arXiv subjects

Weijun Gao

Publications and source records attributed to Weijun Gao.

15 recordsLinked to original sources

arXiv2026

L-COIN: LLM-Assisted Counterfactual Inference for Game-Theoretic Distributed Computation Offloading in Sub-THz LEO Satellite Networks

As Space-Based Information Networks (SBINs) evolve toward high-capacity, intelligence-centric paradigms, integrating sub-Terahertz (sub-THz) communication into Low Earth Orbit (LEO) satellite constellations has emerged as a critical enabler for ultra-broadband and resilient global connectivity. By exploiting the ultra-wide bandwidth of sub-THz links to reduce transmission delays, resource-constrained ground devices can seamlessly offload compute-intensive tasks to LEO edge servers. However, satellite motion, short visibility windows, and limited onboard resources make offloading decisions highly time-varying. Existing distributed offloading schemes typically require repeated inter-device state exchange and poorly adapt to time-varying LEO topology or traffic conditions. To address these limitations, a decentralized game-theoretic offloading framework empowered by large language models (LLMs) and counterfactual inference is proposed in this paper. First, a realistic offloading system is established by integrating time-varying 3D-Walker topology. Second, a game-theoretic scheme using counterfactual inference is introduced to deduce unobserved states from local histories, eliminating global information reliance. Finally, an LLM-empowered semantic fusion algorithm is integrated into the counterfactual inference to enhance adaptability through zero-shot reasoning and self-reflection. Numerical results show that L-COIN reduces offloading cost by 10.9% to 27.7% relative to state-of-the-art baselines.

eess.SY↗

arXiv2026

Where Reasoning Diverges: Localized Multi-Agent Debate for Multi-Hop Question Answering

Multi-agent debate commonly exchanges complete rationales even when disagreements concern only a few intermediate claims. We introduce Localized Multi-Agent Debate (LMAD), an inference-time protocol that represents agent rationales as nodes, locates their earliest conflict, and restricts debate to the corresponding local segments. Guarded resolution extends a shared committed state so that later conflicts can be addressed without reopening accepted steps. We evaluate LMAD on four multi-hop question-answering benchmarks using ten backbones from four model families. Our method achieves the highest macro-averaged judge accuracy across all ten backbones, outperforming the strongest conventional baseline by up to 7.20 percentage points.

cs.AI↗

arXiv2026

CARA: Cognitive Adaptive Recommendation Agent

Recent advances in large language models and agent-based recommendation frameworks have introduced new opportunities for more flexible and context-aware recommendation. However, existing methods still largely rely on semantic matching, end-to-end generation, or loosely structured agent workflows, without explicitly modeling how user preferences are processed and translated into final decisions. To address this limitation, we propose CARA, a cognitively inspired recommendation framework that formulates recommendation as a structured decision-making process. The core intuition of CARA is that user decisions are jointly shaped by two complementary mechanisms: intuitive affective preference and deliberate rational evaluation. Accordingly, CARA organizes recommendation into two coordinated stages: candidate filtering, which narrows the search space based on coarse-grained preference constraints, and dual-perspective decision modeling, which captures recommendation decisions through affective and rational judgment. We further introduce a boundary-aware KTO strategy that prioritizes instructions the model can solve occasionally but not consistently, thereby increasing the density of informative preference signals. Extensive experiments on three Amazon Reviews domains show that CARA achieves the best performance on most evaluation metrics, with relative improvements of up to 10.15% over the baseline.

cs.IR↗

arXiv2026

Topology-Aware Two-Stage Federated Learning via Proxy Models for Sub-THz Heterogeneous LEO Communications

Federated learning (FL) has emerged as a promising distributed training paradigm for Low Earth Orbit (LEO) networks by significantly reducing communication overhead. However, its deployment faces critical challenges, e.g., topology-induced model staleness, short contact windows, and unaddressed computing heterogeneity. To address these issues, a topology-aware two-stage FL framework is proposed in this paper. First, a multi-layer physical architecture utilizing high-altitude platforms (HAPs) and Sub-THz communications is designed to extend satellite-ground contact windows and enlarge available bandwidth. Second, a proxy-model-based approach is adopted to fully utilize heterogeneous resources and enable architecture-agnostic knowledge aggregation. Finally, building upon these foundations, a topology-aware two-stage aggregation mechanism is proposed as the central algorithmic design to overcome the topology-induced staleness. The mechanism dynamically partitions LEO satellites into localized groups based on their transient HAP coverage. Within each group, LEO satellites perform asynchronous aggregation at their associated HAP to naturally tolerate computational delays without penalizing faster nodes. Subsequently, a synchronous inter-group aggregation is executed among all HAPs at the Ground Station (GS) to strictly bound the maximum staleness and guarantee stable global convergence. Numerical results demonstrate the proposed framework extends contact windows and achieves 86.59%--90.57% test accuracy, outperforming the state-of-the-art heterogeneous baseline by 16.26\%--19.80\%. Furthermore, it achieves a 1.5x to 2.2x convergence speedup, which closely approaches the ideal upper bound.

eess.SP↗

arXiv2025

TSRE: Channel-Aware Typical Set Refinement for Out-of-Distribution Detection

Out-of-Distribution (OOD) detection is a critical capability for ensuring the safe deployment of machine learning models in open-world environments, where unexpected or anomalous inputs can compromise model reliability and performance. Activation-based methods play a fundamental role in OOD detection by mitigating anomalous activations and enhancing the separation between in-distribution (ID) and OOD data. However, existing methods apply activation rectification while often overlooking channel's intrinsic characteristics and distributional skewness, which results in inaccurate typical set estimation. This discrepancy can lead to the improper inclusion of anomalous activations across channels. To address this limitation, we propose a typical set refinement method based on discriminability and activity, which rectifies activations into a channel-aware typical set. Furthermore, we introduce a skewness-based refinement to mitigate distributional bias in typical set estimation. Finally, we leverage the rectified activations to compute the energy score for OOD detection. Experiments on the ImageNet-1K and CIFAR-100 benchmarks demonstrate that our method achieves state-of-the-art performance and generalizes effectively across backbones and score functions.

cs.CV↗

arXiv2025

Diffusion-Driven Terahertz Air-Ground Communications under Dynamic Atmospheric Turbulence

The ever-increasing demand for ultra-high data rates in space-air-ground integrated networks (SAGINs) has rendered terahertz THz communications a promising technology owing to its exceptionally broad and continuous spectrum resources. Nevertheless, in air-ground (AG) scenarios, the high mobility of aircraft induces intense and rapidly fluctuating turbulence, leading to additional propagation loss that is often overlooked in existing studies. To bridge this gap, this paper presents an AI-empowered THz AG communication framework that explicitly models turbulence-induced attenuation through fluid dynamics and integrates it into an adaptive optimization paradigm for communication performance enhancement. Specifically, a fluid-dynamics-informed attenuation model is established to characterize aircraft-generated turbulence and quantify its impact on THz signal propagation. Building upon this model, a joint power-attitude optimization problem is formulated to adaptively allocate transmit power and adjust aircraft attitude for maximizing link capacity. The optimization problem is efficiently solved using a diffusion-based algorithm that learns the nonlinear relationship between flight configuration and turbulence-induced attenuation. Comprehensive numerical evaluations demonstrate that the turbulence-induced attenuation ranges from 18 to 28 dB under attacking angles between -10 degree and 10 degree at 0.7 Mach, verifying the pronounced impact of aircraft-induced turbulence on THz propagation. Furthermore, the proposed framework attains an average capacity of 11.241 bps/Hz, substantially outperforming existing strategies by 22.8% and 66.5%, and approaching approximately 98% of the theoretical capacity limit.

eess.SP↗

arXiv2025

Terahertz Aerospace Communications: Enabling Technologies and Future Directions

To achieve ubiquitous connectivity in next-generation networks through aerospace communications while maintaining high data rates, Terahertz (THz) band communications (0.1-10 THz) with large continuous bandwidths are considered a promising candidate technology. However, key enabling techniques and practical implementations of THz communications for aerospace applications remain limited. In this paper, the wireless channel characteristics, enabling communication techniques, and networking strategies for THz aerospace communications are investigated, aiming to assess their feasibility and encourage future research efforts toward system realization. Specifically, the wireless channel characteristics across various altitudes and scenarios are first analyzed, focusing on modeling the interaction between the THz wave and the external environment, from ground to outer space. Next, key enabling communication technologies, including multiple-input multiple-output (MIMO) technique, beam alignment and tracking, integrated communication and radar sensing (ICARS), and resource allocation for networking are discussed. Finally, the existing challenges and possible future directions are summarized and discussed.

eess.SP↗

arXiv2025

Joint Communication and Radar Sensing for Terahertz Space-Air-Ground Integrated Networks (SAGIN)

The transition from isolated systems to integrated solutions has driven the development of space-air-ground integrated networks (SAGIN) as well as the integration of communication and radar sensing functionalities. By leveraging the unique properties of the Terahertz (THz) band, THz joint communication and radar sensing (JCRS) supports high-speed communication and precise sensing, addressing the growing demands of SAGIN for connectivity and environmental awareness. However, most existing THz studies focus on terrestrial and static scenarios, with limited consideration for the dynamic and non-terrestrial environments of SAGIN. In this paper, the THz JCRS techniques for SAGIN are comprehensively investigated. Specifically, propagation characteristics and channel models of THz waves in non-terrestrial environments are analyzed. A link capacity comparison with millimeter-wave, THz, and free-space optical frequency bands is conducted to highlight the advantages of THz frequencies. Moreover, novel JCRS waveform design strategies are presented to achieve mutual merit of communication and radar sensing, while networking strategies are developed to overcome challenges in SAGIN such as high mobility. Furthermore, advancements in THz device technologies, including antennas and amplifiers, are reviewed to assess their roles in enabling practical JCRS implementations.

eess.SP↗

arXiv2024

Modeling and Analysis of Terahertz Wave Propagation in Charged Dust Using Extended Mie Scattering Theory

Terahertz (THz) band (0.1-10 THz) possesses multi-gigahertz continuous bandwidth resources, making it a promising frequency band for high-speed wireless communications and environment sensing. The interaction between the THz wave and the external environment has been studied for various scenarios. However, it has recently been revealed that the friction forces in dust storms as well as the irradiation of sunlight and solar wind lead to the electrification of dust particles on Earth and the Moon. The THz wave propagation in these charged dust has not been fully investigated, which is essential for THz aerial communications in dust storms and lunar communications. In this paper, a channel model for THz wave propagation in charged dust is developed for wireless communications. Specifically, an extended Mie scattering model for charged dust is first introduced, which captures the electrodynamic feature of the interaction between THz wave and charged particles. Then, the diameter and density distributions of dust particles are modeled, based on which the propagation loss of THz wave in charged dust is modeled and elaborated. Finally, numerical results on the additional loss caused by these charged dust with different sizes in the THz band are evaluated and compared. Extensive results demonstrate that as the number of dust charges increases, the extinction cross section of smaller-sized particles significantly increases, and the overall attenuation led by charged dust increases by at most 50% at 0.3 THz.

eess.SP↗

arXiv2024

Dual-Functional FMCW Waveform for Terahertz Space Debris Detection and Inter-Satellite Communications

Terahertz (THz) band communication, ranging from 0.1 THz to 10 THz, is envisioned as a key enabling technology for next-generation networks and future applications such as inter-satellite communications and environmental sensing. The surging number of space debris in Low Earth Orbit poses a big threat to orbital infrastructure and the development of the space economy. In particular, despite the ability to detect and track large-scale space debris, millions of space debris with a radius within the range of 0.1-10 cm and velocity exceeding 1 km/s remains hard to detect with conventional ground-based radars and optical telescopes. In this study, a dual-functional frequency modulated continuous waveform (FMCW) operating in the THz band is adopted for space debris sensing and inter-satellite communications. Specifically, the radar cross section of space debris with various sizes in the THz band is analyzed to demonstrate the feasibility of THz space debris detection. A joint space debris detection and inter-satellite communications based on the FMCW waveform is derived. Then, the parameter estimation and demodulation algorithms are illustrated. Extensive simulations demonstrate that the proposed method can realize high-accuracy parameter estimation of hypervelocity space debris while achieving high reliability for inter-satellite communications.

eess.SP↗

arXiv2024

Attenuation and Loss of Spatial Coherence Modeling for Atmospheric Turbulence in Terahertz UAV MIMO Channels

Terahertz (THz) wireless communications have the potential to realize ultra-high-speed and secure data transfer with miniaturized devices for unmanned aerial vehicle (UAV) communications. The atmospheric turbulence due to random airflow leads to spatial inhomogeneity of the communication medium, which is yet missing in most existing studies, leading to additional propagation loss and even loss of spatial coherence (LoSC) in MIMO systems. In this paper, the attenuation and loss of spatial coherence for atmospheric turbulence are modeled in THz UAV MIMO channels. Specifically, the frequency- and altitude-dependency of the refractive index structure constant (RISC), as a critical statistical parameter characterizing the intensity of turbulence, is first investigated. Then, the LoSC, fading, and attenuation caused by atmospheric turbulence are modeled, where the turbulence-induced fading is modeled by a Gamma-Gamma distribution, and the turbulence attenuation as a function of altitude and frequency is derived. Numerical results show that the turbulence leads to at most 10 dB attenuation with frequency less than 1 THz and distance less than 10 km. Furthermore, when the distance is 10 km and the RISC is 10^-9m^(-2/3), the loss of spatial coherence effect leads to 10 dB additional loss for a 1024*1024 ultra-massive MIMO system.

eess.SY↗

arXiv2023

A Universal Attenuation Model of Terahertz Wave in Space-Air-Ground Channel Medium

Providing continuous bandwidth over several tens of GHz, the Terahertz (THz) band (0.1-10 THz) supports space-air-ground integrated network (SAGIN) in 6G and beyond wireless networks. However, it is still mystery how THz waves interact with the channel medium in SAGIN. In this paper, a universal space-air-ground attenuation model is proposed for THz waves, which incorporates the attenuation effects induced by particles including condensed particles, molecules, and free electrons. The proposed model is developed from the insight into the attenuation effects, namely, the physical picture that attenuation is the result of collision between photons that are the essence of THz waves and particles in the environment. Based on the attenuation model, the propagation loss of THz waves in the atmosphere and the outer space are numerically assessed. The results indicate that the attenuation effects except free space loss are all negligible at the altitude higher than 50 km while they need to be considered in the atmosphere lower than 50 km. Furthermore, the capacities of THz SAGIN are evaluated in space-ground, space-sea, ground-sea, and sea-sea scenarios, respectively.

eess.SY↗

arXiv2023

Scintillation and Attenuation Modelling of Atmospheric Turbulence for Terahertz UAV Channels

Terahertz (THz) wireless communications have the potential to realize ultra-high-speed and secure data transfer with miniaturized devices for unmanned aerial vehicle (UAV) communications. Existing THz channel models for aerial scenarios assume a homogeneous medium along the line-of-sight propagation path. However, the atmospheric turbulence due to random airflow leads to temporal and spatial inhomogeneity of the communication medium, motivating analysis and modelling of the THz UAV communication channel. In this paper, we statistically modelled the scintillation and attenuation effect of turbulence on THz UAV channels. Specifically, the frequency- and altitude-dependency of the refractive index structure constant, as a critical statistical parameter characterizing the intensity of turbulence, is first investigated. Then, the scintillation characteristic and attenuation of the THz communications caused by atmospheric turbulence are modelled, where the scintillation effect is modelled by a Gamma-Gamma distribution, and the turbulence attenuation as a function of altitude and frequency is derived. Numerical simulations on the refractive index structure constant, scintillation, and attenuation in the THz band are presented to quantitatively analyze the influence of turbulence for the THz UAV channels. It is discovered that THz turbulence can lead to at most 10dB attenuation with frequency less than 1THz and distance less than 10km.

eess.SP↗

arXiv2022

On Multiple-Antenna Techniques for Physical-Layer Range Security in the Terahertz Band

Terahertz (THz) communications have naturally promising physical layer security (PLS) performance in the angular domain due to the high directivity feature brought by the ultra-massive multiple-antenna techniques. However, traditional multiple-antenna techniques fail to combat eavesdroppers residing in the THz beam sector, even when the communication distances of legitimate users and eavesdroppers are different. This THz range security challenge motivates us to study new multiple-antenna techniques to provide THz range and angular security. In this paper, we first conduct a theoretical analysis of the secrecy capacity of the multiple antenna channel under the range security scenario. Based on this, the frequency diverse array, as a candidate multiple-antenna technique, is proven ineffective in addressing the range security problem. Then, motivated by the theoretical analysis, a novel widely-spaced array and beamforming design for THz range security are proposed, which realize communications in the near-field regions. A non-constrained optimum approaching (NCOA) algorithm is developed to achieve the optimal secrecy rate. Simulation results illustrate that under the range security scenario where the eavesdropper is inside the beam sector, our proposed widely-spaced antenna communication scheme can ensure a 6 bps/Hz secrecy rate when the transmit power is 10 dBm and the propagation distance is 10 m.

cs.IT↗

arXiv2022

Molecular Absorption Effect: A Double-edged Sword of Terahertz Communications

Communications in the terahertz band (THz) (0.1--10~THz) have been regarded as a promising technology for future 6G and beyond wireless systems, to overcome the challenges of evergrowing wireless data traffic and crowded spectrum. As the frequency increases from the microwave band to the THz band, new spectrum features pose unprecedented challenges to wireless communication system design. The molecular absorption effect is one of the new THz spectrum properties, which enlarges the path loss and noise at specific frequencies. This brings in a double-edged sword for THz wireless communication systems. On one hand, from the data rate viewpoint, molecular absorption is detrimental, since it mitigates the received signal power and degrades the channel capacity. On the other hand, it is worth noticing that for wireless security and covertness, the molecular absorption effect can be utilized to safeguard THz communications among users. In this paper, the features of the molecular absorption effect and their impact on the THz system design are analyzed under various scenarios, with the ultimate goal of providing guidelines to how better exploit this unique THz phenomenon. Specifically, since the molecular absorption greatly depends on the propagation medium, different communication scenarios consisting of various media are discussed, including terrestrial, air and space, sea surface and nano-scale communications. Furthermore, two novel molecular absorption enlightened secure and covert communication schemes are presented, where the molecular absorption effect is utilized as the key and unique feature to boost security and covertness.

cs.IT↗