IEEE Transactions on Signal Processing最新文献

{"title":"2D DOA Estimation of Coherent Signals Exploiting Forward-Backward Covariance Tensor","authors":"Saidur R. Pavel;Yimin D. Zhang;Shunqiao Sun","doi":"10.1109/TSP.2026.3656569","DOIUrl":"10.1109/TSP.2026.3656569","url":null,"abstract":"Coherent signals commonly arise in applications such as wireless communication and radar sensing, necessitating decorrelation before applying conventional direction-of-arrival (DOA) estimation methods. For two-dimensional (2D) sensor arrays, the covariance entries of the received signals form a tensor, which is rank-deficient when the impinging signals are coherent. This paper develops two novel decorrelation strategies to address the rank deficiency issue in 2D DOA estimation. The first strategy reconstructs a decorrelated tensor by rearranging a slice of the coherent covariance tensor and, to enhance dimensionality and increase the degrees of freedom (DOFs) of the array, integrating a backward slice. The second strategy employs forward backward spatial smoothing (FBSS) with an optimal sensor arrangement. In particular, we optimize four FBSS parameters: the numbers of overlapping subarrays along the X and Y axes, and the numbers of antennas in each subarray along the X and Y axes, in order to maximize the DOFs for a given array size. Furthermore, we derive a closed-form expression for the number of achievable DOFs. Simulation results confirm that the proposed methods outperform existing techniques in terms of DOF enhancement and DOA estimation accuracy.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"846-862"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146042759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generalization Error Analysis for Attack-Free and Byzantine-Resilient Decentralized Learning With Data Heterogeneity","authors":"Haoxiang Ye;Tao Sun;Qing Ling","doi":"10.1109/TSP.2026.3657395","DOIUrl":"10.1109/TSP.2026.3657395","url":null,"abstract":"Decentralized learning, which facilitates joint model training across geographically scattered agents, has gained significant attention in the field of signal and information processing in recent years. While the optimization errors of decentralized learning algorithms have been extensively studied, their generalization errors remain relatively under-explored. As the generalization error reflects the scalability of a trained model on unseen data and are crucial in determining the performance of a trained model in real-world applications, understanding the generalization error of decentralized learning is of paramount importance. In this paper, we present fine-grained generalization error analysis for both attack-free and Byzantine-resilient decentralized learning with heterogeneous data as well as under mild assumptions, in contrast to prior studies that consider homogeneous data and/or rely on a stringent bounded stochastic gradient assumption. Our results shed light on the impact of data heterogeneity, model initialization and stochastic gradient noise – factors that have not been closely investigated before – on the generalization error of decentralized learning. We also reveal that Byzantine attacks performed by malicious agents largely affect the generalization error, and their negative impact is inherently linked to the data heterogeneity while remaining independent on the sample size. Numerical experiments on both strongly convex and non-convex tasks are conducted to validate our theoretical findings.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"952-968"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146042847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantitative Error Feedback for Quantization Noise Reduction of Filtering Over Graphs","authors":"Xue Xian Zheng;Weihang Liu;Xin Lou;Stefan Vlaski;Tareq Y. Al-Naffouri","doi":"10.1109/TSP.2026.3664752","DOIUrl":"10.1109/TSP.2026.3664752","url":null,"abstract":"This paper introduces an innovative error feedback framework designed to mitigate quantization noise in distributed graph filtering, where communications are constrained to quantized messages. It draws from error spectrum shaping techniques from state-space digital filters, and therefore establishes connections between quantized filtering processes over different domains. In contrast to existing error compensation methods, our framework quantitatively feeds back the quantization noise for exact compensation. We examine the framework under three key scenarios: (i) deterministic graph filtering, (ii) graph filtering over random graphs, and (iii) graph filtering with random node-asynchronous updates. Rigorous theoretical analysis demonstrates that the proposed framework significantly reduces the effect of quantization noise, and we provide closed-form solutions for the optimal error feedback coefficients. Moreover, this quantitative error feedback mechanism can be seamlessly integrated into communication-efficient decentralized optimization frameworks, enabling lower error floors. Numerical experiments validate the theoretical results, consistently showing that our method outperforms conventional quantization strategies in terms of both accuracy and robustness.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"1047-1063"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146205428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust Filtering for Multi-Object Tracking Against Stealthy Measurement-Oriented Adversarial Attacks","authors":"Vahid Ghorbani;Amirali K. Gostar;Reza Hoseinnezhad;Zahir Tari;Aidan Blair;Amir Ghorbani;Giorgio Battistelli","doi":"10.1109/TSP.2026.3697369","DOIUrl":"https://doi.org/10.1109/TSP.2026.3697369","url":null,"abstract":"Measurement-oriented adversarial attacks, particularly ghost attacks, pose a significant challenge to the robustness of multi-object tracking (MOT) systems due to their covert nature. These attacks not only deceive the system and degrade tracking accuracy but also impose additional computational overhead proportional to the number of ghost targets. Under high-density attacks, the computational capacity of the sensing node can become fully saturated, potentially leading to a denial-of-service (DoS) scenario. To ensure the integrity and accessibility of multi-sensor, multi-object tracking systems, this work introduces the Average Likelihood for Attack Resilient Multi-object (ALARM) filtering method. This approach enhances efficiency and robustness by effectively rejecting false measurements during the Bayesian update. The proposed method mitigates the impact of ghost measurements at an early stage and demonstrates significantly improved computational performance compared to standard filters under high-density attacks. This makes ALARM a practical and efficient solution, particularly for large-scale networks susceptible to attacks<xref>¹</xref><fn><label>¹</label><p>The code is available at: <uri>https://github.com/ghorbani-vahid/alarm-filtering</uri>.</p></fn>.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"2159-2174"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148236211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Orthogonal Approximate Message Passing for Double Linear Transformation Model","authors":"Yufei Chen;Lei Liu;Yuhao Chi;Ying Li;Yong Liang Guan;Chau Yuen","doi":"10.1109/TSP.2026.3692526","DOIUrl":"https://doi.org/10.1109/TSP.2026.3692526","url":null,"abstract":"The double linear transformation (DLT) model <inline-formula><tex-math>$boldsymbol{Y}=boldsymbol{A}boldsymbol{X}boldsymbol{B}+boldsymbol{N}$</tex-math></inline-formula> is widely used in signal processing and wireless communications, where <inline-formula><tex-math>$boldsymbol{X}$</tex-math></inline-formula> is estimated from the noisy observation <inline-formula><tex-math>$boldsymbol{Y}$</tex-math></inline-formula> with known matrices <inline-formula><tex-math>$boldsymbol{A}$</tex-math></inline-formula> and <inline-formula><tex-math>$boldsymbol{B}$</tex-math></inline-formula>. Most existing algorithms reformulate the DLT model into a vectorized linear form with the equivalent transformation matrix <inline-formula><tex-math>$boldsymbol{B}^{rm T}otimesboldsymbol{A}$</tex-math></inline-formula>, upon which message passing principles are utilized to recover the signal. However, the Kronecker product causes numerous correlated submatrices in <inline-formula><tex-math>$boldsymbol{B}^{rm T}otimesboldsymbol{A}$</tex-math></inline-formula>, which significantly increases the matrix dimension and obscures the original statistical distribution properties of <inline-formula><tex-math>$boldsymbol{A}$</tex-math></inline-formula> and <inline-formula><tex-math>$boldsymbol{B}$</tex-math></inline-formula>. To address this challenge, we propose a novel equivalent tri-decoupled double transformation (TDDT) framework that introduces an intermediate variable <inline-formula><tex-math>$boldsymbol{Z}=boldsymbol{X}boldsymbol{B}$</tex-math></inline-formula> to decouple the double linear constraint into two single linear constraints. This decoupling can preserve both the dimension and the statistical properties of <inline-formula><tex-math>$boldsymbol{A}$</tex-math></inline-formula> and <inline-formula><tex-math>$boldsymbol{B}$</tex-math></inline-formula> in the single linear problems. Building on the TDDT framework, a double linear orthogonal approximate message passing (DL-OAMP) algorithm is proposed, which incorporates three estimators for two decoupled single linear constraints and a more sophisticated nonlinear constraint on <inline-formula><tex-math>$boldsymbol{X}$</tex-math></inline-formula> beyond the independent and identically distributed assumption. Meanwhile, a state evolution analysis is established to accurately predict the asymptotic performance of DL-OAMP. Within the TDDT framework, we further apply the proposed DL-OAMP algorithm to compressed sensing reconstruction in signal processing and massive random access for near-field communications. Numerical results demonstrate that the proposed DL-OAMP algorithm achieves superior performance compared with state-of-the-art methods in practical applications.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"2128-2144"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148236256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Byzantine-Resilient Decentralized Optimization for Joint Feature Selection in Multi-Task Networks","authors":"Dazhong Wang;Weidong Liu;Xiaojun Mao","doi":"10.1109/TSP.2026.3688176","DOIUrl":"10.1109/TSP.2026.3688176","url":null,"abstract":"This paper initiates a study of joint feature selection in multi-task networks, where agents collaboratively learn related prediction models over decentralized networks while keeping data local. Despite extensive progress in decentralized optimization and Byzantine-robust consensus, their integration to realize fully decentralized <inline-formula><tex-math>$boldsymbol{ell}_{mathbf{2,1}}$</tex-math></inline-formula>-coupled multi-task feature learning remains an open challenge. To address this gap, the paper proposes a Byzantine-resilient decentralized proximal gradient framework, named Brd-MTFL, for joint feature selection and model learning under the <inline-formula><tex-math>$boldsymbol{ell}_{mathbf{2,1}}$</tex-math></inline-formula>-regularization. The core innovation lies in a fully decentralized row-norm consensus mechanism that allows agents to compute global <inline-formula><tex-math>$boldsymbol{ell}_{mathbf 2}$</tex-math></inline-formula> row norms associated with feature groups through peer-to-peer communication, thus realizing fully decentralized joint feature selection. To counter malicious or faulty updates, a trimmed-mean-based robust aggregation rule is integrated into the row-norm estimation step, providing provable robustness to Byzantine failures. Our theoretical analysis establishes that Brd-MTFL achieves near-optimal convergence rates, thereby closing a critical theoretical gap at the intersection of Byzantine resilience, decentralized consensus on intermediate quantities, and <inline-formula><tex-math>$boldsymbol{ell}_{mathbf{2,1}}$</tex-math></inline-formula>-coupled feature selection in multi-task networks. Specifically, we demonstrate finite-time <inline-formula><tex-math>$boldsymbol{O}mathbf{(1/}boldsymbol{T}mathbf{)}$</tex-math></inline-formula> convergence under convexity and linear convergence under strong convexity, with both guarantees explicitly characterized by their dependence on network topology and the Byzantine budget. Empirical evaluations on synthetic data and real data demonstrate that Brd-MTFL achieves superior accuracy, stability, and resilience.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"1926-1941"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147753200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Shift-and-Add (SAdd): Hardware-Aware Channel Characterization With Polar NRZ Stimuli","authors":"Mitchell Cooke;Nicola Nicolici","doi":"10.1109/TSP.2026.3694993","DOIUrl":"https://doi.org/10.1109/TSP.2026.3694993","url":null,"abstract":"High-bandwidth wireline transceivers rely on stochastic adaptation algorithms, such as the Least Mean Squares (LMS) method for equalizer calibration. While effective, these methods cannot provide upper-bounded latency guarantees, require tuning step-size parameters, and can be burdened by error propagation when calibrating Decision Feedback Equalizers (DFEs). However, if a high-fidelity estimate of the channel response is known, an alternative approach is to configure an equalizer to directly compensate for channel-induced distortion. To address the challenges of low-latency adaptation in wireline transceivers, we co-design a novel hardware-aware algorithm and a digital architecture for training-based channel response identification. We define and leverage a subset of polar Non Return to Zero (NRZ) stimulus that we refer to as Shift-and-Add (SAdd) sequences, which are designed to simultaneously favor an efficient hardware design and provide low noise amplification with precisely defined estimator variance properties. For deployment in practical resource-constrained wireline transceivers, we present an algorithm and architecture with low latency, upper-bounded computation time, low memory usage, and few arithmetic operations.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"2037-2052"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148171829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid Architecture Gets Fluid: A New Paradigm for Direction-of-Arrival Estimation in 6G Networks","authors":"Ye Tian;Jiaji Ren;Tuo Wu;Wei Liu;Maged Elkashlan;Matthew C. Valenti;Naofal Al-Dhahir;Hing Cheung So","doi":"10.1109/TSP.2026.3691979","DOIUrl":"https://doi.org/10.1109/TSP.2026.3691979","url":null,"abstract":"High-precision direction-of-arrival (DOA) estimation, as a key sensing capability for 6G-enabled applications such as autonomous driving and extended reality, is increasingly dependent on the effective exploitation of spatial degrees of freedom (DOFs). This paper integrates two frontier DOFs-oriented paradigms and proposes a fluid antenna-enabled hybrid analog-digital (FA-HAD) architecture, which features an extremely lightweight front-end configuration mechanism and efficient spatial DOFs exploitation. Within this architecture, a collaborative spatial-phase sampling strategy is first developed to enable real-time 2-D DOA estimation under compressive observations, and a single-source CRLB analysis is provided to quantify the achievable performance limit, offering quantitative guidance for accuracy-overhead trade-offs. Furthermore, an efficient virtual-array spatial covariance matrix reconstruction method is proposed to recover a physically meaningful covariance representation, thereby providing a covariance-domain interface that is directly reusable by a broad class of existing covariance-based array processing and array design techniques, which strengthens the scalability and transferability of the proposed architecture. Building upon the reconstructed SCM, a Jacobi-Anger expansion based dimension-reduced MUSIC estimator is further derived for arbitrary planar arrays with a favorable computational cost. Simulation results demonstrate that the proposed FA-HAD framework attains DOA accuracy close to fully digital systems while substantially reducing RF hardware complexity and training overhead.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"2083-2099"},"PeriodicalIF":5.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148171850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sparse Array Beamforming Design With Proximinal Set Constraints","authors":"Mingyu Jiang;Heng Qiao","doi":"10.1109/TSP.2026.3717615","DOIUrl":"10.1109/TSP.2026.3717615","url":null,"abstract":"Sparse array beamforming design with sidelobe level suppression and set constraints is considered in this paper. A novel algorithmic framework named Dual Descent Variable Smoothing ADMM (DDVS-ADMM) is proposed that can incorporate a class of application-oriented constraints including power limit and constant modulus as special cases. Instead of applying any convex or nonconvex relaxations, we stick to the strict sparsity constraint while avoiding any expensive combinatorial procedures. In contrast to relevant literature on sparse array beamforming, we develop rigorous convergence guarantees in lieu of any impractical assumptions based on the recent idea of dual descent for nonconvex constrained programming. Furthermore, by adopting the variable smoothing technique, the DDVS-ADMM admits efficient closed-form gradient based variable updates. Several typical set constraints fitting DDVS-ADMM are discussed in detail and the empirical advantages of DDVS-ADMM over competing methods are demonstrated via extensive experiments.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"3333-3345"},"PeriodicalIF":5.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148596272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-Time Smoothing by Extremum Monte Carlo Methods","authors":"Karim Moussa;Simon Godsill","doi":"10.1109/TSP.2026.3719002","DOIUrl":"10.1109/TSP.2026.3719002","url":null,"abstract":"Many signal estimation problems allow for a delay in producing the estimates. Such smoothed estimates offer improved accuracy, though their solutions are less established than those of filtering. This paper introduces new simulation-based methods for fixed-lag and fixed-point smoothing in general state space models. The methods are broadly applicable and can be implemented when forward simulation of the model is feasible. Most of the computations can be performed offline, which makes the approach suitable for real-time applications, while a univariate treatment of the state vector enables its use in high-dimensional problems. The proposed methods are illustrated through simulation examples in linear, nonlinear, and non-Gaussian models.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"3376-3389"},"PeriodicalIF":5.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148633787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}