Pub Date : 2026-01-20DOI: 10.1109/LCOMM.2026.3656455
Masahiro Higashida;Yoshiaki Inoue
We analyze the mean squared error (MSE) in real-time monitoring of an Ornstein-Uhlenbeck (OU) process with observational noise, using multiple past samples to improve estimation. Assuming a D/GI/1 queueing model for system delay, we show that the time-averaged MSE decomposes into two components: (i) a term involving the Laplace-Stieltjes transform (LST) of the age of information (AoI), and (ii) a term determined by the number of observations $m$ and the parameters of the OU process and noise. We further derive a closed-form expression for this latter term and investigate its limiting behavior.
{"title":"Analysis of the Real-Time Monitoring Error for the Ornstein–Uhlenbeck Process in the D/GI/1 Queue","authors":"Masahiro Higashida;Yoshiaki Inoue","doi":"10.1109/LCOMM.2026.3656455","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3656455","url":null,"abstract":"We analyze the mean squared error (MSE) in real-time monitoring of an Ornstein-Uhlenbeck (OU) process with observational noise, using multiple past samples to improve estimation. Assuming a D/GI/1 queueing model for system delay, we show that the time-averaged MSE decomposes into two components: (i) a term involving the Laplace-Stieltjes transform (LST) of the age of information (AoI), and (ii) a term determined by the number of observations <inline-formula> <tex-math>$m$ </tex-math></inline-formula> and the parameters of the OU process and noise. We further derive a closed-form expression for this latter term and investigate its limiting behavior.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"1002-1006"},"PeriodicalIF":4.4,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the dual-function radar-communication (DFRC) system, due to variations in communication information, the matched filter output of each DFRC waveform exhibits distinct range sidelobe structures, referred as range sidelobe modulation (RSM), which will degrade the clutter cancellation performance and increase residual clutter energy after moving target indication (MTI) processing. To solve this problem, a localized mismatch filter (LMMF) design method is proposed in this letter. First, a joint weighted optimization problem in terms of sidelobe level (SLL), similarity, and signal-to-noise ratio (SNR) loss is formulated. Then, the LMMF design method based on the alternating direction method of multipliers (ADMM) is proposed, and its convergence and complexity are discussed. Finally, several numerical results are given to demonstrate the effectiveness of the proposed method.
{"title":"Localized Mismatch Filter Design for Dual-Function Radar-Communication Waveforms","authors":"Hao Tang;Yongjun Liu;Guisheng Liao;Xuchen Liu;Heming Wang;Xiaoyang Dong","doi":"10.1109/LCOMM.2026.3655560","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3655560","url":null,"abstract":"In the dual-function radar-communication (DFRC) system, due to variations in communication information, the matched filter output of each DFRC waveform exhibits distinct range sidelobe structures, referred as range sidelobe modulation (RSM), which will degrade the clutter cancellation performance and increase residual clutter energy after moving target indication (MTI) processing. To solve this problem, a localized mismatch filter (LMMF) design method is proposed in this letter. First, a joint weighted optimization problem in terms of sidelobe level (SLL), similarity, and signal-to-noise ratio (SNR) loss is formulated. Then, the LMMF design method based on the alternating direction method of multipliers (ADMM) is proposed, and its convergence and complexity are discussed. Finally, several numerical results are given to demonstrate the effectiveness of the proposed method.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"927-931"},"PeriodicalIF":4.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1109/LCOMM.2026.3655827
Mert İlgüy;Berna Özbek;Didier Le Ruyet
Reconfigurable intelligent surfaces (RIS) have emerged as an important technology for next-generation wireless networks by intelligently manipulating the wireless propagation environment. Beyond Diagonal RIS (BD-RIS) extends the traditional RIS architecture by allowing non-diagonal reflection matrices, enabling more flexible signal manipulation. Transmissive RIS (T-RIS), on the other hand, facilitates the transmission of signals through the metasurfaces. In this paper, we propose a novel design called transmissive BD-RIS (T-BD-RIS), which integrates the functionalities of BD-RIS and T-RIS to enhance the user data rate. We design an algorithm for the group connected (GC) configuration, which jointly optimizes the beamforming at the base station, the T-BD-RIS transmission matrix, and the receive combiner at the user side. The fully connected (FC) and single connected (SC) cases are special instances of the proposed generic GC design, obtained by an appropriate choice of the number of groups. We evaluate the performance of various schemes, demonstrating the potential of the proposed approach.
{"title":"On Group Connected Transmissive Beyond Diagonal RIS for MIMO Systems","authors":"Mert İlgüy;Berna Özbek;Didier Le Ruyet","doi":"10.1109/LCOMM.2026.3655827","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3655827","url":null,"abstract":"Reconfigurable intelligent surfaces (RIS) have emerged as an important technology for next-generation wireless networks by intelligently manipulating the wireless propagation environment. Beyond Diagonal RIS (BD-RIS) extends the traditional RIS architecture by allowing non-diagonal reflection matrices, enabling more flexible signal manipulation. Transmissive RIS (T-RIS), on the other hand, facilitates the transmission of signals through the metasurfaces. In this paper, we propose a novel design called transmissive BD-RIS (T-BD-RIS), which integrates the functionalities of BD-RIS and T-RIS to enhance the user data rate. We design an algorithm for the group connected (GC) configuration, which jointly optimizes the beamforming at the base station, the T-BD-RIS transmission matrix, and the receive combiner at the user side. The fully connected (FC) and single connected (SC) cases are special instances of the proposed generic GC design, obtained by an appropriate choice of the number of groups. We evaluate the performance of various schemes, demonstrating the potential of the proposed approach.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"947-951"},"PeriodicalIF":4.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1109/LCOMM.2026.3655158
Yuan Ai;Xidong Mu;Pengbo Si;Yuanwei Liu
This letter proposes a novel pinching antenna systems (PASS) enabled non-orthogonal multiple access (NOMA) multi-access edge computing (MEC) framework. An optimization problem is formulated to minimize the maximum task delay by optimizing offloading ratios, transmit powers, and pinching antenna (PA) positions, subject to constraints on maximum transmit power, user energy budgets, and minimum PA separation to mitigate coupling effects. To address the non-convex problem, a bisection search-based alternating optimization (AO) algorithm is developed, where each subproblem is iteratively solved for a given task delay. Numerical simulations demonstrate that the proposed framework significantly reduces the task delay compared to benchmark schemes.
{"title":"Delay Minimization in Pinching-Antenna-Enabled NOMA-MEC Networks","authors":"Yuan Ai;Xidong Mu;Pengbo Si;Yuanwei Liu","doi":"10.1109/LCOMM.2026.3655158","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3655158","url":null,"abstract":"This letter proposes a novel pinching antenna systems (PASS) enabled non-orthogonal multiple access (NOMA) multi-access edge computing (MEC) framework. An optimization problem is formulated to minimize the maximum task delay by optimizing offloading ratios, transmit powers, and pinching antenna (PA) positions, subject to constraints on maximum transmit power, user energy budgets, and minimum PA separation to mitigate coupling effects. To address the non-convex problem, a bisection search-based alternating optimization (AO) algorithm is developed, where each subproblem is iteratively solved for a given task delay. Numerical simulations demonstrate that the proposed framework significantly reduces the task delay compared to benchmark schemes.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"962-966"},"PeriodicalIF":4.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146082090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this letter, we address efficient communication and resource allocation for vehicular networks in urban environments. Due to spectrum reuse and rapidly varying interference, joint spectrum allocation and power control become particularly challenging in such scenarios. To tackle this problem, we formulate a system-level optimization framework and propose a multi-agent reinforcement learning approach termed the Predictive Residual Q-Network Method (P-RQNM). An LSTM-based predictor is employed to predict short-term interference from historical observations, while a residual Q-network refines individual Q-values using trajectory statistics to improve decision consistency and contribute to more stable training behavior. Simulation results under urban vehicular scenarios show that P-RQNM outperforms baseline multi-agent value-decomposition methods in terms of link capacity, transmission reliability, and convergence behavior, demonstrating robust performance across diverse traffic densities and loads and practical potential for urban vehicular networks.
{"title":"P-RQNM: Multi-Agent Resource Allocation Optimization in Vehicular Networks via Predictive Residual Q-Networks","authors":"Hui Ji;Fei Ding;Xuanjie Tong;Guangying Wang;Yue Zheng","doi":"10.1109/LCOMM.2026.3655445","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3655445","url":null,"abstract":"In this letter, we address efficient communication and resource allocation for vehicular networks in urban environments. Due to spectrum reuse and rapidly varying interference, joint spectrum allocation and power control become particularly challenging in such scenarios. To tackle this problem, we formulate a system-level optimization framework and propose a multi-agent reinforcement learning approach termed the Predictive Residual Q-Network Method (P-RQNM). An LSTM-based predictor is employed to predict short-term interference from historical observations, while a residual Q-network refines individual Q-values using trajectory statistics to improve decision consistency and contribute to more stable training behavior. Simulation results under urban vehicular scenarios show that P-RQNM outperforms baseline multi-agent value-decomposition methods in terms of link capacity, transmission reliability, and convergence behavior, demonstrating robust performance across diverse traffic densities and loads and practical potential for urban vehicular networks.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"1071-1075"},"PeriodicalIF":4.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146223582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1109/LCOMM.2026.3654956
Mingkun Li;Pengyu Wang;Yuhan Dong;Jinshu Chen;Zhaocheng Wang
With the development of wireless communication devices, due to limited spectrum resources and rising interference, the importance of effective spectrum sensing techniques has drawn much attention. Hereby, modulation recognition serves as a cornerstone for non-cooperative communications and anti-jamming operations. While deep learning becomes popular through autonomous feature extraction, its vulnerability to adversarial attacks poses critical security risks. To address this challenge, FlowSlicer is proposed based on diffusion models for the modulation recognition domain. Furthermore, we explore a segmented recognition strategy for communication signals and propose an aggregation algorithm to enhance the modulation recognition. Simulation results validate the robustness of FlowSlicer under various adversarial attack strategies.
{"title":"Adversarial Defense in Modulation Recognition via Diffusion and Segment-Wise Classification","authors":"Mingkun Li;Pengyu Wang;Yuhan Dong;Jinshu Chen;Zhaocheng Wang","doi":"10.1109/LCOMM.2026.3654956","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3654956","url":null,"abstract":"With the development of wireless communication devices, due to limited spectrum resources and rising interference, the importance of effective spectrum sensing techniques has drawn much attention. Hereby, modulation recognition serves as a cornerstone for non-cooperative communications and anti-jamming operations. While deep learning becomes popular through autonomous feature extraction, its vulnerability to adversarial attacks poses critical security risks. To address this challenge, FlowSlicer is proposed based on diffusion models for the modulation recognition domain. Furthermore, we explore a segmented recognition strategy for communication signals and propose an aggregation algorithm to enhance the modulation recognition. Simulation results validate the robustness of FlowSlicer under various adversarial attack strategies.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"987-991"},"PeriodicalIF":4.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1109/LCOMM.2026.3655084
Jingjing Zhao;Qingyi Huang;Kaiquan Cai;Quan Zhou;Xidong Mu;Yuanwei Liu
A point-to-point movable element (ME) enabled reconfigurable intelligent surface (ME-RIS) communication system is investigated, where each element position can be flexibly adjusted to create favorable channel conditions. For maximizing the communication rate, an efficient ME position optimization approach is proposed. Specifically, by characterizing the cascaded channel power gain in an element-wise manner, the position of each ME is iteratively updated by invoking the successive convex approximation method. Numerical results unveil that: 1) proposed element-wise ME position optimization algorithm outperforms the standard gradient ascent algorithm (GAA) which is easily trapped in local optima and 2) ME-RIS significantly improves the communication rate compared to the conventional RIS with fixed-position elements.
{"title":"Movable-Element RIS-Aided Wireless Communications: An Element-Wise Position Optimization Approach","authors":"Jingjing Zhao;Qingyi Huang;Kaiquan Cai;Quan Zhou;Xidong Mu;Yuanwei Liu","doi":"10.1109/LCOMM.2026.3655084","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3655084","url":null,"abstract":"A point-to-point movable element (ME) enabled reconfigurable intelligent surface (ME-RIS) communication system is investigated, where each element position can be flexibly adjusted to create favorable channel conditions. For maximizing the communication rate, an efficient ME position optimization approach is proposed. Specifically, by characterizing the cascaded channel power gain in an element-wise manner, the position of each ME is iteratively updated by invoking the successive convex approximation method. Numerical results unveil that: 1) proposed element-wise ME position optimization algorithm outperforms the standard gradient ascent algorithm (GAA) which is easily trapped in local optima and 2) ME-RIS significantly improves the communication rate compared to the conventional RIS with fixed-position elements.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"967-971"},"PeriodicalIF":4.4,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-16DOI: 10.1109/LCOMM.2026.3651317
Qiyuan Li;Qin Huang
This letter introduces integrated interleaved (II) codes into product codes. In II codes, a shared redundancy check relationship protects the first-layer code, which helps correct errors that individual component codes cannot correct. Thus, this letter introduces integrated interleaving coding scheme to the component codes of the product code, enabling it to correct the minimum error patterns of the original product code. The simulation results show that this code outperforms existing schemes on both the binary erasure channel and the additive white Gaussian noise channel.
{"title":"Product Code With Integrated Interleaved Component Codes","authors":"Qiyuan Li;Qin Huang","doi":"10.1109/LCOMM.2026.3651317","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3651317","url":null,"abstract":"This letter introduces integrated interleaved (II) codes into product codes. In II codes, a shared redundancy check relationship protects the first-layer code, which helps correct errors that individual component codes cannot correct. Thus, this letter introduces integrated interleaving coding scheme to the component codes of the product code, enabling it to correct the minimum error patterns of the original product code. The simulation results show that this code outperforms existing schemes on both the binary erasure channel and the additive white Gaussian noise channel.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"897-901"},"PeriodicalIF":4.4,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145982207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-15DOI: 10.1109/LCOMM.2025.3650674
Gehan Sathsara Vithanage;Dushantha Nalin K. Jayakody;Ioannis Krikidis
In this work a receiver-centric waveform design technique for simultaneous wireless information and power transfer (SWIPT) is proposed, eliminating the traditional trade-off between energy harvesting (EH) efficiency and information transfer (IT) integrity. By injecting pulses into the receiver, the peak-to-average power ratio (PAPR) of the received signal is increased, using diode nonlinearity to enhance EH without affecting IT. Particle swarm optimization (PSO) is used to tune the pulse parameters to obtain the maximum harvest power under practical constraints. The Monte Carlo simulation results demonstrate superior EH performance compared to existing waveform optimization schemes. The method remains robust under common IT optimizations, such as selective mapping (SLM) and partial transmit sequence (PTS), confirming its compatibility and scalability for real-world SWIPT systems.
{"title":"Receiver-Centric Waveform Design: A New Frontier in SWIPT","authors":"Gehan Sathsara Vithanage;Dushantha Nalin K. Jayakody;Ioannis Krikidis","doi":"10.1109/LCOMM.2025.3650674","DOIUrl":"https://doi.org/10.1109/LCOMM.2025.3650674","url":null,"abstract":"In this work a receiver-centric waveform design technique for simultaneous wireless information and power transfer (SWIPT) is proposed, eliminating the traditional trade-off between energy harvesting (EH) efficiency and information transfer (IT) integrity. By injecting pulses into the receiver, the peak-to-average power ratio (PAPR) of the received signal is increased, using diode nonlinearity to enhance EH without affecting IT. Particle swarm optimization (PSO) is used to tune the pulse parameters to obtain the maximum harvest power under practical constraints. The Monte Carlo simulation results demonstrate superior EH performance compared to existing waveform optimization schemes. The method remains robust under common IT optimizations, such as selective mapping (SLM) and partial transmit sequence (PTS), confirming its compatibility and scalability for real-world SWIPT systems.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"932-936"},"PeriodicalIF":4.4,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-14DOI: 10.1109/LCOMM.2026.3654375
Shaoshuai Jiang;Shufang Li;Liang Yin
Modulation recognition of radar signals is a core challenge in electronic countermeasures under complex electromagnetic environments. Deep learning demonstrates significant promise for signal analysis. Yet real-world applications often involve a broad Signal-to-Noise Ratio (SNR) range, spanning from low to high levels. Such scenarios present dual challenges: key feature loss and inflexible feature capture. To tackle these, we propose a progressive solution. It integrates a channel-level fusion mechanism for multiple time-frequency images (TFIs) and multi-scale dynamic cascaded attention (MDCA) module. First, the channel-level fusion mechanism reconstructs TFIs into combined feature maps for enhanced representation. Meanwhile, the MDCA uses variable-sized window groups across layers to capture fine-grained local and global features. Experiments on eight typical radar signals (SNR range from -21 dB to 3 dB) show the proposed method’s average accuracy outperforms traditional models. This verifies the engineering practicality of our approach in radar signal modulation recognition scenarios.
雷达信号的调制识别是复杂电磁环境下电子对抗的核心问题。深度学习在信号分析方面展示了巨大的前景。然而，现实世界的应用通常涉及广泛的信噪比（SNR）范围，从低到高。这样的场景提出了双重挑战：关键特征丢失和不灵活的特征捕获。为了解决这些问题，我们提出了一个渐进的解决方案。它集成了信道级多时频图像（tfi）融合机制和多尺度动态级联注意（MDCA）模块。首先，通道级融合机制将tfi重构为组合特征映射以增强表征。同时，MDCA跨层使用可变大小的窗口组来捕获细粒度的局部和全局特征。在8个典型雷达信号（信噪比范围为-21 dB ~ 3 dB）上的实验表明，该方法的平均精度优于传统模型。验证了该方法在雷达信号调制识别场景下的工程实用性。
{"title":"Feature Fusion Based on Multi-Scale Cascaded Attention for Radar Signal Modulation Recognition","authors":"Shaoshuai Jiang;Shufang Li;Liang Yin","doi":"10.1109/LCOMM.2026.3654375","DOIUrl":"https://doi.org/10.1109/LCOMM.2026.3654375","url":null,"abstract":"Modulation recognition of radar signals is a core challenge in electronic countermeasures under complex electromagnetic environments. Deep learning demonstrates significant promise for signal analysis. Yet real-world applications often involve a broad Signal-to-Noise Ratio (SNR) range, spanning from low to high levels. Such scenarios present dual challenges: key feature loss and inflexible feature capture. To tackle these, we propose a progressive solution. It integrates a channel-level fusion mechanism for multiple time-frequency images (TFIs) and multi-scale dynamic cascaded attention (MDCA) module. First, the channel-level fusion mechanism reconstructs TFIs into combined feature maps for enhanced representation. Meanwhile, the MDCA uses variable-sized window groups across layers to capture fine-grained local and global features. Experiments on eight typical radar signals (SNR range from -21 dB to 3 dB) show the proposed method’s average accuracy outperforms traditional models. This verifies the engineering practicality of our approach in radar signal modulation recognition scenarios.","PeriodicalId":13197,"journal":{"name":"IEEE Communications Letters","volume":"30 ","pages":"997-1001"},"PeriodicalIF":4.4,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146175689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","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：