For the multi-target DOA estimation problem of uniform linear arrays, this paper proposes a DOA estimation method based on the deep convolution neural network. The algorithm adopts the deep convolutional neural network, and the DOA estimation problem of the array signal is transformed into the inverse mapping problem of the array output covariance matrix to a binary sequence in which “1” indicates that there is a target incident in the corresponding angular direction at that position. The upper triangular array of the discrete covariance matrix is used as the data input to realize the DOA estimation of multiple sources. The simulation results show that the DOA estimation accuracy of the proposed algorithm is significantly better than that of the typical super-resolution estimation algorithm under the conditions of low SNR and small snapshot. Under the conditions of high SNR and large snapshot, the estimation accuracy of the proposed algorithm is basically the same as those of the MUSIC algorithm, ESPRIT algorithm, and ML algorithm, which are better than that of the deep fully connected neural network. The analysis of the simulation results shows that the algorithm is effective, and the time and space complexity can be further reduced by replacing the square array with the upper triangular array as the input.
27 June 2019 in Security and Communication Networks
The current common color image encryption algorithms applying “scrambling-diffusion” have some problems, such as the small key space, the cumbersome encryption process, and the security vulnerability. Aiming at these problems, this paper proposes a new color image encryption algorithm based on the hyperchaotic system and applying “transforming-scrambling-diffusion” model. Before scrambling, in accordance with the plaintext itself attributes, the number of iterations was calculated, all the pixel values of color image were transformed into gray code iteratively, and then the chaotic sequence was generated from the four-dimensional hyperchaotic system. Pixel matrix after gray code transformation was converted to one-dimensional matrix. The chaotic sequence was sorted and the one-dimensional matrix was changed positions correspondingly to complete the whole domain scrambling. And then, bit-operation was executed for image diffusion. The ciphertext can be obtained by matrix transformation. The key sensitivity, histogram, information entropy, correlation, and other evaluation indexes were calculated and analyzed through the simulation experiment. Compared with other algorithms, it can be proved that the encryption algorithm has the strong antiattack ability.
This paper proposes a joint estimation method for source number and DOA based on an improved convolutional neural network for unknown source number and undetermined DOA estimation. By analyzing the signal model, the paper designs a convolutional neural network model based on the existence of a mapping relationship between the covariance matrix and both the source number and DOA estimation. The model, which discards the pooling layer to avoid data loss and introduces the dropout method to improve generalization, takes the signal covariance matrix as input and the two branches of source number estimation and DOA estimation as outputs, and achieves the unfixed number of DOA estimation by filling in invalid values. Simulation experiments and analysis of the results show that the algorithm can effectively achieve the joint estimation of source number and DOA. Under the conditions of high SNR and a large snapshot number, both the proposed algorithm and the traditional algorithm have high estimation accuracy, while under the conditions of low SNR and a small snapshot, the algorithm is better than the traditional algorithm, and under the underdetermined conditions, where the traditional algorithm often fails, the algorithm can still achieve the joint estimation.
14 October 2021 in International Journal of Antennas and Propagation
Aiming at low degrees of freedom (DOF) and high mutual coupling (MC) of the existing sparse arrays, an enhanced generalized nested array (EGNA) is proposed in this paper. Specifically, the proposed array adds a single antenna on the basis of generalized nested array (GNA), and the difference of coprime factors is employed as the spacing between the second subarray and the additional antenna. Then, the values of the coprime factors are analyzed in detail, which indicates that Yang-NA can be explained as a special case. Compared with the majority of the existing sparse arrays, EGNA not only has the closed-form expressions of the physical antenna locations, consecutive lags, and unique lags, but also significantly increases DOF and reduces MC. In view of the above advantages, EGNA can obtain superior performance in direction of arrival (DOA) estimation. Numerical simulation results verify the rationality and superiority of the proposed nested array.
For the DOA (direction of arrival) estimation of a low-elevation-angle target under the influence of a multipath effect, this paper proposes a DOA estimation method based on CAE (convolutional autoencoder) and CNN (convolutional neural network). The algorithm firstly inputs the signal covariance matrix of the array of the low-elevation target containing direct and reflected waves into the convolutional autoencoder to realize the de-multipath, and uses the spatial features extracted by the convolutional autoencoder as the input of the extreme learning machine to realize the DOA preclassification of direct waves; based on the preclassification results, one branch of the three parallel convolutional neural nets is selected, and the output of the convolutional autoencoder is used as the input of this branch to realize DOA estimation. The simulation results show that the algorithm has better estimation accuracy and efficiency than the conventional algorithms, especially when the DOA of the target is in the lower range. The analysis of the simulation results shows that the algorithm is effective, in which the convolutional autoencoder can effectively realize the de-multipath, and the use of parallel convolutional neural networks can avoid overfitting and underfitting and realize DOA estimation more accurately.
This paper addresses the problem of the two-dimensional direction-of-arrival (2D DOA) estimation of low-elevation or non-low-elevation targets using L-shaped uniform and sparse arrays by analyzing the signal models’ features and their mapping to 2D DOA. This paper proposes a 2D DOA estimation algorithm based on the dilated convolutional network model, which consists of two components: a dilated convolutional autoencoder and a dilated convolutional neural network. If there are targets at low elevation, the dilated convolutional autoencoder suppresses the multipath signal and outputs a new signal covariance matrix as the input of the dilated convolutional neural network to directly perform 2D DOA estimation in the absence of a low-elevation target. The algorithm employs 3D convolution to fully retain and extract features. The simulation experiments and the analysis of their results revealed that for both L-shaped uniform and L-shaped sparse arrays, the dilated convolutional autoencoder could effectively suppress the multipath signals without affecting the direct wave and non-low-elevation targets, whereas the dilated convolutional neural network could effectively achieve 2D DOA estimation with a matching rate and an effective ratio of pitch and azimuth angles close to 100% without the need for additional parameter matching. Under the condition of a low signal-to-noise ratio, the estimation accuracy of the proposed algorithm was significantly higher than that of the traditional DOA estimation.