检索结果-内蒙古大学图书馆

Autonomous Driving via Brain-Inspired Causality-Aware Contrastive Learning with Time-Frequency Prediction

学校读者我要写书评

暂无评论

IEEE Internet of Things Journal 2025年

作者： Wang, Chengyu Wang, Luhan Lu, Zhaoming Wu, Celimuge Shou, Guochu Wen, Xiangming Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence Beijing Laboratory of Advanced Information Networks Beijing China Meta-Networking Research Center The University of Electro-Communications Tokyo182-8585 Japan

Developing trustworthy reinforcement learning (RL) agents for safety-critical control tasks, such as end-to-end autonomous driving, has been a longstanding challenge due to low sample efficiency. Prior works have attempted to address this challenge by performing self-supervised auxiliary tasks like self-reconstruction or predicting long-term future states. However, there still remain unexplored sequential features and causality relationships inherent in sequential state, action, and reward signals in the frequency domain. To fully exploit the temporal and frequential features, we propose a contrastive RL framework called BRain-Inspired causalitY-Aware coNTrastive learning (BRYANT) to achieve efficient representation learning and human-like autonomous driving. Different from existing temporal predictive methods, we transform the sequential latent representations, reward, and action signals into the frequency domain, followed by the symmetric temporal prediction pattern for real and imaginary parts of the frequential signals. To capture the temporal causality for the latent representations, we introduce a brain-inspired network structure called Closed-form Continuous-time (CfC) network to parameterize the derivative of the latent representations and establish the neural dynamic model. Experimental results conducted in the CARLA simulator demonstrate the effectiveness of BRYANT in efficient representation learning, enabling agents to concentrate on potential risks and decrease the collision rate compared to several state-of-the-art RL methods. Furthermore, through the visualization of the latent representation prediction process, we reveal the causal relationships between the critic Q values and the latent representation vectors in the frequency domain, and demonstrate the effectiveness of the frequency domain prediction. © 2014 IEEE.

关键词： Reinforcement learning

Converged Service-based architecture for Next-Generation Mobile Communication networks

学校读者我要写书评

暂无评论

Converged Service-based Architecture for Next-Generation Mob...

IEEE Conference on Wireless Communications and networking

作者： Keliang Du Luhan Wang Zishen Zhu Yunan Yan Xiangming Wen Beijing Laboratory of Advanced Information Networks BUPT Beijing China Beijing Key Laboratory of Network System Architecture and Convergence BUPT Beijing China

Software Defined network (SDN) and network Function Virtualization (NFV) technologies have driven mobile communication networks to evolve toward a Service-Based architecture (SBA) with great flexibility to meet the highly-dynamic requirements of vertical applications. The SBA has been applied into 5 th Generation (5G) core network and is evolving toward service-based end-to-end networks including virtualized Radio Access networks (vRANs). This paper proposes a converged SBA (cSBA) to decouple access, control, and data planes. Specifically, the access plane with Distributed Units (DUs) functionalities is responsible for accessing User Equipments (UEs); the control plane consists of newly-defined converged services by integrating functionalities of Central Unit Control Planes (CU-CPs) and 5G core network for signaling processing; the data plane comprises Central Unit User Planes (CP-UPs) and User Plane Functions (UPFs), which are responsible for service data transmission. Optional converged services definition, service-based interface implementation, and cognitive service framework are proposed to enable an efficient cSBA. Finally, we evaluate the performance of the cSBA against the monolithic architecture.

关键词： Protocols 5G mobile communication Process control Computer architecture Media Access Protocol network function virtualization Synchronization

Deep Joint Source-Channel Coding Based on Semantics of Pixels for Wireless Image Transmission

学校读者我要写书评

暂无评论

Deep Joint Source-Channel Coding Based on Semantics of Pixel...

IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

作者： Qizheng Sun Caili Guo Yang Yang Rui Tang Chuanhong Liu Beijing Key Laboratory of Network System Architecture and Convergence Beijing Laboratory of Advanced Information Networks Beijing University of Posts and Telecommunications Beijing China

Current image coding methods for semantic communication typically concentrate on intelligent tasks or image reconstruction separately, and seldom consider both aspects simultaneously. To balance these two aspects during wireless image transmission, we propose a joint source-channel coding method based on the semantics of pixels (SP), which can retain both pixel information for reconstruction and semantic information for intelligent tasks. Specifically, we first design a gradient-based mechanism to quantify the semantic importance of downstream intelligent tasks on pixels. Then, we design the SP-based loss function to train the deep joint source-channel coding network. Experiment results demonstrate that the proposed method maintains reconstruction performance and improves the task performance by 1.61% and 4.06%, respectively, compared to the state-of-the-art deep joint source-channel coding method and traditional separate source-channel coding method at the same transmission rate and signal-to-noise ratio.

关键词：

Signal-SGN: A Spiking Graph Convolutional network for Skeleton Action Recognition via Learning Temporal-Frequency Dynamics

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zheng, Naichuan Du, Yuchen Xia, Hailun Liu, Dapeng Beijing Laboratory of Advanced Information Networks Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China

For skeleton-based action recognition, Graph Convolutional networks (GCNs) are effective models. Still, their reliance on floating-point computations leads to high energy consumption, limiting their applicability in battery-powered devices. While energy-efficient, Spiking Neural networks (SNNs) struggle to model skeleton dynamics, leading to suboptimal solutions. We propose Signal-SGN (Spiking Graph Convolutional network), which utilizes the temporal dimension of skeleton sequences as the spike time steps and represents features as multi-dimensional discrete stochastic signals for temporal-frequency domain feature extraction. It combines the 1D Spiking Graph Convolution (1D-SGC) module and the Frequency Spiking Convolution (FSC) module to extract features from the skeleton represented as spiking form. Additionally, the Multi-Scale Wavelet Transform Feature Fusion (MWTF) module is proposed to extract dynamic spiking features and capture frequency-specific characteristics, enhancing classification performance. Experiments across three large-scale datasets reveal Signal-SGN exceeding state-of-the-art SNN-based methods in accuracy and computational efficiency while attaining comparable performance with GCN methods and significantly reducing theoretical energy consumption. Copyright © 2024, The Authors. All rights reserved.

关键词： Musculoskeletal system

MK-SGN: A Spiking Graph Convolutional network with Multimodal Fusion and Knowledge Distillation for Skeleton-based Action Recognition

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zheng, Naichuan Xia, Hailun Liang, Zeyu Du, Yuchen Beijing Laboratory of Advanced Information Networks Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China

In recent years, multimodal Graph Convolutional networks (GCNs) have achieved remarkable performance in skeleton-based action recognition. The reliance on high-energy-consuming continuous floating-point operations inherent in GCN-based methods poses significant challenges for deployment in energy-constrained, battery-powered edge devices. To address these limitations, MK-SGN, a Spiking Graph Convolutional network with Multimodal Fusion and Knowledge Distillation, is proposed to leverage the energy efficiency of Spiking Neural networks (SNNs) for skeleton-based action recognition for the first time. By integrating the energy-saving properties of SNNs with the graph representation capabilities of GCNs, MK-SGN achieves significant reductions in energy consumption while maintaining competitive recognition accuracy. Firstly, we formulate a Spiking Multimodal Fusion (SMF) module to effectively fuse multimodal skeleton data represented as spike-form features. Secondly, we propose the Self-Attention Spiking Graph Convolution (SA-SGC) module and the Spiking Temporal Convolution (STC) module, to capture spatial relationships and temporal dynamics of spike-form features. Finally, we propose an integrated knowledge distillation strategy to transfer information from the multimodal GCN to the SGN, incorporating both intermediate-layer distillation and soft-label distillation to enhance the performance of the SGN. MK-SGN exhibits substantial advantages, surpassing state-of-the-art GCN frameworks in energy efficiency and outperforming state-of-the-art SNN frameworks in recognition accuracy. The proposed method achieves a remarkable reduction in energy consumption, exceeding 98% compared to conventional GCN-based approaches. This research establishes a robust baseline for developing high-performance, energy-efficient SNN-based models for skeleton-based action recognition Copyright © 2024, The Authors. All rights reserved.

关键词： Musculoskeletal system

Differential Electrical Conductivity Strength Based Neighborhood Compactness Model: A Novel Vulnerable Node Identification Method for Power networks

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Min, Zhao Junhan, Ye Jiayun, Li Beijing Laboratory of Advanced Information Networks Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China

The large scale and complex connections of the power networks make the vulnerability of the networks increasingly prominent. Due to the sensitivity of networks to interference, the failure of a single vulnerable node can quickly spread and bring down the system. Therefore, it is of great significance to efficiently and accurately identify the vulnerable nodes to improve the safety and management of the power networks. The traditional global information-based vulnerable node identification methods have high computational complexity and cannot meet the real-time requirements of the power networks. The existing methods based on local information only focus on the information of neighbors but lack the exploration of the interaction between neighbors. This paper focuses on the essential characteristics of the power network topology and attempts to mine the vulnerable nodes through broader and richer neighborhood information. Further, a novel vulnerable node identification method combining the neighborhood compactness model and differential electrical conductivity strength (ECS) is proposed. From the new perspective of the decentered neighborhood, we propose the kth-order neighborhood ECS to quantitatively analyze the strength of the multipath connections of the neighbors. Meanwhile, this paper defines the differential ECS to describe the closeness of the neighbors considering the distance between the central node and the paths between neighbors. And finally, the identification of vulnerable nodes is achieved by the neighborhood compactness model. Our newly proposed parameter, average ECS, as well as the traditional parameters, including network efficiency, connectivity, vulnerability, capacity, ranking distribution, and complexity, are used for the performance analysis of the algorithm. The proposed algorithm generally outperforms the benchmark methods in the above six dimensions in our comparison experiments on nine power networks. © 2024, The Authors. All rights reserv

关键词： network topology

Deep Joint Source-Channel Coding for Wireless Image Transmission with Semantic Importance

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Sun, Qizheng Guo, Caili Yang, Yang Chen, Jiujiu Tang, Rui Liu, Chuanhong Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence Beijing Laboratory of Advanced Information Networks Beijing100876 China

The sixth-generation mobile communication system proposes the vision of smart interconnection of everything, which requires accomplishing communication tasks while ensuring the performance of intelligent tasks. A joint source-channel coding method based on semantic importance is proposed, which aims at preserving semantic information during wireless image transmission and thereby boosting the performance of intelligent tasks for images at the receiver. Specifically, we first propose semantic importance weight calculation method, which is based on the gradient of intelligent task’s perception results with respect to the features. Then, we design the semantic loss function in the way of using semantic weights to weight the features. Finally, we train the deep joint source-channel coding network using the semantic loss function. Experiment results demonstrate that the proposed method achieves up to 57.7% and 9.1% improvement in terms of intelligent task’s performance compared with the source-channel separation coding method and the deep source-channel joint coding method without considering semantics at the same compression rate and signal-to-noise ratio, respectively. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantics

Importance-Aware Image Segmentation-based Semantic Communication for Autonomous Driving

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lv, Jie Tong, Haonan Pan, Qiang Zhang, Zhilong He, Xinxin Luo, Tao Yin, Changchuan School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China Beijing Key Laboratory of Network System Architecture and Convergence Beijing100876 China

This article studies the problem of image segmentation-based semantic communication in autonomous driving. In real traffic scenes, detecting the key objects (e.g., vehicles, pedestrians and obstacles) is more crucial than that of other objects to guarantee driving safety. Therefore, we propose a vehicular image segmentation-oriented semantic communication system, termed VIS-SemCom, where image segmentation features of important objects are transmitted to reduce transmission redundancy. First, to accurately extract image semantics, we develop a semantic codec based on Swin Transformer architecture, which expands the perceptual field thus improving the segmentation accuracy. Next, we propose a multi-scale semantic extraction scheme via assigning the number of Swin Transformer blocks for diverse resolution features, thus highlighting the important objects’ accuracy. Furthermore, the importance-aware loss is invoked to emphasize the important objects, and an online hard sample mining (OHEM) strategy is proposed to handle small sample issues in the dataset. Experimental results demonstrate that the proposed VIS-SemCom can achieve a coding gain of nearly 6 dB with a 60% mean intersection over union (mIoU), reduce the transmitted data amount by up to 70% with a 60% mIoU, and improve the segmentation intersection over union (IoU) of important objects by 4%, compared to traditional transmission scheme. © 2024, CC BY-NC-SA.

关键词： Autonomous vehicles

Multi-Attention Adaptive Graph Transformer network for Data Inference and Prediction in Vehicular Crowdsensing

学校读者我要写书评

暂无评论

Multi-Attention Adaptive Graph Transformer Network for Data ...

International Conference on Intelligent Transportation

作者： Shouyu Huang Luhan Wang Rui Zheng Jie Huo Xiangming Wen Zhaoming Lu Beijing Key Laboratory for Network System Architecture and Convergence Beijing University of Posts and Telecommunications Beijing China School of Traffic and Transportation Beijing Jiaotong University Beijing China

ISBN: (数字)9798331505929

ISBN: (纸本)9798331505936

Vehicular crowdsensing has emerged as an important sensing paradigm in the Internet of Things (IoT), which recruits Intelligent Vehicles to collect data. As its variant, sparse MCS collects data from some sensing areas and infer data for other unsensed areas, addressing the challenge of maintaining high-quality sensing at manageable costs. In practical vehicular crowdsensing scenarios, the diverse requirements of urban applications make it crucial not only to infer data for the current period but also to predict the entire sensing map for the future. This predictive capability not only reduces the cost of sensing but also provides critical data support for various urban applications, including intelligent transportation systems. This paper introduces the Multi-Attention Adaptive Graph Transformer network (MAAGTN) aimed at improving the accuracy of data inference and prediction in vehicular crowdsensing. We present a novel embedding method that integrates spatial-temporal information of sensing maps into the model. An adaptive learnable graph module and a multiattention module are designed to dynamically capture complex spatial-temporal correlations among data. Furthermore, a dynamic multi-task learning framework is proposed to mitigate error propagation by adjusting task weights during training. Experimental evaluations on real-world datasets demonstrate the superiority of MAAGTN for data inference and prediction.

关键词： Training Adaptive systems Costs Correlation Crowdsensing Transformers Multitasking Sensors Internet of Things Vehicle dynamics