A sequential GM-based PHD filter for a linear Gaussian system

Abstract

The probability hypothesis density (PHD) filter provides an efficiently parallel processing method for multi-target tracking. However, measurements have to be gathered for a scan period before the PHD filter can perform a recursion, therefore, significant delay may arise if the scan period is long. To reduce the delay in the PHD filter, we propose a sequential PHD filter which updates the posterior intensity whenever a new measurement becomes available. An implementation of the sequential PHD filter for a linear Gaussian system is also developed. The unique characteristic of the proposed filter is that it can retain the useful information of missed targets in the posterior intensity and sequentially handle the received measurements in time.