Haichao Jiang

Multiframe Radar Detection of Fluctuating Targets Using Phase Information

This paper considers the detection of fluctuating targets via dynamic-programming-based track-before-detect (DP-TBD) in radar systems. Swerling targets of types 0, 1, and 3 are considered. DP-TBD usually integrates either squared amplitude or the logarithm of the envelope likelihood ratio (LELR) scoring functions. Thus, only amplitude information is taken into account regardless of the fact that the measurements are often complex valued. In this paper, the phase information is used in the integration process of DP-TBD to enhance radar detection performance. More precisely, the logarithm of the complex-measurement-based likelihood ratio (LCLR) is used, taking the place of squared amplitude or the LELR. First, we derive the expressions for the LELR and the LCLR for the three Swerling types. Then, to reduce the complexity of computing LELR and LCLR, we also propose efficient but accurate approximations for the LELR and the LCLR. Simulations are used to assess the performance of different DP-TBD strategies.

Track-before-detect strategies for range distributed target detection in compound-Gaussian clutter

This paper considers range distributed target detection in compound-Gaussian clutter, which is often arisen in high resolution radars. Two dynamic programming based track-before-detect (DP-TBD) strategies, i.e., noncoherent integration based DP-TBD (NCI-DP-TBD) and the generalized likelihood ratio test based DP-TBD (GLRT-DP-TBD) strategies, are proposed to address the detection problem. The detection performances of the proposed strategies are examined via numerical simulations. HighlightsRange distributed targets in compound-Gaussian clutter is considered.Noncoherent integration and GLRT based DP-TBD strategies are proposed.A covariance matrix estimation strategy is proposed.

Track-before-detect for fluctuating targets in heterogeneous clutter

This paper considers the radar detection and tracking of weak fluctuating targets in heterogeneous clutter via dynamic programming based track-before-detect (DP-TBD). The target fluctuating satisfies the well-known Swerling I model and radar clutter is modeled by G0 distribution, which is widely used to model heterogeneous clutter received by small grazing angle or high resolution radar. In this case, the log-likelihood ratio (LLR), which utilizes the clutter distribution information and the target fluctuating information, is required during the integration process of DP-TBD. Since no closed-form solution of LLR exists under this condition, we present a fast but accurate LLR approximation using variable resolution grid based method. Various simulations are used to examine the performance of the DP-TBD using the approximate LLR.

Track-Before-Detect Strategies for Radar Detection in G0-Distributed Clutter

This paper considers target detection via dynamic-programming based track-before-detect (DP-TBD) for radar systems. The clutter is modeled usingenlr G0 distribution, which is usually used to model clutter received from high-resolution radars and radars working at small grazing angles. Two target models, namely, Swerling 0 and 1 models, are considered to capture the radar cross section changes over time. DP-TBD techniques that integrate amplitude suffer from significant performance loss in this case due to the high likelihood of target-like outliers. In this paper, the log-likelihood ratio (LLR) is used in the integration process of DP-TBD, taking the place of amplitude, to enhance radar detection performance. The expressions for the LLR for the above target models are derived first. However, neither of them has a closed-form solution. In order to reduce the complexity of evaluating the LLR, efficient but accurate approximation methods are proposed. Then the approximated LLR is used in the integration process of DP-TBD. Simulations are used to examine the efficiency of the approximation methods as well as the performances of different DP-TBD strategies.

Knowledge-Based Track-Before-Detect Strategies for Fluctuating Targets in

In this paper, we address the problem of detecting fluctuating targets in heavy-tailed clutter through the use of dynamic programming based track-before-detect (DP-TBD) in radar systems. The clutter is modeled in terms of K distribution, while the well-known Swerling targets of types 1 and 3 are considered to capture the target amplitude fluctuation scan-toscan. Conventional DP-TBD suffers significant performance loss in this case due to the high frequency of target-like outliers. In this paper, we resort to the knowledge-based techniques, i.e., use a priori information to enhance radar detection performance. More precisely, knowledge-based DP-TBD (KB-DP-TBD) strategies are developed by incorporating the amplitude information into the integration process of DP-TBD. Simulations are used to assess performances and computational complexities of different DP-TBD strategies. The relevant result is that KB-DP-TBD can improve the detection performance over the conventional DP-TBD, especially for very heavy-tailed K-distributed clutter.

K

This paper considers the target tracking problem through the use of dynamic programming based track-before-detect (DP-TBD) in G0 clutter, where G0 clutter is used to model radar clutter returns received by small grazing angle or high resolution radars. The amplitude probability density functions of the clutter-only and target-plus-clutter cases are derived and the latter one is approximated using the sum of weighted incomplete gamma functions since no close-form solution exists. Then loglikelihood (LL) and log-likelihood ratio (LLR) are derived and used in the integration process of DP-TBD respectively. Various simulations are used to examine the performance of DP-TBD using different scoring functions.

-Distributed Clutter

The fluctuating target tracking problem through the use of dynamic programming based track-before-detect (DP-TBD) is considered in this paper. The target is modelled by the well-known Swerling family of target amplitude fluctuation models in order to capture the effect of radar cross-section changes that a target would present to radar over time. For each of the target fluctuation models, the complex likelihood ratio (CLLR), which utilizes the target amplitude fluctuation information together with the phase information, is derived and used in the integration process of DP-TBD. Simulation results show a significant gain in detection and tracking performance through accurately modeling the target amplitude fluctuations.

Tracking targets in G0 clutter via dynamic programming based track-before-detect

This paper focuses on the signal detection problem using compressive measurements in the presence of correlated noise, and derives closed probability expressions of detection and false alarm, also provide the signal with which the best detection performance can be obtained. Moreover, the theoretical lower- and upper-bounds are provided. Finally, several numerical simulations are provided and discussed.

Track-before-detect for fluctuating targets using phase information

In this paper, radar detection of Swerling target of type 3 in GO-distributed clutter via dynamic programming based track-before-detect (DP-TBD) is considered. Conventional DP-TBD suffers from significant performance loss due to the high frequency of target-like outliers. To enhance the detection performance of DP-TBD, the log-likelihood ratio (LLR) of the measurements is derived and used in the integration process of DP-TBD taking the place of amplitude that used in the conventional DP-TBD strategy. Various simulations are used to examine the detection performances of different DP-TBD strategies. Simulation results show that significant performance improvement can be achieved, especially for vary heavy-tailed clutter.

Signal detection with compressive measurements in correlated noise

The knowledge of the operating environment (whether scan-to-scan varying or stationary) has a significant impact on the performance of the original adaptive track-before-detect (TBD) strategy for space-time adaptive processing (STAP) radars. If the knowledge assumption is not met, the performance degrades dramatically due to the mismatch between the true covariance matrix and the estimated one. To deal with this problem, an improved adaptive TBD strategy is proposed in this paper, which reduces the impact of data contaminated by signal or interferer components on the estimation of covariance matrix and thus improves the performance. Computational complexity and performance of the proposed adaptive TBD strategy are analyzed and compared with the original adaptive TBD strategy.