arXiv subjects

Sumei Sun

Publications and source records attributed to Sumei Sun.

At least 19 recordsLinked to original sources

arXiv2026

Toward Intelligent Skies: Signal Processing and AI Foundations of Low-Altitude Wireless Networks

The rapid growth of low-altitude aerial services and applications, driven by uncrewed aerial vehicles (UAVs), calls for a new class of digital infrastructure beyond conventional terrestrial networks. The low-altitude wireless network (LAWN) has been proposed as dynamically reconfigurable three-dimensional architectures that integrate aerial and ground nodes to provide connectivity, sensing, and control in open, safety-critical airspace. This tutorial presents a comprehensive treatment of LAWNs from the joint perspectives of artificial intelligence (AI) and signal processing. We first review the historical evolution and architectural foundations of LAWNs, introducing altitude-based layers and functional planes, and summarizing the regulatory and standardization landscape. Building on this system view, we then discuss signal processing fundamentals for LAWNs, including 3D channel and system models, performance metrics, waveform and receiver design, localization and tracking, and multi-functionality co-design. Next, we survey AI techniques for LAWNs, covering discriminative and generative models for perception, control, resource management, and security, as well as emerging paradigms such as foundation models, large language models, and digital twins for mission planning and closed-loop optimization. To illustrate AI-signal processing integration in practice, we provide a case study of an AI-driven multi-tier LAWN with hybrid satellite, high-altitude, and ground nodes. The tutorial concludes by outlining key research challenges in architecture design, signal processing-AI co-design, safety and security, experimentation, and standardization, and by highlighting opportunities for LAWNs to evolve into dependable, AI-native infrastructure for the intelligent skies.

eess.SP↗

arXiv2026

TIDE: Trustworthy and Interpretable Battery Degradation Estimation with Contextual Learning and Symbolic Distillation

Battery health estimation is fundamental for battery management in battery-powered systems, where inaccurate health states may affect control, maintenance, and service life. It becomes even more critical in intelligent connected systems, where estimation errors can propagate across interconnected devices and downstream decisions. In this paper, we propose TIDE, a trustworthy and interpretable battery degradation estimator for reliable battery health estimation. TIDE jointly considers accuracy, trustworthiness, and interpretability, which are all essential for practical deployment and downstream decision making. To realize these objectives, TIDE combines battery-domain knowledge with operational measurements in a three-component backbone. A knowledge-guided degradation prior promotes trustworthy estimation, a monotone residual component provides interpretable aging-consistent refinement, and a contextual learning component captures battery-specific operational effects for improved accuracy. The trained backbone is then distilled into a compact symbolic surrogate to provide model-level interpretability and support deployment. Experiments show that TIDE achieves strong estimation accuracy, improving overall estimation fidelity by an average of 19.7% over representative baselines. Its knowledge-guided prior and monotone residual modelling substantially reduce aging-consistency violations, supporting trustworthy estimation. Meanwhile, the backbone enables component-level interpretation, while symbolic distillation provides a compact model-level representation of the learned estimation logic. These results support the practical use of TIDE for battery health monitoring and decision support in intelligent connected systems.

cs.LG↗

arXiv2026

Physical Layer Security for FAS-Aided Short-Packet Systems: A Variable Block-Correlation Approach

This paper presents a comprehensive physical layer security (PLS) framework for fluid antenna system (FAS)-aided short-packet communications under the variable block-correlation model (VBCM). We consider a downlink wiretap scenario in which a base station transmits confidential short packets to a legitimate receiver user (RU) in the presence of an eavesdropper user (EU), where both the RU and EU are equipped with fluid antennas. Unlike existing FAS security analyses that rely on constant block-correlation models or infinite-blocklength assumptions, we incorporate the VBCM to accurately capture the non-uniform spatial correlation structure inherent in practical FAS deployments. By employing a piecewise linear approximation of the decoding error probability and Gauss-Chebyshev quadrature, we derive closed-form and asymptotic expressions for the average achievable secrecy throughput (AAST). We further prove that the AAST is monotonically non-decreasing in the number of RU ports, which reduces the three-dimensional joint optimization of transmit power, blocklength, and port number to a two-dimensional grid search (GS). Numerical results demonstrate that the FAS-aided system achieves up to an order-of-magnitude secrecy throughput improvement over conventional fixed-position antenna systems, and reveal that blocklength selection is the most critical design parameter in the joint optimization.

eess.SP↗

arXiv2026

Symbiotic Backscatter Communication: A Design Perspective on the Modulation Scheme of Backscatter Devices

Symbiotic Backscatter Communication (SBC) has emerged as a spectrum-efficient and low-power communication technology, where backscatter devices (BDs) modulate and reflect incident radio frequency (RF) signals from primary transmitters (PTs). While previous studies have assumed a circularly symmetric complex Gaussian (CSCG) distribution for the BD's signal, this assumption may not be practical because the high complexity of generating CSCG signals is not supported by the low-cost BD. In this paper, we address this gap by investigating SBC for two low-complexity modulation schemes, i.e., $M$-ary amplitude-shift keying (MASK) and $M$-ary phase-shift keying (MPSK), where BD's signals inherently deviate from CSCG distribution. Our goal is to derive the achievable rate of the PT and BD under the MASK/MPSK and to design MASK/MPSK modulation scheme for maximizing the PT's rate. Towards this end, we first derive the expressions of both the PT's rate and BD's rate. Theoretical results reveal that whether or not the BD improves the PT's rate depends on the phase of MASK/MPSK modulation, while the BD's rate is independent of this phase. We then formulate two optimization problems to maximize the PT's rate by adjusting the phase under the MASK and MPSK modulation schemes, respectively, and derive the optimal phases for each modulation scheme in closed forms. Simulation results demonstrate that the optimal phase of MASK/MPSK can ensure an improvement in the PT's rate, and reveal that a low-order ASK modulation is better than a low-order PSK for the BD in terms of improving PT's rate, especially when the direct link is not significantly weaker than the backscatter link in SBC.

cs.IT↗

arXiv2025

Symbol Detection in Ambient Backscatter Communications Under Residual Time Synchronization Errors

Ambient backscatter communications (AmBC), where a backscatter transmitter (BT) modulates and reflects ambient signals to a backscatter receiver (BR), have been deemed a low-power communication technology for the Internet of Things. Previous work on symbol detection in AmBC assumed perfect time synchronization (TS), which is unrealistic in practice. The residual TS errors (RTSE) cause \emph{partial sample mismatch}, degrading symbol detection performance. To address this, we propose a new AmBC symbol detection framework that incorporates the BT's current and adjacent symbols, as well as channel coefficients. Using energy detector (ED) as a case study, we derive both exact and approximate bit error rate (BER) expressions. Our results show that the ED's BER performance degrades significantly under RTSE, with the symbol detection threshold optimized under the assumption of perfect TS. We then derive a closed-form expression for a near-optimal symbol detection threshold that minimizes BER under RTSE. To estimate the required parameters for the detection threshold, we propose a novel method exploiting the attributes of the BR's received signal samples. The analytical results are verified by simulation results.

cs.IT↗

arXiv2025

DWM-RO: Decentralized World Models with Reasoning Offloading for SWIPT-enabled Satellite-Terrestrial HetNets

Wireless networks are undergoing a paradigm shift toward massive connectivity with energy-efficient operation, driving the integration of satellite-terrestrial architectures with simultaneous wireless information and power transfer (SWIPT). Optimizing transmit beamforming and power splitting in such systems faces formidable challenges, e.g., time-varying channels and multi-tier interference, which create a complex decision landscape where conventional model-free multi-agent reinforcement learning (MARL) suffers from sample inefficiency due to rarely-encountered state transitions and poor coordination as decentralized agents act independently. This paper proposes the Decentralized World Model with Reasoning Offloading (DWM-RO) framework to address these fundamental limitations. Specifically, each agent employs a world model to learn compact predictive representations of environment dynamics, enabling imagination-based policy training that dramatically reduces required environment interactions. An uncertainty-aware offloading gate monitors local interference levels and model reconstruction errors to trigger selective edge coordination. When activated, a lightweight latent decorrelation mechanism at the edge refines agents' strategic representations, guiding them toward orthogonal actions that minimize resource conflicts. Extensive simulations demonstrate that DWM-RO converges 5 times faster than state-of-the-art baselines while achieving 34.7% higher spectral efficiency and reducing constraint violations by 40%. In dense network scenarios with 10 users, DWM-RO maintains violation rates below 20% while baselines exceed 70%, validating superior robustness.

cs.DC↗

arXiv2025

Adaptive Phase Shift Information Compression for IRS Systems: A Prompt Conditioned Variable Rate Framework

Intelligent reflecting surfaces (IRSs) have become a vital technology for improving the spectrum and energy efficiency of forthcoming wireless networks. Nevertheless, practical implementation is obstructed by the excessive overhead associated with the frequent transmission of phase shift information (PSI) over bandwidth-constrained control lines. Current deep learning-based compression methods mitigate this problem but are constrained by elevated decoder complexity, inadequate flexibility to dynamic channels, and static compression ratios. This research presents a prompt-conditioned PSI compression system that integrates prompt learning inspired by large models into the PSI compression process to address these difficulties. A hybrid prompt technique that integrates soft prompt concatenation with feature-wise linear modulation (FiLM) facilitates adaptive encoding across diverse signal-to-noise ratios (SNRs), fading kinds, and compression ratios. Furthermore, a variable rate technique incorporates the compression ratio into the prompt embeddings through latent masking, enabling a singular model to adeptly balance reconstruction accuracy. Additionally, a lightweight depthwise convolutional gating (DWCG) decoder facilitates precise feature reconstruction with minimal complexity. Comprehensive simulations indicate that the proposed framework significantly reduces NMSE compared to traditional autoencoder baselines, while ensuring robustness across various channel circumstances and accommodating variable compression ratios within a single model. These findings underscore the framework's promise as a scalable and efficient solution for real-time IRS control in next-generation wireless networks.

eess.SP↗

arXiv2025

Fairness-Aware Computation Offloading in Wireless-Powered MEC Systems with Cooperative Energy Recycling

In this paper, cooperative energy recycling (CER) is investigated in wireless-powered mobile edge computing systems. Unlike conventional architectures that rely solely on a dedicated power source, wireless sensors are additionally enabled to recycle energy from peer transmissions. To evaluate system performance, a joint computation optimization problem is formulated that integrates local computing and computation offloading, under an alpha-fairness objective that balances total computable data and user fairness while satisfying energy, latency, and task size constraints. Due to the inherent non-convexity introduced by coupled resource variables and fairness regularization, a variable-substitution technique is employed to transform the problem into a convex structure, which is then efficiently solved using Lagrangian duality and alternating optimization. To characterize the fairness-efficiency tradeoff, closed-form solutions are derived for three representative regimes: zero fairness, common fairness, and max-min fairness, each offering distinct system-level insights. Numerical results validate the effectiveness of the proposed CER-enabled framework, demonstrating significant gains in throughput and adaptability over benchmark schemes. The tunable alpha fairness mechanism provides flexible control over performance-fairness trade-offs across diverse scenarios.

cs.IT↗

arXiv2025

Error Floor of ML-Decoded Spinal Codes in the Finite Blocklength Regime

Spinal codes is a new family of capacity-achieving rateless codes that has been shown to achieve better rate performance compared to Raptor codes, Strider codes, and rateless Low-Density Parity-Check (LDPC) codes. This correspondence addresses the performance limitations of Spinal codes in the finite block length regime, uncovering an error floor phenomenon at high Signal-to-Noise Ratios (SNRs). We develop an analytical expression to approximate the error floor and devise SNR thresholds at which the error floor initiates. Numerical results across {Additive White Gaussian Noise (AWGN), rayleigh, and nakagami-m fading channels} verify the accuracy of our analysis. The analysis and numerical results also show that transmitting more passes of symbols can lower the error floor but does not affect the SNR threshold, providing insights on the performance target, the working SNR region, and the code design.

cs.IT↗

arXiv2025

Joint Probing and Scheduling for Cache-Aided Hybrid Satellite-Terrestrial Networks

Caching is crucial in hybrid satellite-terrestrial networks to reduce latency, optimize throughput, and improve data availability by storing frequently accessed content closer to users, especially in bandwidth-limited satellite systems, requiring strategic Medium Access Control (MAC) layer. This paper addresses throughput optimization in satellite-terrestrial integrated networks through opportunistic cooperative caching. We propose a joint probing and scheduling strategy to enhance content retrieval efficiency. The strategy leverages the LEO satellite to probe satellite-to-ground links and cache states of multiple cooperative terrestrial stations, enabling dynamic user scheduling for content delivery. Using an optimal stopping theoretic approach with two levels of incomplete information, we make real-time decisions on satellite-terrestrial hybrid links and caching probing. Our threshold-based strategy optimizes probing and scheduling, significantly improving average system throughput by exploiting cooperative caching, satellite-terrestrial link transmission, and time diversity from dynamic user requests. Simulation results validate the effectiveness and practicality of the proposed strategies.

eess.SP↗

arXiv2025

Drone Controller Localization Based on TDoA

We study time difference of arrival (TDoA)-based algorithms for drone controller localization and analyze TDoA estimation in multipath channels. Building on TDoA estimation, we propose two algorithms to enhance localization accuracy in multipath environments: the Maximum Likelihood (ML) algorithm and the Least Squares Bancroft with Gauss-Newton (LS-BF-GN) algorithm. We evaluate these proposed algorithms in two typical outdoor channels: Wireless Local Area Network (WLAN) Channel F and the two-ray ground reflection (TRGR) channel. Our simulation results demonstrate that the ML and LS-BF-GN algorithms significantly outperform the LS-BF algorithm in multipath channels. To further enhance localization accuracy, we propose averaging multiple tentative location estimations. Additionally, we evaluate the impact of time synchronization errors among sensors on localization performance through simulation.

cs.IT↗

arXiv2025

Knowledge-Aware Modeling with Frequency Adaptive Learning for Battery Health Prognostics

Battery health prognostics are critical for ensuring safety, efficiency, and sustainability in modern energy systems. However, it has been challenging to achieve accurate and robust prognostics due to complex battery degradation behaviors with nonlinearity, noise, capacity regeneration, etc. Existing data-driven models capture temporal degradation features but often lack knowledge guidance, which leads to unreliable long-term health prognostics. To overcome these limitations, we propose Karma, a knowledge-aware model with frequency-adaptive learning for battery capacity estimation and remaining useful life prediction. The model first performs signal decomposition to derive battery signals in different frequency bands. A dual-stream deep learning architecture is developed, where one stream captures long-term low-frequency degradation trends and the other models high-frequency short-term dynamics. Karma regulates the prognostics with knowledge, where battery degradation is modeled as a double exponential function based on empirical studies. Our dual-stream model is used to optimize the parameters of the knowledge with particle filters to ensure physically consistent and reliable prognostics and uncertainty quantification. Experimental study demonstrates Karma's superior performance, achieving average error reductions of 50.6% and 32.6% over state-of-the-art algorithms for battery health prediction on two mainstream datasets, respectively. These results highlight Karma's robustness, generalizability, and potential for safer and more reliable battery management across diverse applications.

cs.LG↗

arXiv2025

Coordinated FMCW and OFDM for Integrated Sensing and Communication

We propose a coordinated FMCW-OFDM (Co-FMCW-OFDM) system that enables integrated sensing and communication (ISAC) by allowing sensing and communication to share the same RF front end, antennas, and spectral resources. In the proposed ISAC system, the FMCW signal is superimposed on the OFDM signal and serves dual purposes: facilitating bistatic sensing and enabling channel estimation at the receiver end. Based on proposed Co-FMCW-OFDM waveform, we propose two efficient sensing algorithms-fast cyclic correlation radar (FCCR) and digital mixing and down-sampling (DMD)- which significantly reduce system complexity while accurately estimating target range and velocity. We consider a realistic channel model where delays can take any value, not just integer multiples of the sampling period. This leads to a significantly larger number of effective paths compared to the actual number of targets, which makes the sensing, channel estimation, and interference cancellation more challenging. Leveraging the sensing results, we develop a sensing-aided effective channel estimation method which effectively reconstructs the channel under arbitrary delay condition based on successive interference cancellation and propose an interference cancellation scheme that removes the FMCW signal before the OFDM demodulation. Simulation results demonstrate that the proposed system achieves superior sensing accuracy, improved channel estimation, and lower bit error rate (BER) compared to conventional OFDM systems with embedded pilots. The proposed scheme demonstrates superior BER performance in comparison to the conventional OFDM-plus-FMCW approach.

cs.IT↗

arXiv2025

Pilot design, channel estimation, and target detection for integrated sensing and communication with OTFS

Recent studies shows that the orthogonal time frequency space (OTFS) waveform is a promising candidate for future communication. To meet users' potential demand for Integrated Sensing and Communication (ISAC) applications in 6G, the usage of OTFS for both radar sensing and wireless communication needs to be explored. In this paper, we propose a Fast Algorithm OTFS radar (FAOR) that can perform radar sensing in low complexity to detect the range and speed of the targets. It computes the 2D cyclic correlation of transmitted signal with the reordered delay Doppler (DD) domain received signals, and then generates the 2D range-Doppler map. It can be applied not only to monostatic radar but also to bistatic radar with a much lower computational complexity compared to state-of-the-art radar sensing technology. With the detected time delays and Doppler frequencies of the targets after the radar sensing, we propose a pilot-aided channel estimation method. The multifunction pilot symbol can serve the purpose of both bistatic radar sensing and channel estimation without any guard symbol added, while reducing the peak-to-average power ratio (PAPR) considerably compared to the conventional pilot design. The simulation results show that the proposed scheme outperforms the compared algorithms and gives decent performance in both radar sensing and channel estimation.

cs.IT↗

arXiv2025

FCT O-RAN: Design and Deployment of a Multi-Vendor End-to-End Private 5G Testbed

The transformation of 5G networks into software-defined, agile, intelligent and programmable architectures necessitates a paradigm shift in deployment strategies. To deliver superior performance and surpass traditional systems, public and private 5G networks must adopt software-centric cloud native frameworks that enable flexibility through tailored configurations and optimized deployment approaches. In Singapore, the Infocomm Media Development Authority (IMDA) and the National Research Foundation Singapore (NRF) launched the Future Communications Research and Development Programme (FCP) to advance the nation's communications and connectivity landscape. At the core of this initiative is the Future Communications Translation Lab (FCT) at the Singapore Institute of Technology (SIT), which focuses on advancing 5G technologies to higher readiness levels, facilitating their adoption across various industries. A key component is the deployment of FCT O-RAN, a state-of-the-art multi-vendor private 5G platform. The setup includes a 5G core network powered by Microsoft Affirmed and ENEA, O-RAN Centralized and Distributed Units from Radisys. Indoor Remote Units are deployed with Foxconn, while outdoor RUs are deployed with Benetel. To optimize the deployment of remote units, a digital twin was created using Wireless InSite, and performance evaluations were conducted for both the digital twin and the private 5G deployment. Smartphones equipped with QualiPoc were used to measure network performance. The testbed demonstrated effective performance with optimized bandwidth allocations for both indoor and outdoor environments. In the indoor setup, utilizing 50 MHz of bandwidth, a downlink throughput of 713 Mbps and an uplink throughput of 66 Mbps were achieved. Meanwhile, the outdoor setup, utilizing 40 MHz of bandwidth, achieved a downlink throughput of 371 Mbps and an uplink throughput of 55 Mbps.

cs.NI↗

arXiv2025

Vehicle-to-Everything Cooperative Perception for Autonomous Driving

Achieving fully autonomous driving with enhanced safety and efficiency relies on vehicle-to-everything cooperative perception, which enables vehicles to share perception data, thereby enhancing situational awareness and overcoming the limitations of the sensing ability of individual vehicles. Vehicle-to-everything cooperative perception plays a crucial role in extending the perception range, increasing detection accuracy, and supporting more robust decision-making and control in complex environments. This paper provides a comprehensive survey of recent developments in vehicle-to-everything cooperative perception, introducing mathematical models that characterize the perception process under different collaboration strategies. Key techniques for enabling reliable perception sharing, such as agent selection, data alignment, and feature fusion, are examined in detail. In addition, major challenges are discussed, including differences in agents and models, uncertainty in perception outputs, and the impact of communication constraints such as transmission delay and data loss. The paper concludes by outlining promising research directions, including privacy-preserving artificial intelligence methods, collaborative intelligence, and integrated sensing frameworks to support future advancements in vehicle-to-everything cooperative perception.

cs.CV↗

arXiv2025

Toward Edge General Intelligence with Agentic AI and Agentification: Concepts, Technologies, and Future Directions

The rapid expansion of sixth-generation (6G) wireless networks and the Internet of Things (IoT) has catalyzed the evolution from centralized cloud intelligence towards decentralized edge general intelligence. However, traditional edge intelligence methods, characterized by static models and limited cognitive autonomy, fail to address the dynamic, heterogeneous, and resource-constrained scenarios inherent to emerging edge networks. Agentic artificial intelligence (Agentic AI) emerges as a transformative solution, enabling edge systems to autonomously perceive multimodal environments, reason contextually, and adapt proactively through continuous perception-reasoning-action loops. In this context, the agentification of edge intelligence serves as a key paradigm shift, where distributed entities evolve into autonomous agents capable of collaboration and continual adaptation. This paper presents a comprehensive survey dedicated to Agentic AI and agentification frameworks tailored explicitly for edge general intelligence. First, we systematically introduce foundational concepts and clarify distinctions from traditional edge intelligence paradigms. Second, we analyze important enabling technologies, including compact model compression, energy-aware computing strategies, robust connectivity frameworks, and advanced knowledge representation and reasoning mechanisms. Third, we provide representative case studies demonstrating Agentic AI's capabilities in low-altitude economy networks, intent-driven networking, vehicular networks, and human-centric service provisioning, supported by numerical evaluations. Furthermore, we identify current research challenges, review emerging open-source platforms, and highlight promising future research directions to guide robust, scalable, and trustworthy Agentic AI deployments for next-generation edge environments.

cs.NI↗

arXiv2025

A Deep Learning based Signal Dimension Estimator with Single Snapshot Signal in Phased Array Radar Application

Signal dimension, defined here as the number of copies with different delays or angular shifts, is a prerequisite for many high-resolution delay estimation and direction-finding algorithms in sensing and communication systems. Thus, correctly estimating signal dimension itself becomes crucial. In this paper, we present a deep learning-based signal dimension estimator (DLSDE) with single-snapshot observation in the example application of phased array radar. Unlike traditional model-based and existing deep learning-based signal dimension estimators relying on eigen-decomposition and information criterion, to which multiple data snapshots would be needed, the proposed DLSDE uses two-dimensional convolutional neural network (2D-CNN) to automatically develop features corresponding to the dimension of the received signal. Our study shows that DLSDE significantly outperforms traditional methods in terms of the successful detection rate and resolution. In a phased array radar with 32 antenna elements, DLSDE improves detection Signal to Noise Ratio (SNR) by >15dB and resolution by >1°. This makes the proposed method suitable for distinguishing multiple signals that are spatially correlated or have small angular separation. More importantly, our solution operates with a single snapshot signal, which is incompatible with other existing deep learning-based methods.

eess.SP↗