Passive radar is known to suffer from high clutter. In this paper a generalized subband cancellation (GSC) algorithm is proposed to achieve robust clutter rejection. Four major factors are taken into account, namely carrier frequency offset (CFO), sampling frequency offset (SFO), fractional delay, and hardware frequency response. The GSC algorithm utilizes subband signal processing and tailored clutter subspace construction to solve the problems caused by these factors. Quantitative evaluations demonstrate that the GSC algorithm approaches the optimal clutter cancellation performance. Its effectiveness is also validated using field experimental data.

Geometry configuration is crucial to the system performance in distributed multiple-input-multiple-output (MIMO) radars. Sensor placement in distributed MIMO radars is different from the conventional sensor placement problem as transmitters and receivers need to cooperate with each other to perform sensing. In this paper, we establish a combinatorial optimization model for the joint placement of transmitters and receivers. The proposed algorithm first transforms the original model into an equivalent model with convex constraints via convex relaxation. Then, the nonconvex objective function is further replaced with a new convex surrogate. The surrogate is shown to converge to a good approximation of the original optimal solution. The performance of the proposed algorithm is validated using numerical simulations.

In frequency-modulation (FM)-based passive radar, the strong side peaks randomly appearing in the ambiguity function will generate a false alarm of target detection. To mitigate this side peak interference, this paper starts with a detailed analysis of the structure and ambiguity function of FM stereo signal. Then, it expounds the formation and characteristics of side peaks, together with a side peak identification method, which identifies and discards the false detections of the same range and Doppler caused by side peaks, respectively. The performance analysis, conducted using both simulated data and real recorded datasets, proves that the proposed method can eliminate the false targets caused by side peaks thus improving the detection performance in FM-passive radar.

In recent years, digital broadcasting signals, like Digital Audio/Video Broadcast (DAB/DVB), have received increasing attention in passive radar system. Most of these digital broadcasting signals use orthogonal frequency division multiplex (OFDM) modulation with cyclic prefix (CP-OFDM). Benefitting from the distinctive characteristics of the CP-OFDM signal, several efficient methods have been proposed to handle the high clutter problem in the CP-OFDM based passive radar. These methods are all operated in the subcarrier domain. In this paper, an evaluation of the inner relationship and suppression performance of these subcarrier-domain methods are presented, along with the discussion of their Doppler response and computational complexity. The mathematical derivations demonstrate that two of the methods have almost the equivalent suppression mechanism and their suppression performance depends on the modulation scheme of the illuminators of opportunity (IOs). Besides, these methods can achieve comparable suppression performance to that of the temporal method but have lower computational and memory complexity. Numerical experiments on simulated and real data as well validate the evaluation results.

The conventional calibration methods only took into account the mutual coupling (MC) and the gain/phase error. However, the antenna element pattern could also be an important factor in calibration as the antenna element pattern could be direction-dependent. This paper proposes a calibration method to jointly estimate array errors and antenna element patterns for the uniform circular array (UCA). The array errors are composed of mutual coupling and gain/phase error. The antenna patterns of different elements are presumed rotationally similar due to the structure of the UCA. Simulation and field experiments demonstrated the effectiveness of the proposed method.

In a passive radar, when the reference antenna is set to receive a direct-path signal, target signals will be inevitably mixed in the reference channel due to the antenna pattern. In this case, adaptive clutter cancellation in the temporal domain will affect the target steering vector due to the correlation between the target signal in the reference and surveillance channels, which results in a direction of arrival (DOA) estimation bias. The spatial-domain adaptive clutter rejection will also cause a similar influence. To solve this problem, in this article, we derive and analyze the effect of the adaptive clutter rejection on the target steering vector. Two kinds of DOA estimation methods are hence proposed to correct the estimation bias. The applicable boundaries of these two methods are also discussed. Simulations and real radar data show that the proposed methods can correct the DOA bias caused by adaptive clutter rejection, thereby improving the location accuracy in a passive radar.