Mission Critical Sensor Networks Research Articles

Intelligent Resource Collaboration in Mobile Target Tracking Oriented Mission-Critical Sensor Networks

Mobile target tracking-oriented sensor networks are a special kind of Mission-critical Sensor Networks (MCSN), in which the various missions with the diverse priorities exist. However, it is challenging to achieve real time tracking while keeping the MCSN a long life time with limited energy provision in a complicated environment. In this paper, we develop a collaborative perception and intelligent scheduling scheme, which jointly optimizes the system responding latency and tracking accuracy with the constraint of the available energy. A new hierarchical architecture is proposed to realize the coupled function of perception and computation. In particular, the multi-node collaborative perception scheme is applied to obtain the excellent sensing capacity, and the Unmanned Aerial Vehicles (UAVs) play as the edge nodes to provide the computing service for those resource-constrained sensor nodes. To reach the sustained target tracking, we propose an intelligent tracking policy by exploiting the deep deterministic policy gradient (DDPG) method. Simulation results demonstrate that the proposed intelligent collaboration scheme can improve the tracking accuracy by 45.5% compared with the random selection scheme. The system cost is also reduced approximately by 17.3% while guaranteeing the tracking accuracy.

Adaptive Fuzzy Tree System for Target Tracking in Mission Critical Sensor Networks

The emergence of mobile sensors dramatically improves the availability of mission critical sensors and sensor networks (MC-SSN), enabling it to be utilized in more challenging tasks. However, current researches aimed to target tracking in mobile MC-SSN are very limited. This paper first proposes an adaptive fuzzy tree system (AFS) for target tracking in MC-SSN. To avoid oversimplifying the target mobile pattern, which is unrealistic in real tracking tasks, we introduce the target circle into the model considerations to add a bit of uncertainty about the target position. At each time step of the tracking process, strategy-selection layer will activate the pursuit strategy or the diffusion strategy for the specific situation, which makes the system more intelligent and can be applicable to various scenarios. The pursuit strategy is built by a two-layer fuzzy tree to select and mobilize sensors that have not detected target, while the diffusion strategy uses a fuzzy inference system to balance the density of detected sensors. And all the hyper-parameters are tuned by particle swarm optimization (PSO). We performed a large number of simulations with two target trajectories: line and irregular. The simulation results show that the AFS significantly outperforms the state-of-the-art algorithm. Moreover, it is highly robust to various target motion patterns, making it competent for a variety of real target tracking scenarios.