Exploiting Information About the Structure of Signals of Opportunity for Passive Radar Performance Increase

In this paper, we study a passive radar system employing digital communication signals of opportunity. We develop the frequency domain model for the received observations of a passive radar and investigate the problem of joint target position and velocity estimation. We first consider the case where we know exactly the transmitted signals. This performance is compared with that for the case of known transmitted signal form, but where the signals have unknown parameters which include the unknown information bits. Next, we consider the case where we treat the samples of the received signal as deterministic unknowns. Finally, we consider the case where the samples of the received signals in a discrete-time formulation are modeled as being random with known distribution. In each case we derive the corresponding Cramer-Rao bound (CRB). Our focus is on the impact of employing extra information on the estimation performance. We show that for orthogonal frequency division multiplexing (OFDM) signals, knowing the signal form provides numerical performance results close to those for the completely known signal case (active radar) when the number of samples per bit is large enough. It is also shown that incorporating the direct path signals in the received observations can have a positive influence on estimating target parameters. The estimation performance improves with an increasing direct path signal strength for all cases.