In this paper, a new time-frequency analysis method known as the polynomial chirplet transform (PCT) is developed by extending conventional (CT). By using function instead of linear chirp kernel in CT, PCT can produce distribution with excellent concentration for wide range signals continuous instantaneous frequency (IF). addition, an effective IF estimation algorithm proposed based on PCT, and effectiveness validated applying it to estimate signal nonlinear component seriously contaminated...

Interest in parameterized time-frequency analysis for non-stationary signal processing is increasing steadily. An important advantage of such to provide highly concentrated representation with signal-dependent resolution. In this paper, a general scheme, named as transform (GPTF transform), proposed carrying out analysis. The GPTF defined by applying generalized kernel based rotation operator and shift operator. It provides the availability single applications on signals different natures....

The conventional time-frequency analysis (TFA) methods, including continuous wavelet transform, short-time Fourier and Wigner-Ville distribution, have played important roles in analyzing nonstationary signals. However, they often show less capability dealing with signals time-varying frequency due to the bad energy concentration plane. On other hand, by introducing an extra transform kernel that matches instantaneous of signal, parameterized TFA methods powerful ability characterizing...

Abstract The segmentation accuracy of bridge crack images is influenced by high‐frequency light, complex scenes, and tiny cracks. Therefore, an integration–competition network (complex [CCSNet]) proposed to address these problems. First, a grayscale‐oriented adjustment algorithm solve the light problem. Second, mechanism detach backgrounds grayscale features Finally, attention extract shallow CCSNet outperforms seven state‐of‐the‐art methods in both generalization comparison experiments on...

An analytical and closed-form optimal solution expressed in elementary functions for superdirectivity of a circular sensor array is proposed this paper. By utilizing the circulant property data covariance matrix arrays, such solutions are derived to accurately calculate beamforming vector, beam pattern, corresponding directivity factor, eigenbeams. Based on these solutions, it proved that possesses several important properties which facilitate an eigenbeam decomposition synthesis approach...

This paper presents a detailed study of the high-order superdirectivity circular sensor arrays, which is aimed at completing authors' recently proposed analytical model. From limit expression maximum directivity factor, it shown that arrays possess good potential for improvement. It found sensitivity function used as robustness measurement can also be accurately decomposed into series closed-form functions eigenbeams, similar to optimal beampattern and its corresponding factor. Moreover,...

In deep-sea source localization, some of the existing methods only estimate range, while others produce large errors in distance estimation when estimating both range and depth. Here, a convolutional neural network-based method with high accuracy is introduced, which localization problem solved as regression problem. The proposed network trained by normalized acoustic matrix used to predict position. Experimental data from western Pacific indicate that this performs satisfactorily: mean...

In the field of underwater acoustic recognition, machine learning methods rely on a large number datasets to achieve high accuracy, while actual collected signal samples are often very scarce, which has great impact recognition performance. This paper presents method an target by data augmentation technique and residual convolutional neural network (CNN) model, is used expand training improve As representative model in CNN, ResNet18 for recognition. The whole process mainly includes...

Chronic neuropathic pain and comorbid depression syndrome (CDS) is a major worldwide health problem that affects the quality of life patients imposes tremendous socioeconomic burden. More than half with chronic also suffer from moderate or severe depression. Due to complex pathogenesis CDS, there are no effective therapeutic drugs available. The lack research on neural circuit mechanisms CDS limits development treatments. purpose this article provide an overview various circuits involved in...

Horizontal refraction notably influences propagation characteristics with the variation of waveguide environment. In this study, horizontal phenomenon at low frequencies was investigated in a sloping sea region an incomplete vertical sound speed profile. Using mode coupling theory, research explores relationship between and energy exchange among modes, examining impact environmental conditions on angle. Theoretical derivations numerical simulations reveal mechanisms by which source depth...

The regulatory mechanisms underlying embryogenic callus (EC) formation in polyploid bananas remain unexplored, posing challenges for genetic transformation and biotechnological applications. Here, we conducted transcriptome sequencing on cultured explants, non-embryogenic callus, EC, browning the ABB cultivar ‘MJ’ (Musa spp. cv. Bengal). Our analysis of differentially expressed genes (DEGs) revealed significant enrichment plant hormones, MAPK, zeatin biosynthesis pathways. Notably, most MJ...

Within the conventional sparse Bayesian learning (SBL) framework, only Gaussian scale mixtures have been adopted to model sparsity-inducing priors that guarantee exact inverse recovery. In light of relative scarcity formal SBL tools in enforcing a proper sparsity profile signal vectors, we explore use hierarchical synthesis lasso (HSL) for representing same small subset features among multiple responses. We outline viable approximation this particular choice prior, leading tractable...

The evaporation duct which forms above the ocean surface has a significant influence on electromagnetic wave propagation 2 GHz over ocean. effects of horizontal inhomogeneity are investigated, both in numerical simulation and experimental observation methods, this paper. Firstly, features discussed. Then, two typical inhomogeneous cases simulated compared with homogeneous case. result shows that path loss is significantly higher than case when height (EDH) at receiver lower transmitter. It...

Localized multiple kernel learning (LMKL) is an attractive strategy for combining heterogeneous features with regard to their discriminative power each individual sample. However, the of numerous local solutions may not scale well even a moderately sized training set, and independently learned models suffer from overfitting. Hence, in existing methods, distributed samples are typically assumed share same weights, various unsupervised clustering methods applied as preprocessing. In this...