In this paper, we consider nonsubsampled graph filter banks (NSGFBs) to process data on a sparse graph. The analysis of NSGFBs have small bandwidth, pass/block the normalized constant signal, and stability ℓ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> . Given an bank with introduce algebraic optimization methods construct well-localized synthesis such that corresponding provide perfect signal reconstruction in noiseless setting. We...

In this article, a multifeature detector based on isolation forest (iForest) algorithm is developed to detect floating small targets in sea clutter. The conventional can only process three features or less. proposed aims break the limitation of feature dimensions' number existed feature-based detectors and improve detection performance. It transforms target into an anomaly problem high-dimensional space, breaking features. First, modified constructed from multiple extracted Meanwhile,...

Graph Signal Processing (GSP) leverages pair-wise relationship between nodes of a graph to formulate operators on signals defined over the nodes. Most existing signal in literature are linear, and can be described by linear transformation matrices. Recently, works emerging that consider time correlation signals, leading time-vertex processing. By exploiting joint correlations across topology time, better results obtained. In this brief, we propose median operator will leverage correlation,...

Graph signal processing (GSP) is a field that deals with data residing on irregular domains, i.e. graph signals. In this field, the filter bank one of most important developments, owing to its ability provide multiresolution analysis However, current research focuses static The does not exploit temporal correlations time-varying signals in real-world applications, such as wireless sensor networks. paper, theory and design joint time-vertex nonsubsampled are developed, using generalized...

The recovery of missing samples from available noisy measurements is a fundamental problem in signal processing. This process also sometimes known as graph inpainting, reconstruction, forecasting or inference. Many the existing algorithms do not scale well with size and/or they cannot be implemented efficiently distributed manner. In this paper, we develop efficient for time-varying signals. priori assumptions are that smooth respect to topology and correlative across time. These can...

Graph filter banks play a crucial role in the vertex and spectral representation of graph signals.The notion twochannel nonsubsampled (NSGFBs) on an undirected was introduced recently.The absence downsampling/upsampling operators allows greater flexibility design NSGFBs that achieve perfect reconstruction.However take response into account has not been adequately addressed yet.Based polynomial/rational lifting scheme, this letter presents simple method to with good...

Hyperspectral image (HSI) is often corrupted by various kinds of noises. This letter proposes an innovative HSI denoising approach leveraging the graph signal processing (GSP) theory and low-rank (LR) matrix recovery model. With GSP, piecewise smoothness (PWS) property can be efficiently characterized, leading to a new regularization for denoising, termed adaptive weight total variation (AWGTV) regularization. Then, problem formulated into constrained optimization that incorporates AWGTV LR...

Target position estimation is one of the important research directions in array signal processing. In recent years, target azimuth based on graph processing (GSP) has sprung up, which provides new ideas for Direction Arrival (DoA) application. this article, by extending GSP-based DOA to joint and distance constructing a fully connected model, multi-target method GSP proposed. Firstly, connection model established related phase information linear array. For graph, Fourier transform used solve...

Hyperspectral image (HSI) is often disturbed by various kinds of noise, which brings great challenges to subsequent applications. Many the existing restoration algorithms do not scale well for HSI with large size. This article proposes a novel mixed-noise removal method size, leveraging superpixel segmentation-based technology and distributed algorithm based on graph signal processing (GSP). First, underlying structure modeled two-layer architecture graph. The upper layer, called skeleton...

A distributed iterative method is proposed to solve the problem of node (sensor) localization for large-scale wireless sensor network (WSN), by leveraging graph topology decomposition and gradient descent method. First, undirected representing WSN divided into several overlapping subgraphs. Based on subgraphs, splitting a series subproblems each which resides one subgraph. The procedure proceeded subgraphs iteration consists two operators. first operator solving subproblem in every subgraph...

This study proposes a modified Newton&apos;s algorithm to design oversampled single-prototype causal FIR nearly perfect reconstruction DFT-modulated filter banks allowing low system delays. The problem boils down an unconstrained fourth-order optimisation with respect the prototype (PF), which minimises weighted sum of transfer function distortion bank and stopband energy PF. analytic forms gradient vector Hessian matrix objective are derived, based on which, method exact line search is...

In this paper, we consider the inverse graph filtering process when original filter is a polynomial of some shift on simple connected graph. The Chebyshev approximation high order has been widely used to approximate filter. propose an iterative (ICPA) algorithm implement procedure, which feasible eliminate restoration error even using lower order. We also provide detailed convergence analysis for ICPA and distributed implementation spatially network. Numerical results are included...

This paper investigates two-dimensional (2D) 2 oversampled DFT modulated filter banks and 2D critically sampled modified (MDFT) as well their design. The structure perfect reconstruction (PR) condition of 2× are presented in terms the polyphase decompositions prototype filters (PFs). In double-prototype case, part solutions PR parameterized by imposing two-channel lifting on each pair components analysis synthesis PFs. Based parametric structure, PFs separately designed constrained quadratic...

With the development of radio technology, passive bistatic radar (PBR) will suffer from interferences not only base station that is used as illuminator opportunity (BS-IoO), but also with co-frequency or adjacent frequency (BS-CF/AF). It difficult for clutter cancellation algorithm to suppress all interferences, especially BS-CF/AF. The residual seriously affect target detection and DOA estimation. To solve this problem, a novel estimation method PBR based on compressed sensing sparse...

In this study, the lifting scheme is first employed to design two-channel biorthogonal graph filter bank. The condition parameterised by imposing a single-level structure on analysis and synthesis kernels. Based parametric structure, two kernels are separately optimised constrained quadratic programming. obtained banks of structurally perfect reconstruction. Numerical results comparison included show proposed algorithm can lead with improved performance.