27 May 2024 in IEEE Journal of Selected Topics in Signal Processing
Beamforming improves the received signal power and eliminates undesirable interference by sharpening the transmitted signal toward a specific direction, enhancing service quality in the future vehicle network. However, the traditional beam codebook has gradually failed to cope with high-speed mobile services and complex pavement conditions due to beam misalignment and channel fading. To address the challenges above, this paper proposes a transformer-based beamforming approach to achieve sensing-assisted high reliable communication. We use the multimodal data collected by the sensors at the base station for beamforming to optimize the communication performance. The proposed model employs three-dimensional (3D) ResNet-18 to extract multimodal features and leverages the transformer's merged-attention mechanism to fuse these features for beamforming. The experimental result based on real-world vision, radar, LiDAR, and position data shows the advance of our proposed method, which achieves 91.59% top-3 accuracy on average and exceeds over 30% top-1 accuracy than single-modal schemes in the high-speed environment.
20 September 2023 in IEEE Transactions on Network and Service Management
Due to the rapid development of Internet of Things, a large number of data are collected and published. Nevertheless, the process of data publication entails the risk of data privacy disclosure. Most of the related works can be largely classified into data publication based on anonymity, differential privacy, and graph. However, these existing works either cannot provide theoretically provable privacy protection, or only considered one kind of data attribute and thus cannot guarantee the desirable data utility. To this end, we propose a graph-based data publication scheme via differentially structural inference that can provide theoretically provable differential privacy for individuals, and maintain desirable data utility in many practical applications rather than a certain kind of statistics or data mining results. The main idea is to map the dataset to be published into a data graph, and further use the hierarchical random graph model in statistics to encode the structure of the data graph into dendrograms. Then, we use the Markov Chain Monte Carlo to infer an optimal dendrogram, and moreover design threshold strategy to differentially disturb the optimal dendrogram. Finally, we generate the sanitized data graph based on the disturbed optimal dendrogram, and further map the sanitized data graph to the sanitized dataset to be published. Thereafter, we theoretically prove the performance boundaries of both the privacy preservation and the data utility guarantees provided in our work. Furthermore, the extensive experimental results on two real-world datasets demonstrate that the proposed scheme is superior to the existing work and Baseline, guaranteeing the data utility and preserving the data privacy in many practical applications.
14 June 2024 in IEEE Internet of Things Journal
The 6G transport network facilitates the Internet of Everything (IoE), carrying numerous services and emphasizing the paramount importance of its reliability. However, within the transport network, the issue of co-route fibers arises. The co-route fibers, encompassing both co-cable and co-trench fibers, presents a significant latent hazard for service disruptions, posing a substantial threat to the seamless connectivity envisioned for the 6G era of pervasive IoE. The segregation of communication and sensing in the transmission network results in mutual interference between communication and sensing signals, rendering it difficult to promptly address sudden fiber interruptions. This article proposes an integrated sensing and communication (ISAC) architecture within transport networks, aiming at the online discernment of co-cable fibers, characterization of fiber optic trenches, and real-time classification of fiber vibration events. In the domain of co-cable fiber identification, our approach has successfully reduced the nuisance alarm rate to an impressive 5.3%, while simultaneously elevating the recognition accuracy to an outstanding 99.7%. As for co-trench fiber identification, our proposed methodology not only facilitates the discernment of co-trench fibers but also achieves an impressive accuracy of 97.7% in classifying fiber trenches. Moreover, in the realm of fiber state prediction, our solution has achieved a remarkable recognition accuracy of 98% across six distinct vibration events. These results underscore the robust performance of the proposed ISAC architecture, which will effectively safeguard the survivability of 6G IoE.
20 March 2024 in IEEE Internet of Things Journal
In the realm of emergency disaster relief, it is paramount to attain a thorough comprehension of the semantic information associated with the local disaster scene for strategic rescue path planning and immediate rescue operations for affected individuals. Unmanned aerial vehicle (UAV) networks are widely utilized for rapid data collection in the aftermath of disasters due to their flexibility and maneuverability, assisting in rescue decision-making. However, some disasters, such as seismic events and floods, have disrupted the initially structured ground shape information, leading to a disparate distribution of data collected by various UAV groups. This exposes traditional semantic segmentation models susceptible to shortcut bias, posing challenges in adapting to semantic segmentation tasks in disaster scenarios. Thus, this article proposes a bias-compensation augmentation learning-based semantic segmentation framework, which substantially enhances the extraction capability of semantic information. Initially, we exploit an artificial augmentation neural network for bias-awareness to determine the relative bias values of the collected image data. Subsequently, considering the limited computing power resources in UAV networks, we present a bias compensation computation offloading strategy to achieve a relatively balanced distribution of semantic information across UAV nodes, optimizing the tradeoff between network scheduling efficiency and model accuracy. We conduct reconstruction validation on the FloodNet data set, and a plethora of experimental results demonstrate that, compared to traditional methods, this approach greatly improves the accuracy of pixel-level semantic segmentation by over 86.5%. Moreover, the average combined processing time is also reduced by over 50%, enhancing the utilization efficiency of limited computational resources.
1 April 2025 in IEEE Transactions on Artificial Intelligence
Automatic driving systems face critical challenges, including limited computational resources, complex data processing demands, and disruptions caused by high vehicular mobility, all of which hinder real-time decision-making and system accuracy. Existing solutions, such as edge computing and distributed architectures, partially address these issues but often fail to integrate semantic communication and mobility-aware optimizations. To tackle these challenges, we propose a Temporal Semantic Transformer-based edge computing network architecture (TSTNet) that enhances decision-making accuracy and reduces latency in automatic driving systems. TSTNet overcomes three key challenges in automatic driving. First, it efficiently utilizes limited computational resources by optimizing the processing of large-scale multimodal data through lightweight semantic extractors and attention-based feature integration, significantly reducing computational overhead. Second, it preserves semantic and behavioral continuity by ensuring seamless transitions of vehicle behavior and surrounding scene semantics during mobility and service handovers. This ensures consistent situational awareness even in highly dynamic vehicular environments. Third, TSTNet reduces decision-making latency by leveraging semantic inheritance to minimize redundant computations, enabling real-time performance and improving the reliability of driver-assist systems. Experimental results demonstrate that TSTNet improves task accuracy by over 90% while reducing decision-making latency by over 50% compared to conventional methods. This architecture offers a scalable, efficient, and robust solution to the computational and mobility challenges in automatic driving, enabling enhanced real-time adaptability in complex traffic scenarios.