In this study, the problem of detection and parameter estimation of manoeuvring targets in passive radar is addressed. Classical processing scheme used in passive radars works well under certain conditions. If the conditions are not met, such as in the case of detection of highly manoeuvring targets, performance degradation can be expected. In this study, an extended processing is proposed, which makes it possible to detect manoeuvring targets without the loss of the signal-to-noise ratio. Moreover, an estimate of the bistatic acceleration is available, which can greatly facilitate tracking of manoeuvring targets. The proposed methods are tested using real-life data recorded with an FM-based passive radar.