In this work, the Transformer Network (TRN) is applied to automatic modulation classification (AMC) problem for first time. Different from other deep networks, TRN can incorporate global information of each sample sequence and exploit that semantically relevant classification. order illustrate performance proposed model, it compared with four models two traditional methods. Simulation results show one has a higher accuracy especially at low signal noise ratios (SNRs), number training...

Text classification is one of the most widely used natural language processing technologies. Common text applications include spam identification, news classification, information retrieval, emotion analysis, and intention judgment, etc. Traditional classifiers based on machine learning methods have defects such as data sparsity, dimension explosion poor generalization ability, while deep network greatly improve these defects, avoid cumbersome feature extraction process, strong ability...

An expanding and shift scheme for efficient fourth-order difference coarray construction is proposed. It consists of two sparse subarrays, where one them modified shifted according to the analysis provided. The number consecutive lags proposed structure at fourth order consistently larger than previously methods. Two effective examples are provided with second subarray chosen be a two-level nested array, as such choice can increase further. Simulations performed show improved performance by...

A direction of arrival (DOA) estimation algorithm is proposed using the concept sparse representation. In particular, a new signal representation model called smoothed covariance vector (SCV) established, which constructed lower left diagonals matrix. DOA then achieved from SCV by recovering, where two distinguished error limit methods constrained optimization are to make algorithms more robust. The shows robust performance on in uniform array, especially for coherent signals. Furthermore,...

The data volume and computation task of MIMO radar is huge; a very high-speed necessary for its real-time processing. In this paper, we mainly study the time division signal processing flow, propose an improved algorithm, raising algorithm speed combined with previous algorithms, and, on basis, parallel simulation system based CPU/GPU architecture proposed. outer layer framework coarse-grained OpenMP acceleration CPU, inner fine-grained accelerated GPU. Its performance significantly faster...

The existing research on deep learning for radar signal intra–pulse modulation classification is mainly based supervised leaning techniques, which performance relies a large number of labeled samples. To overcome this limitation, self–supervised framework, contrastive (CL), combined with the convolutional neural network (CNN) and focal loss function proposed, called CL––CNN. A two–stage training strategy adopted by CL–CNN. In first stage, model pretrained using abundant unlabeled...

Radar intrapulse signal modulation classification is an important work for the electronic countermeasure and there are mainly two categories of algorithms. The deep learning-based algorithms usually outperform traditional feature extraction-based ones, but they may rely on massive labeled samples training, which limits their practical applications. To solve this problem, SS-LWCNN model combines semisupervised learning (SI-SL) with virtual adversarial training (VAT) light weight technology...

A novel orthogonal frequency-division multiplexing (OFDM) multiple-input multiple-output (MIMO) radar is proposed in this paper a scenario of coexisting with communication system. For the purpose avoiding interference to while maintaining MIMO radar's capabilities, such as measuring signal's direction departure (DOD), collocated antenna array divided into several overlapped subarrays. Mutually OFDM waveforms exploiting an space-time block code are transmitted through these subarrays obtain...

In order to further improve the performance of radar signal modulation recognition, recognition algorithm based on an improved convolutional neural networks (CNN) model is proposed in this paper. As CNN has some shortcomings such as long training time and poor generalization, dense connection block layer global pooling are added its performance. experiment, eight types signals used verify feasibility algorithm, results show that advantages high rate, short good generalization.

Recently, link flooding attacks (LFA) have been observed as a serious threat for cutting off the Internet connectivity through congesting critical links. A LFA typically utilizes legitimate and low-rate flows, which makes it extremely hard to be detected and, subsequently, mitigated. In this paper, we present LF-Shield, that is deep convolutional neural network (ConvNet) based countermeasure accurately detect efficiently mitigate LFAs using software-defined (SDN) paradigm. LF-Shield can...

The optical approach to estimate the direction-of-arrival (DOA) estimation of microwave signals has attracted a lot attention recently. Most existing methods are based on one-sensor array system, which converts DOA problem into an power problem. Their main disadvantage is that additional work needed represent relationship between phase shifts and powers before estimation. algorithm proposed in this paper <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> <mml:mrow...

A time synchronization algorithm for hidden mobile node (HMB), which can only receive signals, joining an existing synchronized underwater Acoustic (UWA) sensor network (UASN) is proposed. In order to obtain the location of HMB or communicate with it, local should synchronize UASN. However, propagation delay in UWA channels could not be ignored compared electromagnetic radio channels. proposed algorithm, do uniform linear motion a certain direction, and clock drift solved. After derivation...

The under-determined direction of arrival (DOA) estimation problem for a mixture circular and non-circular signals is studied in the context sparse arrays novel compressive sensing based DOA algorithm proposed. Compared to direct application existing algorithm, new one can make more effective use degree freedoms provided by both difference co-array sum co-array, which are generated vectorising covariance matrix pseudo array, respectively. Simulation results presented show improved...

In this work, the omni-dimensional dynamic convolution (ODConv) layer based network (OD-CNN) with focal loss function is applied to radar intra-pulse signal modulation classification, which greatly improves classification accuracy. Compared layer, ODConv employs a novel multi-dimensional attention mechanism learn four types of attentions along dimensions kernel space in parallel manner, further feature mining ability model. order illustrate superior proposed model, it compared other three...

With the development of science and requirement, Integration Communication Radar with properties more functions, lower buck safety is becoming a trend draws attention researchers. Current researches are focusing on communication or radar respectively just signal generation. This paper proposes solution integrated system for SAR based OFDM techniques in high speed scenario. The receiver can reconstruct reference which adopted pulse compression procedure by directly. In scenario, channel...

It is well known that the presence of non-circular signals can provide extra degrees freedom for a sensor array system, which then be exploited improved performance. Recently, by exploiting non-circularity information and difference sum co-array concept, new approach was developed direction arrival estimation in mixture circular noncircular signals. In this paper, problem reformulated algorithm finding under same conditions. As shown extensive simulation results, so-called MUSIC (IMUSIC)...

In order to improve the speed of DOA estimation, an efficient MUSIC algorithm using subspace projection is proposed in this paper. algorithm, covariance matrix, which causes high computational complexity (SP) tracking field, approximated simplify processing procedure. The shows similar angle accuracy compared with SP-MUSIC algorithm. However, it much more and needs less storages than since doesn't require matrix estimation eigenvalue decomposition. Finally, simulation results are used...

In this paper, a gridless DOA estimation method with coexistence of non-circular and circular signals is proposed by employing an enhanced sparse nested array, whose virtual array has no holes. The derived from both sum difference co-arrays are constructed based on atomic norm minimization. Simulation results provided to demonstrate the performance method.

The Cramer-Rao bound (CRB) offers insights into the inherent performance benchmark of any unbiased estimator developed for a specific parametric model, which is an important tool to evaluate direction-of-arrival (DOA) estimation algorithms. In this paper, closed-form stochastic CRB mixture circular and noncircular uncorrelated Gaussian signals derived. As general one, it can be transformed some existing representative results. existence condition also analysed based on sparse arrays, allows...