arXiv subjects

Rongqing Zhang

Publications and source records attributed to Rongqing Zhang.

At least 19 recordsLinked to original sources

arXiv2026

SoM-MTM: Synesthesia of Machines (SoM)-Driven Masked Token Model for Cooperative Perception over Packet Loss Channel

To support the large-scale and heterogeneous visual cooperative perception (CP) demands in next-generation mobile networks, intelligent and efficient sensory data transmission is a critical challenge. Under the emerging convergence of communication networks and agentic artificial intelligence (AI), existing research emphasizes utilizing end-to-end neural networks to simplify communication modules, which has shown promising potential for CP. However, these studies are still limited to specific channel models, cooperation modes, and perception tasks, failing to fully leverage powerful visual processing approaches to enhance universality. To address this, we propose a Synesthesia of Machines (SoM)-driven Masked Token Model, referred to as SoM-MTM, as a plug-and-play paradigm for generic visual CP. Inspired by masked image modeling methods such as MAE, it possesses great perceptual context learning capabilities to recover distorted features over packet loss channels, thereby improving information carrying efficiency. Building upon Swin Transformer, SoM-MTM further embeds prior masked information through an External Routing MoE mechanism, maximally repairing and enhancing environmental perception features during cooperation. Comprehensive experimental results confirm that SoM-MTM can consistently enhance perception performances on various tasks, especially strong generalization to unseen scenarios, while maintaining favorable model cost and scalability.

eess.IV↗

arXiv2025

Synesthesia of Machines (SoM)-Empowered Wireless Image Transmission over Complex Dynamic Channel

Wireless image transmission underpins diverse networked intelligent services and becomes an increasingly critical issue. Existing works have shown that deep learning-based joint source-channel coding (JSCC) is an effective framework to balance image transmission fidelity and data overhead. However, these studies oversimplify the communication system as a mere pipeline with noise, failing to account for the complex dynamics of wireless channels and concrete physical-layer transmission process. To address these limitations, we propose a Synesthesia of Machines (SoM)-empowered Dynamic Channel Adaptive Transmission (DCAT) scheme, designed for practical implementation in real communication scenarios. Building upon the Swin Transformer backbone, our DCAT scheme demonstrates robust adaptability to time-selective fading and channel aging effects by effectively utilizing the physical-layer transmission characteristics of wireless channels. Comprehensive experimental results confirm that DCAT consistently achieves superior performance compared with JSCC baseline approaches across all conditions. Furthermore, our neural network architecture demonstrates high scalability due to its interpretable design, offering substantial potential for cost-efficient deployment in practical applications.

eess.SP↗

arXiv2025

Synesthesia of Machines (SoM)-Aided LiDAR Point Cloud Transmission for Collaborative Perception

Collaborative perception enables more accurate and comprehensive scene understanding by learning how to share information between agents, with LiDAR point clouds providing essential precise spatial data. Due to the substantial data volume generated by LiDAR sensors, efficient point cloud transmission is essential for low-latency multi-agent collaboration. In this work, we propose an efficient, robust and applicable LiDAR point cloud transmission system via the Synesthesia of Machines (SoM), termed LiDAR Point Cloud Feature Transmission (LPC-FT), to support collaborative perception among multiple agents. Specifically, we employ a density-preserving deep point cloud compression method that encodes the complete point cloud into a downsampled efficient representation. To mitigate the effects of the wireless channel, we design a channel encoder module based on self-attention to enhance LiDAR point cloud features and a feature fusion module based on cross-attention to integrate features from transceivers. Furthermore, we utilize the nonlinear activation layer and transfer learning to improve the training of deep neural networks in the presence the digital channel noise. Experimental results demonstrate that the proposed LPC-FT is more robust and effective than traditional octree-based compression followed by channel coding, and outperforms state-of-the-art deep learning-based compression techniques and existing semantic communication methods, reducing the Chamfer Distance by 30% and improving the PSNR by 1.9 dB on average. Owing to its superior reconstruction performance and robustness against channel variations, LPC-FT is expected to support collaborative perception tasks.

eess.SP↗

arXiv2025

Synesthesia of Machines (SoM)-Based Task-Driven MIMO System for Image Transmission

To support cooperative perception (CP) of networked mobile agents in dynamic scenarios, the efficient and robust transmission of sensory data is a critical challenge. Deep learning-based joint source-channel coding (JSCC) has demonstrated promising results for image transmission under adverse channel conditions, outperforming traditional rule-based codecs. While recent works have explored to combine JSCC with the widely adopted multiple-input multiple-output (MIMO) technology, these approaches are still limited to the discrete-time analog transmission (DTAT) model and simple tasks. Given the limited performance of existing MIMO JSCC schemes in supporting complex CP tasks for networked mobile agents with digital MIMO communication systems, this paper presents a Synesthesia of Machines (SoM)-based task-driven MIMO system for image transmission, referred to as SoM-MIMO. By leveraging the structural properties of the feature pyramid for perceptual tasks and the channel properties of the closed-loop MIMO communication system, SoM-MIMO enables efficient and robust digital MIMO transmission of images. Experimental results have shown that compared with two JSCC baseline schemes, our approach achieves average mAP improvements of 6.30 and 10.48 across all SNR levels, while maintaining identical communication overhead.

eess.SP↗

arXiv2025

Decentralized Handover Parameter Optimization with MARL for Load Balancing in 5G Networks

In cellular networks, cell handover refers to the process where a device switches from one base station to another, and this mechanism is crucial for balancing the load among different cells. Traditionally, engineers would manually adjust parameters based on experience. However, the explosive growth in the number of cells has rendered manual tuning impractical. Existing research tends to overlook critical engineering details in order to simplify handover problems. In this paper, we classify cell handover into three types, and jointly model their mutual influence. To achieve load balancing, we propose a multi-agent-reinforcement-learning (MARL)-based scheme to automatically optimize the parameters. To reduce the agent interaction costs, a distributed training is implemented based on consensus approximation of global average load, and it is shown that the approximation error is bounded. Experimental results show that our proposed scheme outperforms existing benchmarks in balancing load and improving network performance.

cs.NI↗

arXiv2024

AirFogSim: A Light-Weight and Modular Simulator for UAV-Integrated Vehicular Fog Computing

Vehicular Fog Computing (VFC) is significantly enhancing the efficiency, safety, and computational capabilities of Intelligent Transportation Systems (ITS), and the integration of Unmanned Aerial Vehicles (UAVs) further elevates these advantages by incorporating flexible and auxiliary services. This evolving UAV-integrated VFC paradigm opens new doors while presenting unique complexities within the cooperative computation framework. Foremost among the challenges, modeling the intricate dynamics of aerial-ground interactive computing networks is a significant endeavor, and the absence of a comprehensive and flexible simulation platform may impede the exploration of this field. Inspired by the pressing need for a versatile tool, this paper provides a lightweight and modular aerial-ground collaborative simulation platform, termed AirFogSim. We present the design and implementation of AirFogSim, and demonstrate its versatility with five key missions in the domain of UAV-integrated VFC. A multifaceted use case is carried out to validate AirFogSim's effectiveness, encompassing several integral aspects of the proposed AirFogSim, including UAV trajectory, task offloading, resource allocation, and blockchain. In general, AirFogSim is envisioned to set a new precedent in the UAV-integrated VFC simulation, bridge the gap between theoretical design and practical validation, and pave the way for future intelligent transportation domains. Our code will be available at https://github.com/ZhiweiWei-NAMI/AirFogSim.

cs.NI↗

arXiv2024

Cloud-Edge-Terminal Collaborative AIGC for Autonomous Driving

In dynamic autonomous driving environment, Artificial Intelligence-Generated Content (AIGC) technology can supplement vehicle perception and decision making by leveraging models' generative and predictive capabilities, and has the potential to enhance motion planning, trajectory prediction and traffic simulation. This article proposes a cloud-edge-terminal collaborative architecture to support AIGC for autonomous driving. By delving into the unique properties of AIGC services, this article initiates the attempts to construct mutually supportive AIGC and network systems for autonomous driving, including communication, storage and computation resource allocation schemes to support AIGC services, and leveraging AIGC to assist system design and resource management.

eess.SY↗

arXiv2024

Multi-Modal Fusion-Based Multi-Task Semantic Communication System

In recent years, there has been significant progress in semantic communication systems empowered by deep learning techniques. It has greatly improved the efficiency of information transmission. Nevertheless, traditional semantic communication models still face challenges, particularly due to their single-task and single-modal orientation. Many of these models are designed for specific tasks, which may result in limitations when applied to multi-task communication systems. Moreover, these models often overlook the correlations among different modal data in multi-modal tasks. It leads to an incomplete understanding of complex information, causing increased communication overhead and diminished performance. To address these problems, we propose a multi-modal fusion-based multi-task semantic communication (MFMSC) framework. In contrast to traditional semantic communication approaches, MFMSC can effectively handle various tasks across multiple modalities. Furthermore, we design a fusion module based on Bidirectional Encoder Representations from Transformers (BERT) for multi-modal semantic information fusion. By leveraging the powerful semantic understanding capabilities and self-attention mechanism of BERT, we achieve effective fusion of semantic information from different modalities. We compare our model with multiple benchmarks. Simulation results show that MFMSC outperforms these models in terms of both performance and communication overhead.

eess.SP↗

arXiv2024

A Survey on Federated Learning in Intelligent Transportation Systems

The development of Intelligent Transportation System (ITS) has brought about comprehensive urban traffic information that not only provides convenience to urban residents in their daily lives but also enhances the efficiency of urban road usage, leading to a more harmonious and sustainable urban life. Typical scenarios in ITS mainly include traffic flow prediction, traffic target recognition, and vehicular edge computing. However, most current ITS applications rely on a centralized training approach where users upload source data to a cloud server with high computing power for management and centralized training. This approach has limitations such as poor real-time performance, data silos, and difficulty in guaranteeing data privacy. To address these limitations, federated learning (FL) has been proposed as a promising solution. In this paper, we present a comprehensive review of the application of FL in ITS, with a particular focus on three key scenarios: traffic flow prediction, traffic target recognition, and vehicular edge computing. For each scenario, we provide an in-depth analysis of its key characteristics, current challenges, and specific manners in which FL is leveraged. Moreover, we discuss the benefits that FL can offer as a potential solution to the limitations of the centralized training approach currently used in ITS applications.

cs.NI↗

arXiv2023

TJDR: A High-Quality Diabetic Retinopathy Pixel-Level Annotation Dataset

Diabetic retinopathy (DR), as a debilitating ocular complication, necessitates prompt intervention and treatment. Despite the effectiveness of artificial intelligence in aiding DR grading, the progression of research toward enhancing the interpretability of DR grading through precise lesion segmentation faces a severe hindrance due to the scarcity of pixel-level annotated DR datasets. To mitigate this, this paper presents and delineates TJDR, a high-quality DR pixel-level annotation dataset, which comprises 561 color fundus images sourced from the Tongji Hospital Affiliated to Tongji University. These images are captured using diverse fundus cameras including Topcon's TRC-50DX and Zeiss CLARUS 500, exhibit high resolution. For the sake of adhering strictly to principles of data privacy, the private information of images is meticulously removed while ensuring clarity in displaying anatomical structures such as the optic disc, retinal blood vessels, and macular fovea. The DR lesions are annotated using the Labelme tool, encompassing four prevalent DR lesions: Hard Exudates (EX), Hemorrhages (HE), Microaneurysms (MA), and Soft Exudates (SE), labeled respectively from 1 to 4, with 0 representing the background. Significantly, experienced ophthalmologists conduct the annotation work with rigorous quality assurance, culminating in the construction of this dataset. This dataset has been partitioned into training and testing sets and publicly released to contribute to advancements in the DR lesion segmentation research community.

eess.IV↗

arXiv2020

Robust Two-Stream Multi-Feature Network for Driver Drowsiness Detection

Drowsiness driving is a major cause of traffic accidents and thus numerous previous researches have focused on driver drowsiness detection. Many drive relevant factors have been taken into consideration for fatigue detection and can lead to high precision, but there are still several serious constraints, such as most existing models are environmentally susceptible. In this paper, fatigue detection is considered as temporal action detection problem instead of image classification. The proposed detection system can be divided into four parts: (1) Localize the key patches of the detected driver picture which are critical for fatigue detection and calculate the corresponding optical flow. (2) Contrast Limited Adaptive Histogram Equalization (CLAHE) is used in our system to reduce the impact of different light conditions. (3) Three individual two-stream networks combined with attention mechanism are designed for each feature to extract temporal information. (4) The outputs of the three sub-networks will be concatenated and sent to the fully-connected network, which judges the status of the driver. The drowsiness detection system is trained and evaluated on the famous Nation Tsing Hua University Driver Drowsiness Detection (NTHU-DDD) dataset and we obtain an accuracy of 94.46%, which outperforms most existing fatigue detection models.

cs.CV↗

arXiv2020

Towards Adaptive Semantic Segmentation by Progressive Feature Refinement

As one of the fundamental tasks in computer vision, semantic segmentation plays an important role in real world applications. Although numerous deep learning models have made notable progress on several mainstream datasets with the rapid development of convolutional networks, they still encounter various challenges in practical scenarios. Unsupervised adaptive semantic segmentation aims to obtain a robust classifier trained with source domain data, which is able to maintain stable performance when deployed to a target domain with different data distribution. In this paper, we propose an innovative progressive feature refinement framework, along with domain adversarial learning to boost the transferability of segmentation networks. Specifically, we firstly align the multi-stage intermediate feature maps of source and target domain images, and then a domain classifier is adopted to discriminate the segmentation output. As a result, the segmentation models trained with source domain images can be transferred to a target domain without significant performance degradation. Experimental results verify the efficiency of our proposed method compared with state-of-the-art methods.

cs.CV↗

arXiv2018

Flexible Energy Management Protocol for Cooperative EV-to-EV Charging

In this paper, we investigate flexible power transfer among electric vehicles (EVs) from a cooperative perspective in an EV system. First, the concept of cooperative EV-to-EV (V2V) charging is introduced, which enables active cooperation via charging/discharging operations between EVs as energy consumers and EVs as energy providers. Then, based on the cooperative V2V charging concept, a flexible energy management protocol with different V2V matching algorithms is proposed, which can help the EVs achieve more flexible and smarter charging/discharging behaviors. In the proposed energy management protocol, we define the utilities of the EVs based on the cost and profit through cooperative V2V charging and employ the bipartite graph to model the charging/discharging cooperation between EVs as energy consumers and EVs as energy providers. Based on the constructed bipartite graph, a max-weight V2V matching algorithm is proposed in order to optimize the network social welfare. Moreover, taking individual rationality into consideration, we further introduce the stable matching concepts and propose two stable V2V matching algorithms, which can yield the EV-consumer-optimal and EV-provider-optimal stable V2V matchings, respectively. Simulation results verify the efficiency of our proposed cooperative V2V charging based energy management protocol in improving the EV utilities and the network social welfare as well as reducing the energy consumption of the EVs.

eess.SY↗

arXiv2013

Truthful Mechanisms for Secure Communication in Wireless Cooperative System

To ensure security in data transmission is one of the most important issues for wireless relay networks, and physical layer security is an attractive alternative solution to address this issue. In this paper, we consider a cooperative network, consisting of one source node, one destination node, one eavesdropper node, and a number of relay nodes. Specifically, the source may select several relays to help forward the signal to the corresponding destination to achieve the best security performance. However, the relays may have the incentive not to report their true private channel information in order to get more chances to be selected and gain more payoff from the source. We propose a Vickey-Clark-Grove (VCG) based mechanism and an Arrow-d'Aspremont-Gerard-Varet (AGV) based mechanism into the investigated relay network to solve this cheating problem. In these two different mechanisms, we design different "transfer payment" functions to the payoff of each selected relay and prove that each relay gets its maximum (expected) payoff when it truthfully reveals its private channel information to the source. And then, an optimal secrecy rate of the network can be achieved. After discussing and comparing the VCG and AGV mechanisms, we prove that the AGV mechanism can achieve all of the basic qualifications (incentive compatibility, individual rationality and budget balance) for our system. Moreover, we discuss the optimal quantity of relays that the source node should select. Simulation results verify efficiency and fairness of the VCG and AGV mechanisms, and consolidate these conclusions.

cs.CR↗

arXiv2013

Relay Selection for Bidirectional AF Relay Network with Outdated CSI

Most previous researches on bidirectional relay selection (RS) typically assume perfect channel state information (CSI). However, outdated CSI, caused by the the time-variation of channel, cannot be ignored in the practical system, and it will deteriorate the performance. In this paper, the effect of outdated CSI on the performance of bidirectional amplify-and-forward RS is investigated. The optimal single RS scheme in minimizing the symbol error rate (SER) is revised by incorporating the outdated channels. The analytical expressions of end-to-end signal to noise ratio (SNR) and symbol error rate (SER) are derived in a closed-form, along with the asymptotic SER expression in high SNR. All the analytical expressions are verified by the Monte-Carlo simulations. The analytical and the simulation results reveal that once CSI is outdated, the diversity order degrades to one from full diversity. Furthermore, a multiple RS scheme is proposed and verified that this scheme is a feasible solution to compensate the diversity loss caused by outdated CSI.

cs.IT↗

arXiv2013

Capacity Analysis of Bidirectional AF Relay Selection with Imperfect Channel State Information

In this letter, we analyze the ergodic capacity of bidirectional amplify-and-forward relay selection (RS) with imperfect channel state information (CSI), i.e., outdated CSI and imperfect channel estimation. Practically, the optimal RS scheme in maximizing the ergodic capacity cannot be achieved, due to the imperfect CSI. Therefore, two suboptimal RS schemes are discussed and analyzed, in which the first RS scheme is based on the imperfect channel coefficients, and the second RS scheme is based on the predicted channel coefficients. The lower bound of the ergodic capacity with imperfect CSI is derived in a closed-form, which matches tightly with the simulation results. The results reveal that once CSI is imperfect, the ergodic capacity of bidirectional RS degrades greatly, whereas the RS scheme based on the predicted channel has better performance, and it approaches infinitely to the optimal performance, when the prediction length is sufficiently large.

cs.IT↗

arXiv2012

Physical Layer Security for Two-Way Untrusted Relaying with Friendly Jammers

In this paper, we consider a two-way relay network where two sources can communicate only through an untrusted intermediate relay, and investigate the physical layer security issue of this two-way relay scenario. Specifically, we treat the intermediate relay as an eavesdropper from which the information transmitted by the sources needs to be kept secret, despite the fact that its cooperation in relaying this information is essential. We indicate that a non-zero secrecy rate is indeed achievable in this two-way relay network even without external friendly jammers. As for the system with friendly jammers, after further analysis, we can obtain that the secrecy rate of the sources can be effectively improved by utilizing proper jamming power from the friendly jammers. Then, we formulate a Stackelberg game model between the sources and the friendly jammers as a power control scheme to achieve the optimized secrecy rate of the sources, in which the sources are treated as the sole buyer and the friendly jammers are the sellers. In addition, the optimal solutions of the jamming power and the asking prices are given and a distributed updating algorithm to obtain the Stakelberg equilibrium is provided for the proposed game. Finally, the simulations results verify the properties and the efficiency of the proposed Stackelberg game based scheme.

cs.GT↗

arXiv2011

Performance Analysis of Bidirectional Relay Selection with Imperfect Channel State Information

In this paper, we investigate the performance of bidirectional relay selection using amplify-and-forward protocol with imperfect channel state information, i.e., delay effect and channel estimation error. The asymptotic expression of end-to-end SER in high SNR regime is derived in a closed form, which indicates that the delay effect causes the loss of both coding gain and diversity order, while the channel estimation error merely affects the coding gain. Finally, analytical results are verified by Monte-Carlo simulations.

cs.NI↗