Distributed Particle Filter for Target Tracking: With Reduced Sensor Communications

Tadesse Ghirmai

doi:10.3390/s16091454

Abstract

For efficient and accurate estimation of the location of objects, a network of sensors can be used to detect and track targets in a distributed manner. In nonlinear and/or non-Gaussian dynamic models, distributed particle filtering methods are commonly applied to develop target tracking algorithms. An important consideration in developing a distributed particle filtering algorithm in wireless sensor networks is reducing the size of data exchanged among the sensors because of power and bandwidth constraints. In this paper, we propose a distributed particle filtering algorithm with the objective of reducing the overhead data that is communicated among the sensors. In our algorithm, the sensors exchange information to collaboratively compute the global likelihood function that encompasses the contribution of the measurements towards building the global posterior density of the unknown location parameters. Each sensor, using its own measurement, computes its local likelihood function and approximates it using a Gaussian function. The sensors then propagate only the mean and the covariance of their approximated likelihood functions to other sensors, reducing the communication overhead. The global likelihood function is computed collaboratively from the parameters of the local likelihood functions using an average consensus filter or a forward-backward propagation information exchange strategy.

Highlights

A network of large number of sensors can be deployed to collect and process data to monitor a certain event
We have developed an algorithm that is flexible to employ both the forward-backward propagation and the average consensus filter information exchange strategies to compute the global likelihood function
We have presented distributed target tracking algorithms for sensor networks based on a particle filtering method in which a network of sensors collaborates to estimate the kinematic parameters of the target

Summary

Introduction

A network of large number of sensors can be deployed to collect and process data to monitor a certain event. Advances in sensor technology have allowed the development of “smart” sensors that are able to collect and process data [1]. Modern sensors are inexpensive and can be deployed in large numbers to form a sensor network so that the sensors collaborate in collecting, analyzing and processing information [1]. Deployment of large sensors in a network has the benefit of increasing the resolutions of the data collected and, as a result, increasing the accuracy of the processed information. Sensor networks, by virtue of the multiplicity of the data collected from a large number of sensors, provide protections against sensor failures, mitigates the effect of shadowing due to no line-of-sight (LOS)

Objectives

Results

Conclusion