Pub Date : 2026-01-13DOI: 10.1109/TSP.2026.3653846
Yifan Wang;Xianghui Cao;Shi Jin;Mo-Yuen Chow
Federated learning (FL) has been widely regarded as a promising paradigm for privacy preservation of raw data in machine learning. Although data privacy in FL is locally protected to some extent, it is still a desideratum to enhance privacy and alleviate communication overhead caused by repetitively transmitting model parameters. Typically, these challenges are addressed separately, or jointly via a unified scheme that consists of noise-injected privacy mechanism and communication compression, which may lead to model corruption due to the introduced composite noise. In this work, we propose a novel model-splitting privacy-preserving FL (MSP-FL) scheme to achieve private FL with precise accuracy guarantee. Based upon MSP-FL, we further propose a model-splitting privacy-preserving FL with dynamic quantization (MSPDQ-FL) to mitigate the communication overhead, which incorporates a shrinking quantization interval to reduce the quantization error. We provide privacy and convergence analysis for both MSP-FL and MSPDQ-FL under non-i.i.d. dataset, partial clients participation and finite quantization level. Numerical results are presented to validate the superiority of the proposed schemes.
{"title":"A Novel Privacy Enhancement Scheme with Dynamic Quantization for Federated Learning","authors":"Yifan Wang;Xianghui Cao;Shi Jin;Mo-Yuen Chow","doi":"10.1109/TSP.2026.3653846","DOIUrl":"10.1109/TSP.2026.3653846","url":null,"abstract":"Federated learning (FL) has been widely regarded as a promising paradigm for privacy preservation of raw data in machine learning. Although data privacy in FL is locally protected to some extent, it is still a desideratum to enhance privacy and alleviate communication overhead caused by repetitively transmitting model parameters. Typically, these challenges are addressed separately, or jointly via a unified scheme that consists of noise-injected privacy mechanism and communication compression, which may lead to model corruption due to the introduced composite noise. In this work, we propose a novel model-splitting privacy-preserving FL (MSP-FL) scheme to achieve private FL with precise accuracy guarantee. Based upon MSP-FL, we further propose a model-splitting privacy-preserving FL with dynamic quantization (MSPDQ-FL) to mitigate the communication overhead, which incorporates a shrinking quantization interval to reduce the quantization error. We provide privacy and convergence analysis for both MSP-FL and MSPDQ-FL under non-i.i.d. dataset, partial clients participation and finite quantization level. Numerical results are presented to validate the superiority of the proposed schemes.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"453-470"},"PeriodicalIF":5.8,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961945","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}
Pub Date : 2026-01-13DOI: 10.1109/tsp.2026.3653790
Shouharda Ghosh, Nithin V. George
{"title":"A Generalized Family of Saturation Composition Cost Function based Robust Adaptive Filters","authors":"Shouharda Ghosh, Nithin V. George","doi":"10.1109/tsp.2026.3653790","DOIUrl":"https://doi.org/10.1109/tsp.2026.3653790","url":null,"abstract":"","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"138 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961947","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}
Pub Date : 2026-01-12DOI: 10.1109/TSP.2026.3653322
Lin Chen;Li Ge;Xue Jiang;Zhiyuan Jiang;Hongbin Li
Most existing target sensing approaches in integrated sensing and communication (ISAC) systems assume a regular time-frequency resource allocation. However, in practical ISAC systems, resources are often allocated irregularly because of the randomness of user scheduling. This paper addresses such resource-irregular scenarios by integrating the CANDECOMP/PARAFAC decomposition (CPD) framework with tensor completion. The proposed structured tensor completion and decomposition (STCD) method enhances target sensing by not only processing echo signals from irregularly allocated resource regions but also interpolating those from unallocated ones. Moreover, tensor completion reconstructs the Vandermonde structure of steering matrices. By enforcing a tensor rank-1 constraint, the STCD method leverages the Vandermonde structure to establish more relaxed uniqueness conditions for CPD compared with existing approaches. Additionally, we present the Cramér-Rao bound results for STCD in angle-range-velocity estimation, extending prior analyses from resource-regular to resource-irregular scenarios. Simulation results validate the effectiveness of the proposed STCD method for resource-irregular target sensing, demonstrating improved performance over traditional methods and its unstructured counterpart.
{"title":"Tensor-Based Target Sensing for Resource-Irregular ISAC Systems","authors":"Lin Chen;Li Ge;Xue Jiang;Zhiyuan Jiang;Hongbin Li","doi":"10.1109/TSP.2026.3653322","DOIUrl":"https://doi.org/10.1109/TSP.2026.3653322","url":null,"abstract":"Most existing target sensing approaches in integrated sensing and communication (ISAC) systems assume a regular time-frequency resource allocation. However, in practical ISAC systems, resources are often allocated irregularly because of the randomness of user scheduling. This paper addresses such resource-irregular scenarios by integrating the CANDECOMP/PARAFAC decomposition (CPD) framework with tensor completion. The proposed structured tensor completion and decomposition (STCD) method enhances target sensing by not only processing echo signals from irregularly allocated resource regions but also interpolating those from unallocated ones. Moreover, tensor completion reconstructs the Vandermonde structure of steering matrices. By enforcing a tensor rank-1 constraint, the STCD method leverages the Vandermonde structure to establish more relaxed uniqueness conditions for CPD compared with existing approaches. Additionally, we present the Cramér-Rao bound results for STCD in angle-range-velocity estimation, extending prior analyses from resource-regular to resource-irregular scenarios. Simulation results validate the effectiveness of the proposed STCD method for resource-irregular target sensing, demonstrating improved performance over traditional methods and its unstructured counterpart.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"605-621"},"PeriodicalIF":5.8,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299592","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}
Pub Date : 2026-01-12DOI: 10.1109/TSP.2026.3651805
Ángel F. García-Fernández;Giorgio Battistelli
This paper presents the distributed Poisson multi-Bernoulli (PMB) filter based on the generalised covariance intersection (GCI) fusion rule for distributed multi-object filtering. Since the exact GCI fusion of two PMB densities is intractable, we derive a principled approximation. Specifically, we approximate the power of a PMB density as an unnormalised PMB density, which corresponds to an upper bound of the PMB density. Then, the GCI fusion rule corresponds to the normalised product of two unnormalised PMB densities. We show that the result is a Poisson multi-Bernoulli mixture (PMBM), which can be expressed in closed form. Future prediction and update steps in each filter preserve the PMBM form, which can be projected back to a PMB density before the next fusion step. Experimental results show the benefits of this approach compared to other distributed multi-object filters.
{"title":"Distributed Poisson Multi-Bernoulli Filtering via Generalized Covariance Intersection","authors":"Ángel F. García-Fernández;Giorgio Battistelli","doi":"10.1109/TSP.2026.3651805","DOIUrl":"https://doi.org/10.1109/TSP.2026.3651805","url":null,"abstract":"This paper presents the distributed Poisson multi-Bernoulli (PMB) filter based on the generalised covariance intersection (GCI) fusion rule for distributed multi-object filtering. Since the exact GCI fusion of two PMB densities is intractable, we derive a principled approximation. Specifically, we approximate the power of a PMB density as an unnormalised PMB density, which corresponds to an upper bound of the PMB density. Then, the GCI fusion rule corresponds to the normalised product of two unnormalised PMB densities. We show that the result is a Poisson multi-Bernoulli mixture (PMBM), which can be expressed in closed form. Future prediction and update steps in each filter preserve the PMBM form, which can be projected back to a PMB density before the next fusion step. Experimental results show the benefits of this approach compared to other distributed multi-object filters.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"246-257"},"PeriodicalIF":5.8,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026549","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}
Modern software-defined networks, such as Open Radio Access Network (O-RAN) systems, rely on artificial intelligence (AI)-powered applications running on controllers interfaced with the radio access network. To ensure that these AI applications operate reliably at runtime, they must be properly calibrated before deployment. A promising and theoretically grounded approach to calibration is conformal prediction (CP), which enhances any AI model by transforming it into a provably reliable set predictor that provides error bars for estimates and decisions. CP requires calibration data that matches the distribution of the environment encountered during runtime. However, in practical scenarios, network controllers often have access only to data collected under different contexts – such as varying traffic patterns and network conditions – leading to a mismatch between the calibration and runtime distributions. This paper introduces a novel methodology to address this calibration-test distribution shift. The approach leverages meta-learning to develop a zero-shot estimator of distribution shifts, relying solely on contextual information. The proposed method, called meta-learned context-dependent weighted conformal prediction (ML-WCP), enables effective calibration of AI applications without requiring data from the current context. Additionally, it can incorporate data from multiple contexts to further enhance calibration reliability.
{"title":"Calibrating Wireless AI via Meta-Learned Context-Dependent Conformal Prediction","authors":"Seonghoon Yoo;Sangwoo Park;Petar Popovski;Joonhyuk Kang;Osvaldo Simeone","doi":"10.1109/TSP.2026.3650912","DOIUrl":"10.1109/TSP.2026.3650912","url":null,"abstract":"Modern software-defined networks, such as Open Radio Access Network (O-RAN) systems, rely on artificial intelligence (AI)-powered applications running on controllers interfaced with the radio access network. To ensure that these AI applications operate reliably at runtime, they must be properly calibrated before deployment. A promising and theoretically grounded approach to calibration is conformal prediction (CP), which enhances any AI model by transforming it into a provably reliable set predictor that provides error bars for estimates and decisions. CP requires calibration data that matches the distribution of the environment encountered during runtime. However, in practical scenarios, network controllers often have access only to data collected under different contexts – such as varying traffic patterns and network conditions – leading to a mismatch between the calibration and runtime distributions. This paper introduces a novel methodology to address this calibration-test distribution shift. The approach leverages meta-learning to develop a zero-shot estimator of distribution shifts, relying solely on contextual information. The proposed method, called meta-learned context-dependent weighted conformal prediction (ML-WCP), enables effective calibration of AI applications without requiring data from the current context. Additionally, it can incorporate data from multiple contexts to further enhance calibration reliability.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"423-438"},"PeriodicalIF":5.8,"publicationDate":"2026-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145902789","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}
Pub Date : 2025-12-29DOI: 10.1109/TSP.2025.3649224
Ruofan Liu;Bo Jiu;Danlei Xu;Youlin Fan;Hongwei Liu
Frequency Agile (FA) radar achieves superior anti-jamming capabilities by modulating the carrier frequency across different pulses. However, the array manifold changes with the carrier frequencies of different pulses, making it challenging to utilize multi-frequency information for Direction of Arrival (DOA) estimation effectively. This paper presents a DOA estimation method for FA radar in both random frequency mode and adaptive mode. Initially, in random frequency mode, the algorithm runs at the receiver while the transmitter’s carrier frequency varies randomly. A dictionary matrix that integrates multi-carrier frequency information is developed to estimate the DOA of multiple targets using Sparse Recovery (SR) theory. Subsequently, the theoretical performance bounds for DOA estimation in the random frequency mode are analytically derived. Utilizing the pre-estimated DOA results obtained from the random frequency mode, the adaptive mode jointly optimizes both carrier frequency and dictionary matrix to lower theoretical performance bounds, thereby improving the accuracy and resolution of multi-target DOA estimation. The experimental results validate the effectiveness of the proposed algorithm, showing significant improvements in DOA estimation performance compared to traditional methods.
{"title":"Adaptive DOA Estimation Method Based on Frequency Agile Radar With Joint Transmit-Receive Processing","authors":"Ruofan Liu;Bo Jiu;Danlei Xu;Youlin Fan;Hongwei Liu","doi":"10.1109/TSP.2025.3649224","DOIUrl":"10.1109/TSP.2025.3649224","url":null,"abstract":"Frequency Agile (FA) radar achieves superior anti-jamming capabilities by modulating the carrier frequency across different pulses. However, the array manifold changes with the carrier frequencies of different pulses, making it challenging to utilize multi-frequency information for Direction of Arrival (DOA) estimation effectively. This paper presents a DOA estimation method for FA radar in both random frequency mode and adaptive mode. Initially, in random frequency mode, the algorithm runs at the receiver while the transmitter’s carrier frequency varies randomly. A dictionary matrix that integrates multi-carrier frequency information is developed to estimate the DOA of multiple targets using Sparse Recovery (SR) theory. Subsequently, the theoretical performance bounds for DOA estimation in the random frequency mode are analytically derived. Utilizing the pre-estimated DOA results obtained from the random frequency mode, the adaptive mode jointly optimizes both carrier frequency and dictionary matrix to lower theoretical performance bounds, thereby improving the accuracy and resolution of multi-target DOA estimation. The experimental results validate the effectiveness of the proposed algorithm, showing significant improvements in DOA estimation performance compared to traditional methods.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"483-498"},"PeriodicalIF":5.8,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145894660","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}
Pub Date : 2025-12-29DOI: 10.1109/TSP.2025.3649010
Xiaotong Cheng;Setareh Maghsudi
The nodes’ interconnections on a social network often reflect their dependencies and information-sharing behaviors. Nevertheless, abnormal nodes, which significantly deviate from most of the network concerning patterns or behaviors, can lead to grave consequences. Therefore, it is imperative to design efficient online learning algorithms that robustly learn users’ preferences while simultaneously detecting anomalies. We introduce a novel bandit algorithm to address this problem. Through network knowledge, the method characterizes the users’ preferences and residuals of feature information. By learning and analyzing these preferences and residuals, it develops a personalized recommendation strategy for each user and simultaneously detects anomalies. We rigorously prove an upper bound on the regret of the proposed algorithm and experimentally compare it with several state-of-the-art collaborative contextual bandit algorithms on both synthetic and real-world datasets.
{"title":"Anomaly Detection in Networked Bandits","authors":"Xiaotong Cheng;Setareh Maghsudi","doi":"10.1109/TSP.2025.3649010","DOIUrl":"10.1109/TSP.2025.3649010","url":null,"abstract":"The nodes’ interconnections on a social network often reflect their dependencies and information-sharing behaviors. Nevertheless, abnormal nodes, which significantly deviate from most of the network concerning patterns or behaviors, can lead to grave consequences. Therefore, it is imperative to design efficient online learning algorithms that robustly learn users’ preferences while simultaneously detecting anomalies. We introduce a novel bandit algorithm to address this problem. Through network knowledge, the method characterizes the users’ preferences and residuals of feature information. By learning and analyzing these preferences and residuals, it develops a personalized recommendation strategy for each user and simultaneously detects anomalies. We rigorously prove an upper bound on the regret of the proposed algorithm and experimentally compare it with several state-of-the-art collaborative contextual bandit algorithms on both synthetic and real-world datasets.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"230-245"},"PeriodicalIF":5.8,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145894659","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}
Pub Date : 2025-12-22DOI: 10.1109/TSP.2025.3646485
Hengyu Chen;Jiazhi Ma;Mengyuan Dong;Xitong Yang;Yongzhen Li
Polarimetric phased array radar (PPAR) has the capability to suppress mainlobe and sidelobe interferences by fully utilizing multiple polarization channels. However, adaptive beamforming technique for PPAR, which generates nulls in the space-polarization domain, causes the received signals from multiple polarization channels to be weighted into one. This results in an inherent loss of target polarization, preventing the measurement of target polarization scattering matrix (PSM). In this paper, a robust adaptive beamforming (RAB) approach for PPAR is proposed to estimate the PSM while suppressing the mainlobe and sidelobe interferences. In our solution, dual-beams characterized by a pair of optimal orthogonal polarizations are adaptively formed to reconstruct the polarization channels. Furthermore, multiple uncertainty sets are devised, including a spatial uncertainty set and a pair of polarization-associated uncertainty sets, which respectively improve performance in beam polarization stability and mainlobe interference suppression. This problem is then formulated as a non-convex quadratically constrained quadratic programming (QCQP), which is transformed into a difference of convex (DC) programming problem. Subsequently, it is efficiently solved via a sequential convex programming (SCP) algorithm, incorporating an initial point selection strategy. We further conduct a thorough performance analysis focusing on three critical aspects. As a result, the dual-beams can suppress mainlobe and sidelobe interferences in the space-polarization domain while estimating the target PSM accurately. Simulation results demonstrate the validity of the proposed method.
{"title":"Robust Adaptive Beamforming for Radar Target Polarization Scattering Matrix Estimation","authors":"Hengyu Chen;Jiazhi Ma;Mengyuan Dong;Xitong Yang;Yongzhen Li","doi":"10.1109/TSP.2025.3646485","DOIUrl":"10.1109/TSP.2025.3646485","url":null,"abstract":"Polarimetric phased array radar (PPAR) has the capability to suppress mainlobe and sidelobe interferences by fully utilizing multiple polarization channels. However, adaptive beamforming technique for PPAR, which generates nulls in the space-polarization domain, causes the received signals from multiple polarization channels to be weighted into one. This results in an inherent loss of target polarization, preventing the measurement of target polarization scattering matrix (PSM). In this paper, a robust adaptive beamforming (RAB) approach for PPAR is proposed to estimate the PSM while suppressing the mainlobe and sidelobe interferences. In our solution, dual-beams characterized by a pair of optimal orthogonal polarizations are adaptively formed to reconstruct the polarization channels. Furthermore, multiple uncertainty sets are devised, including a spatial uncertainty set and a pair of polarization-associated uncertainty sets, which respectively improve performance in beam polarization stability and mainlobe interference suppression. This problem is then formulated as a non-convex quadratically constrained quadratic programming (QCQP), which is transformed into a difference of convex (DC) programming problem. Subsequently, it is efficiently solved via a sequential convex programming (SCP) algorithm, incorporating an initial point selection strategy. We further conduct a thorough performance analysis focusing on three critical aspects. As a result, the dual-beams can suppress mainlobe and sidelobe interferences in the space-polarization domain while estimating the target PSM accurately. Simulation results demonstrate the validity of the proposed method.","PeriodicalId":13330,"journal":{"name":"IEEE Transactions on Signal Processing","volume":"74 ","pages":"104-120"},"PeriodicalIF":5.8,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145807780","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：