
Lazy Inference on Object Identities in Wireless Sensor Networks
Jaewon Shin, Nelson Lee, Sebastian Thrun and Leonidas GuibasTracking the identities of moving objects is an important aspect of most multiobject tracking applications. Uncertainty in sensor data, coupled with the intrinsic difficulty of the data association problem, suggests probabilistic formulations over the set of possible identities. While an explicit representation of a distribution over all associations may require exponential storage and computation, in practice the information provided by this distribution is accessed only in certain stylized ways, as when asking for the identity of a given track, or the track with a given identity. Exploiting this observation, we proposed in [?] a practical solution to this problem based on maintaining marginal probabilities and demonstrated its effectiveness in the context of tracking within a wireless sensor network. That method, unfortunately, requires extensive communication in the network whenever new identity observations are made, in order for normalization operations to keep the marginals consistent [?]. In this paper, we propose a very different solution based on accumulated loglikelihoods, which can postpone all normalization computations until actual identity queries are made. In this manner the continuous communication and computational expense of repeated normalizations is avoided and that effort is expended only when actual queries are made of the network. We compare the two methods in terms of their computational complexities, inference accuracies, and distributed implementations. Simulation and experimental results from a RFID system are also presented.
The full paper is available in PDF and gzipped Postscript
@UNPUBLISHED{Shin04a, AUTHOR = {Shin, J. and Lee, N. and Thrun, S. and Guibas, L.}, TITLE = {Lazy inference ob object identities in wireless sensor networks}, YEAR = {2004}, ORGANIZATION = {Stanford University}, ADDRESS = {Stanford, CA}, NOTE = {Submitted for publication} } 