back to the corresponding sensor by the sensor information access point, where the information streams from all sensors are fused into coherent picture. Self-organization takes place in two time scales: in the fast time scale sensors form and optimize multi-hop Mobile Ad-hoc NETwork (MANET) with sensors cooperating by relaying other sensor information possibly in addition to transmitting their own information. Optimized in the fast time scale system utility determines potential fields and virtual forces guiding sensor repositioning in the slow time scale.