In recent years, with the rapid increase in the number of drones, the abuse of drones for activities such as terrorist attacks are bound to pose a threat to public safety and privacy. Therefore, it is very important to identify and classify them. MMW radar provides an effective means to detect drones. And with the development of B5G/6G technologies, a large number of MMW base stations will be deployed in cities in the future. These MMW base stations can be used as MMW radar through simple modification, which greatly saves the detection cost. Radar cross section, as a common data of radar, can provide a data source for drone recognition and classification. Therefore, using the publicly available RCS data set of drones in the MMW band. We firstly encode the RCS series into a 2D GAF representation and design a 2D ResNet-10 to classify them. Secondly, we propose a deep fusion network, which can be used as a RTR with RCS as the information source. The experimental results show that the classification accuracy of 2D ResNet-10 and 2D ResNet-18 is almost the same in GAF representation; Compared with other RCS based classification methods, the performance of deep fusion network is the best.
The radar cross section (RCS) is an important parameter that reflects the scattering characteristics of radar targets. Based on the monostatic radar RCS time series' statistical features by sliding window segmentation, a novel sliding window‐statistical‐gated recurrent unit (SW‐S‐GRU) method for radar target recognition (RTR) is proposed using a GRU. The sliding window method divides the RCS time series into many segments for obtaining the statistical features as the input of the GRU. The deeper features are extracted from the segmented time series and then classified by the GRU. Under the model of precession motion, simulation results show that the proposed method achieves a great recognition accuracy. Even under the condition of a low SNR, the proposed method can obtain better recognition accuracy. For the precession dataset, the effect of sliding window parameters on RTR is analysed. The simulation results show that the recognition result of the model is better using the longer series overlap length between adjacent times. And when the number of hidden layers is 8 or above 8, the recognition accuracy does not change. Using the publicly available measured drone RCS data, the proposed method can also identify the drone well.
The problem of the constant false alarm rate (CFAR) detection of bistatic space based radar (B-SBR) is addressed under the background of strong and nonhomogeneous clutter. We propose a novel and simple cell averaging (CA)-CFAR-like (CA-CFAR-L) detection method based on the Gerschgorin circle theorem for B-SBR. First, we investigate the potential of a novel test statistic TCA-CFAR-L exhibiting better properties about discrimination and CFAR using the Gerschgorin circle theorem, and the test statistic TCA-CFAR-L is constructed as the ratio of the sum of the Gerschgorin radii of the covariance matrix of the cell under test to the sum of the Gerschgorin radii of the mean covariance matrix of the reference cells. Second, we investigate the CA-CFAR-L detector for B-SBR using the constructed test statistic TCA-CFAR-L. Under the background of the K-distribution clutter model and the clutter model for B-SBR, respectively, we verify the performance of the proposed CA-CFAR-L detector, the LE mean matrix CFAR detector and the Fast Fourier Transform (FFT)–CFAR detector. Also, we analyze the experimental data for the IPIX radar. The simulation results and experimental data analysis demonstrate that the detection performance of the detector is better than the LE mean matrix CFAR detector and the FFT–CFAR detector, and the proposed method has less computational complexity. These will provide an important theoretical basis and new ideas for radar target detection in the future.
With the development of multimedia editing technologies, the copyright protection has attacked more attentions. Reversible data hiding (RDH), in which the cover can be recovered losslessly, is an effect method to eliminate embedding distortions. As a typical RDH method, histogram shifting (HS) is used widely. Most existing RDH schemes based on HS usually build sharp histograms by predicting and sorting techniques. To make use of spatial correlations of multimedia, several RDH schemes based on multiple HS (MHS) are proposed to protect copyright, in which some rigid rules are used to build multiple histograms. Against images, videos have more spatial and temple correlations and it is easier to acquire sharper histograms. In this paper, a video MHS scheme based on compression sensing (CS) is proposed. As a linear sensing algorithm, CS can measure macroblock residuals by reducing corrections among pixels to acquire distinguishable macroblock features, while keeping their statistical characteristics immutable. By employing CS, macroblocks with similar characteristics cluster together to formulate multiple histograms. For each of these histograms, data embedding is implemented to reduce shifting distortions by expanding the outermost bins while other bins are unchanged. Experimental results show that the quality of most test videos in our scheme are higher than that in the state-of-art schemes.