Search

To address the problem of target detection in distributed MIMO radar, three styles of distributed MIMO radar systems are given in this paper. With respect to the geometric distribution relations of distributed MIMO radar, the styles are distributed coherent MIMO radar, random phase MIMO radar, and random amplitude-phase MIMO radar. Next, the square law detector structures of random phase and random amplitude-phase MIMO radar are derived in the paper when there is a low signal-to-noise ratio, and the performance of the two detectors are analyzed as well. Finally, simulation results demonstrate the theoretical analysis of this paper are of guiding significance for the actual engineering.

Only a subset of transmitters and receivers in a distributed Multi-Input Multi-Output (MIMO) radar network is allowed to actively track a target at a particular instance due to the limited time and energy resource of a MIMO radar network. It is therefore desirable to obtain an efficient method to overcome the resource constraints while optimizing the tracking performance. In this study, posterior Cramer-Rao lower bound is used as the performance metric and the selection problem is formulated as a Boolean programming problem aiming at optimizing the worst tracking performance of multiple targets. It is later relaxed to a semidefinite programming and solved by the block coordinate descend method. Numerical results show that proposed method superior to the fixed selection method. In addition, with less computation complexity, the proposed method obtains nearly equivalent performance compared with exhaustive search method.

Heterogeneous change detection has a wide range of applications in many fields. However, to date, many existing problems of heterogeneous change detection, such as false alarm suppression, have not been specifically addressed. In this article, we discuss the problem of false alarm suppression and propose a new method based on the combination of a convolutional neural network (CNN) and graph convolutional network (GCN). This approach employs a two-channel CNN to learn the feature maps of multitemporal images and then calculates difference maps of different scales, which means that both low-level and high-level features contribute equally to the change detection. The GCN, with a newly built convolution kernel (called the partially absorbing random walk convolution kernel), classifies these difference maps to obtain the inter-feature relationships between true targets and false ones, which can be represented by an adjacent matrix. We use pseudo-label samples to train the whole network, which means our method is unsupervised. Our method is verified on two typical data sets. The experimental results indicate the superiority of our method compared to some state-of-the-art approaches, which proves the efficacy of our method in false alarm suppression. [ABSTRACT FROM AUTHOR]