Writing on water, a lightweight soft-state tracking framework for dense mobile ad hoc networks

Abstract

Abstract-In a mobile ad hoc network, tracking protocols need to deal with-in addition to the mobility of the target-the mobility of the intermediate nodes that maintain a track toward the target. To address this problem, we propose the MDQT (Mobility-enhanced Distributed QuadTree) tracking framework. MDQT employs a static cell abstraction to mask the mobility of the nodes and provide the illusion of a logical static network overlaid on the mobile network. MDQT implements this virtual static network layer in a lightweight/communication-free manner by exploiting the soft-state principle and the snooping feature of wireless communication.Using simulations, we study the effects of the mobility speed and the percentage of mobile nodes on the performance of our MDQT framework. We find that even at very high mobility speeds (50 meters per second), low update rates (1 update per second), and 100% node mobility, the success rate of MDQT tracking is above 85% and the latency is comparable with that of static networks.

Model-based Tracking for Mobile Ad Hoc Networks

“…This way tracking messages can escape from local maxima and eventually be forwarded to the target. In [9] we show that a cell abstraction can be achieved easily given the assumption that the coordinates of any nodes are available(a node does not need any message exchange with other nodes to decide which cell it belongs to). Gradients are based on nodes rather than cells, and they are not exchanged between nodes.…”

“…As we mentioned before, HMM models haves limitations in nature when applied in WSNs: HMM predicts next move only based on the last location, hence leads to large errors against real trails; Moreover, HMM is more appropriate for offline processing than online tracking. Dynamic hierarchical/cluster based structures are another category of approaches for mobile target tracking applications [8,9,10,20]). GLS is a scalable and distributed location service structure which divides the global map into hierarchical grids with increased size at high levels.…”

“…DCTC (Dynamic Convoy Tree-based Collaboration) [10] uses a tree structure called 'Convoy Tree', which is rooted at the target's location and is dynamically adjusted as the target moves. MDQT [9] overlays a virtually static distributed Quad-Tree structure in the field. As a node detects the target, it pushes the event into the tree until the event reaches the root.…”

A Holistic Solution to Pursuer-Evader Tracking in Sensor Networks

“…The first type of protocols dynamically maintain certain structures, such as trees or overlay graphs, such that a message may simply follow the structure to reach the target [2], [6], [12], [13], [22]. In structure-free protocols, nodes take local decisions by making predictions over available knowledge to generate a path on demand for tracking messages to be forwarded to the target [7], [18], [19], [21].…”

“…There has also been several works on PET problem using WSNs [3]- [5], [13], [16]. These works take a one-sided view of the problem, and follow an evader-centric approach by ignoring the dynamics of the pursuer.…”