Multi-vehicle collaborative sensing has emerged as a new paradigm to boost the safety of autonomous vehicles. The cornerstone of this vision is the real-time and accurate registration of mmWave radar point clouds among multiple vehicles. To accomplish this, we designmmReg, an innovative system capable of achieving centimeter-level and real-time sensing fusion between vehicles.mmRegconsists of three major components: (i) aSAR imaging-driven point cloud generationcomponent leverages SAR imaging to image sparse and disordered radar point clouds to generate high-quality point clouds; (ii) amotion-aware frame synchronizationcomponent can achieve the spatio-temporal alignment of point clouds between vehicles for effectively mitigating the impact of asynchronous radar frames; (iii) ashared object-based registrationcomponent can capture and understand the unique global position of shared objects, supporting real-time and accurate registration. We implement and evaluatemmRegon CARLA and real-world campus datasets. The results demonstrate thatmmRegcan improve the vehicle’s sensing range by 117% in an average of 99.91 ms, achieving a 4.82x improvement in accuracy.

Video transmission, as one of the indispensable core services in the Internet of Vehicles, is confronted with numerous challenges such as the dynamics in the Internet of Vehicles environment, the limited resources, and the demand of vehicles for high-quality user Experience (Quality of Experience, QoE). In real-time video streaming scenarios, multicast optimization strategies, including convex optimization, game theory, stochastic optimization, etc., usually only group based on channel quality without considering task types and priorities, which may lead to the mixture of urgent tasks and ordinary tasks, affecting QoE. Therefore, we propose a task-aware three-stage collaborative optimization framework (TARA-IoV). The objective is to prioritize critical mission video services under limited bandwidth while maximizing the overall QoE. The framework first performs dynamic and adaptive vehicle grouping based on multi-dimensional features including task type, priority, geographical location, and channel state. Second, it conducts QoE-driven video quality layer selection leveraging Scalable Video Coding (SVC). Finally, it employs a deep reinforcement learning agent to dynamically allocate bandwidth with explicit task priority awareness under resource constraints. Evaluations on a real-world vehicle trajectory dataset demonstrate that TARA-IoV achieves improved QoE performance and more stable video delivery compared with existing schemes.

Digital twin (DT) migration technology is pivotal for ensuring seamless synchronization between physical entities and their virtual counterparts. However, in complex urban environments, high device mobility, dynamic network topology, and uncertain wireless channel conditions often lead to suboptimal migration decisions. Such inefficiencies exacerbate resource contention and diminish migration timeliness, thereby increasing DT state deviation and compromising service quality. To address these challenges, this paper proposes an AP group-enhanced digital twin edge network model, where multi-antenna access points (APs) are integrated as relay nodes to optimize channel quality and transmission reliability. Building on this, a Group-Collaborative Multi-Agent Proximal Policy Optimization (GC-MAPPO) migration strategy is presented. The strategy formulates the migration problem as a partially observable Markov decision process (POMDP). Specifically, the K-means++ clustering algorithm is first employed to construct optimal AP collaborative groups for mobile devices; subsequently, the MAPPO algorithm is utilized to derive optimal migration policies in dynamic environments. Experimental results demonstrate that, compared to existing baselines, the proposed GC-MAPPO scheme reduces the average synchronization delay by 6.69% to 37.05% and decreases DT state deviation by 29.24% to 82.91%.

3D object detection is a crucial task for autonomous vehicles to perceive traffic environments. Existing methods typically employ tightly coupled feature fusion strategies with fixed sensor combinations, but it fails to adequately capture the modality-specific characteristics, resulting in suboptimal object detection performance. To this end, we propose Chameleon, a novel multi-sensor decoupling system capable of selecting the appropriate sensor combination for object detection across diverse traffic conditions, which consists of two key components: (i) anuncertainty-aware contribution calculationcomponent leverages the uncertainty perceived by sensors to predict the parameters of traffic factors, followed by evaluating their importance; (ii) amutual information-enhanced sensor combination optimizationcomponent leverages mutual information calculations to enhance the mixture of experts and improve the reliability of sensor combination predictions, followed by achieving adaptive multi-sensor decoupling to ensure accurate 3D object detection while reducing inference latency. We implement and evaluate Chameleon using the nuScenes and nuScenes-C datasets. The experimental results show that Chameleon achieves average improvements of 1.10% in the mAP and 0.61% in the NDS compared to the state-of-the-art method across various traffic scenes.

Millimeter wave radar has recently emerged as a promising modality for enabling pervasive gesture recognition while protecting user privacy. However, personalized user behaviors, interference from unexpected actions, and long-term variability in user gestures significantly degrade the accuracy of gesture and user identity recognition, thereby compromising the quality of user experiences. To this end, we designmmGeswith four key modules: (i) afine-grained feature extractorextracts micro-level features from the time-series radar data to identify users' personalized behaviors; (ii) auser-specific feature classifierextracts coarse-grained features from a global perspective, followed by analyzing the micro-details of gesture features to recognize the user; (iii) avoting-based multi-user recognizerretrieves all pre-trained models from the user model database, followed by obtaining the probability indicators of each to return recognition results; (iv) alifelong learning modelretains previous knowledge while adjusting its feature selection capabilities using newly collected data to adapt to gesture changes. We implement and evaluatemmGesusing three self-collected real-world radar datasets, demonstrating its superior performance compared to other state-of-the-art gesture recognition methods.

Rapid urban development and the steady increase in vehicle ownership have made traffic congestion a persistent challenge for transportation management. In this context, the capability to achieve both high-precision and efficient traffic flow prediction is paramount for modern transportation systems. In response to this challenge, this paper presents GFLDT, a novel prediction framework grounded in the integration of digital twin and graph federated learning, which leverages collaborative intelligence and the distributed computing capabilities of IoT devices. The framework organizes the digital twin network into two components: client twins and a global twin. By employing graph federated learning, the model can extract temporal patterns from local data on client twins while capturing complex spatial dependencies through the global twin. Leveraging realworld traffic datasets for testing, we show that GFLDT achieves accurate traffic flow prediction while effectively ensuring data privacy through the collaborative mechanism between client and global twins.

Millimeter-wave (mmWave) radar enables privacy-preserving gesture recognition but suffers from limited training data, particularly for lying postures. Existing mmWave radar data generation methods are ineffective due to insufficient 2D video data. To this end, we design a novel system named Venus to generate realistic radar data for lying postures using few 2D videos, which addresses two key challenges including i) the simulation of diverse reflected signals and ii) few real-world data leading to low data fidelity. Venus consists of two key components: (i) a gesture sequence generation and signal simulation network, which combines several key modules, movement information extractor, spatio-temporal latent diffusion model, and mmWave signal simulator, to generate diverse gesture vertex sequences under certain conditions and simulate signal propagation characteristics to obtain coarse radar data; (ii) a meta-learning domain adaption network generates realistic radar data with few real-world data via ''meta-learning'' strategy. Extensive experiments on both generated and self-collected datasets demonstrate that Venus significantly outperforms state-of-the-art methods in recognizing gestures performed in lying postures.

Infrastructure-assisted autonomous driving has become a new paradigm that enables autonomous vehicles to fuse sensor data and improve driving safety, where a key enabling technology for achieving this vision is to real-time and accurate registering 3-D mmWave radar point clouds between the infrastructure and the vehicle. To this end, we proposeArtemis, a novel lightweight system capable of achieving real-time registration with decimeter-level localization.Artemisconsists of three components: 1) a modal association-based salient object extraction component leverages the complementary advantages of cameras and radars to extract semantics and areas of salient objects for radar point clouds; 2) a salient object shape construction component extracts the shape contour of salient objects based on their inherent geometries; and 3) a contour-guided 3-D point cloud registration component combines two key strategies, keypoint matching strategy and early exit strategy, to quickly select keypoints and transformation directions for achieving accurate registration in real-time. We implement and evaluateArtemiswith two multiview datasets collected in the CARLA platform and campus. The experiment results show thatArtemisachieves an average registration error of 0.33 m within 32.26 ms.

With the wide application of emerging technologies such as ultra-high definition video in Vehicular Edge Computing (VEC), the massive heterogeneous video data generated by vehicles have put forward higher requirements for real-time performance, energy efficiency and accuracy of processing. However, higher video analysis accuracy often leads to an increase in delay and energy consumption. How to balance the relationship between the three is an urgent problem to be solved. Meanwhile, the balanced or fixed bandwidth allocation mechanism adopted by most studies often ignores the characteristic differences of video tasks, resulting in inefficient resource allocation. At the same time, the security risks in the Internet of vehicles cannot be ignored. In order to deal with these challenges, this paper proposed a distributed task offloading framework combining Analytic Hierarchy Process (AHP) and Deep Deterministic Policy Gradient (DDPG). An adaptive bandwidth allocation mechanism based on the characteristics of video tasks is designed, and an improved blockchain consensus mechanism is introduced to ensure the optimal offloading decision in a trusted environment. Experimental results show that compared with the existing offloading schemes, the proposed algorithm reduces the task offloading delay by about 7.54%, reduces the energy consumption by about 6.37%, and improves the accuracy of video analysis by about 5.02% while ensuring security.

Multi-modal data-driven object detection typically depends on large-scale datasets. However, under adverse weather conditions, these datasets often exhibit a long-tail distribution, resulting in a substantial decline in detection performance and posing risks to system safety in applications such as autonomous driving. To this end, we propose Cerberus, a real-time and high-accuracy object detection system, which consists of two key components: (i) asimilarity-driven proposal extractioncomponent, which utilizes pre-trained detectors and convolutional neural networks to enhance low-quality features and suppress noise across image, LiDAR, and radar data, followed by aiming to extract and refine multi-modal features, thereby improving the quality of object proposals; (ii) anadaptive weight-based box enhancementcomponent, which integrates a frozen-weight regionbased convolutional network sub-network with a sensor fusion mechanism, followed by refining bounding boxes by normalizing confidence vectors and dynamically weighting multi-modal features, ensuring accurate detection under adverse weather conditions. Experimental results demonstrate that Cerberus surpasses existing baselines, yielding performance gains of 2.7% in the mean average precision and 2.0% in the nuScenes detection score under adverse weather conditions.

Millimeter wave radar is gaining traction recently as a promising modality for enabling pervasive and privacy-preserving gesture recognition. However, the lack of rich and fine-grained radar datasets hinders progress in developing generalized deep learning models for gesture recognition across various user postures (e.g., standing, sitting), positions, and scenes. To remedy this, we resort to designing a software pipeline that exploits wealthy 2D videos to generate realistic radar data, but it needs to address the challenge of simulating diversified and fine-grained reflection properties of user gestures. To this end, we designG3Rwith three key components: i) agesture reflection point generatorexpands the arm's skeleton points to form human reflection points; ii) asignal simulation modelsimulates the multipath reflection and attenuation of radar signals to output the human intensity map; iii) anencoder-decoder modelcombines asampling moduleand afitting moduleto address the differences in number and distribution of points between generated and real-world radar data for generating realistic radar data. We implement and evaluateG3Rusing 2D videos from public data sources and self-collected real-world radar data, demonstrating its superiority over other state-of-the-art approaches for gesture recognition.

Millimeter wave radar is gaining traction recently for enabling privacy-preserving human sensing. However, the lack of large-scale, dynamic radar datasets impedes progress in developing robust and generalized deep learning models for mobile sensing applications. To address this problem, we resort to designing a software pipeline that leverages wealthy dynamic videos to generate synthetic radar data, but it faces two key challenges including i) incorrect camera and human positions leading to erroneous superposition of signal intensity and ii) the signal reflection of the background and humans in mobile scenes. To this end, we designMidas++to utilize rich videos to generate realistic radar data via two components: (i) ahuman mesh fitting and calibrationcomponent calculates the camera ego-motion parameters to calibrate the extracted human positions; (ii) areflection and noise signal estimationcomponent combines several key modules,depth prediction,reflection model, andspatiotemporal noise estimation, to output coarse radar data, followed by aU-Netmodel to generate realistic radar data. We implement and evaluateMidas++with video data from public data sources and real-world radar data, demonstrating thatMidas++outperforms other state-of-the-art approaches for both activity recognition and object detection tasks.

Object detection with camera has achieved promising results using deep learning methods, but it suffers degraded performance under adverse conditions (e.g., foggy weather, poor illumination). To remedy this, some recent studies resort to leveraging the complementary mmWave radar, which is less affected by adverse conditions, and designing effective fusion methods. However, the existing early fusion methods are vulnerable to data noise, while the existing late fusion methods ignore the association of object information between feature maps in the early stage. To overcome these shortcomings, we propose a Global-Local Feature Enhancement Network (GLE-Net), a two-stage deep fusion detector, which first generates anchors from two sensors and uses an auxiliary module to locally enhance the single-branch missing proposals, and then fuses the global features from the multimodal sensors to improve final detection results. We collect two datasets under foggy weather and poor illumination conditions with diverse scenes, and conduct extensive experiments, verifying that the proposed GLE-Net surpasses other state-of-the-art methods in terms of Average Precision (AP).

With the rapid development of the Internet of Things (IoT) in 4G/5G deployments, the massive amount of network data generated by users has exploded, which has not only brought a revolution to human’s living, but also caused some malicious actors to utilize these data to attack the privacy of ordinary users. Therefore, it is crucial to identify the entity users behind multiple virtual accounts. Due to the low precision of user identification in the many-to-many mechanism of user identification, a random forest confirmation algorithm based on stable marriage matching (RFCA-SMM) is proposed in this study. It consists of three key steps: we first employ the stable marriage matching model to calculate the similarity between multiple users and utilize a scoring model to calculate the overall similarity of the users, after which candidate matching pairs are selected; second, we construct the random forest model that exploits a user similarity vector training set; afterward, the candidate matching pairs combine the secondary confirmation of the random forest model, which both improve the precision of the many-to-many user identification and protect private user data in the IoT. Extensive experiments are provided to demonstrate that the proposed algorithm improves precision rate, recall rate, and F-Measure (F1), as well as Area Under Curve (AUC).