Flocking based sensor deployment in mobile sensor networks

Based on the flocking behaviors, this paper proposes three distributed, scalable, and stable deployment algorithms in homogeneous mobile sensor networks. We first address the deployment problem of driving sensors to surround a target of interest (TOI) which is either stationary or mobile. To maximize coverage area without coverage hole inside the network, sensors are relocated to approximate an equilateral triangle tessellation centered at the TOI. Since each sensor calculates the control input based on local and 1-hop neighborhood information, the algorithm is scalable. It is proved that a steady network state can be reached when the velocities of all sensors coincide with that of the TOI. We further extend the algorithm to handle sensor deployment under two other situations: (1) sensors are deployed uniformly along a predicted path of the TOI; (2) sensors are deployed according to a distribution density function where algorithms using both global map and local information are provided. Simulation results demonstrate the proposed algorithm can achieve reliable deployment with satisfactory moving distance.