Table of Contents | 2024 Issue 3 Special Topic: 6G Integrated Sensing and Computing
Discussion on the Application of Integrated Sensing and Computing in the Internet of Vehicles
Immersive XR Practices and Outlook Based on 6G Integrated Sensing and Computing
Key Technologies for Integrated Sensing and Computing in Three-Dimensional Transportation Systems
Exploration of Integrated Sensing and Computing Architecture and Scenario Empowerment in 6G
Wireless Resource Management under Deep Integration of Communication, Sensing, Computing, and Storage
“Mobile Communications” 2024 Issue 3
06【6G Integrated Sensing and Computing】 Special Topic
Integrated Sensing, Communication, and Computing for Edge Intelligent Networks: Architecture, Challenges, and Outlook
Qi Qiao1, Chen Xiaoming2
(1. College of Information Science and Technology, Hangzhou Normal University, Hangzhou, Zhejiang 311121;
2. College of Information and Electronic Engineering, Zhejiang University, Hangzhou, Zhejiang 310017)
【Abstract】Integrated sensing and computing serves as an empowering means for edge intelligent networks, effectively enabling real-time collection, transmission, and processing of massive data, thereby providing efficient and real-time intelligent services across various industries. Firstly, based on the characteristics of edge intelligent networks, the definitions of communication, sensing, and computing functions are clarified, and a network architecture and system architecture for integrated sensing and computing empowerment are proposed, along with typical application scenarios. Secondly, key challenges in achieving integrated sensing and computing in edge intelligent networks are analyzed, including massive data processing, complex interference coordination, and joint resource management, with corresponding solutions provided. Finally, feasible research directions for the future are discussed.
【Keywords】Edge Intelligent Networks; Integrated Sensing and Computing; Massive Data
doi:10.3969/j.issn.1006-1010.20240110-0002
Classification Number: TN929.5 Document Code: A
Article Number: 1006-1010(2024)03-0040-07
Citation Format: Qi Qiao, Chen Xiaoming. Integrated Sensing, Communication, and Computing for Edge Intelligent Networks: Architecture, Challenges, and Outlook[J]. Mobile Communications, 2024,48(3): 40-46.
QI Qiao, CHEN Xiaoming. Integrated Sensing, Communication, and Computing for Edge Intelligent Networks: Architecture, Challenges, and Outlook[J]. Mobile Communications, 2024,48(3): 40-46.
0 Introduction
In recent years, with the rapid development of mobile communication, artificial intelligence, and Internet of Things (IoT) technologies, the number of global device connections and data traffic has experienced explosive growth. It is predicted that by 2030, nearly 30 billion IoT devices will be online simultaneously[1], while mobile data traffic will exceed 5,000 EB/month[2]. In the current 5G networks, the application of artificial intelligence is mainly concentrated in the cloud to meet the growing data demands. In a centralized cloud intelligence model, a large amount of data is uploaded from terminal devices to the cloud for intelligent processing and analysis[3-4]. However, faced with the scenarios of large-scale nodes and massive data in the future, this centralized processing model faces a series of challenges, such as bandwidth pressure for communication transmission, real-time data processing, and data security and privacy issues. Based on this, the edge intelligent network, which combines edge computing and artificial intelligence, has emerged to perform real-time analysis, processing, and decision-making on the massive data collected at the network edge, close to the widely distributed network elements and terminals, thereby providing ubiquitous intelligent real-time services across various industries[5].
Driven by the vision of 6G’s “Intelligent Connectivity for All Things,” new data-driven intelligent services are emerging, including autonomous driving, immersive augmented reality (AR)/virtual reality (VR), and smart cities, which place higher demands on wireless networks[6-7]. For example, according to Google data, smart driving cars generate about 1 GB of data per second. In this context, edge intelligent networks need to achieve perception, transmission, and processing of massive information under limited spectrum, computing, and storage resources while ensuring millisecond-level latency response. However, traditional wireless network designs have a fundamental issue: communication, sensing, and computing functions are deployed independently. For instance, sensor networks focus on data collection, communication networks focus on data transmission, and cloud computing centers focus on data processing, as shown in Figure 1 (a). This separation design not only leads to inefficient utilization of system resources and performance bottlenecks but also makes it difficult to meet the high-performance demands of new intelligent applications. Therefore, edge intelligent networks urgently need a design for integrated sensing, communication, and computing, as shown in Figure 1 (b), to fully utilize network resources and coordinate communication, sensing, and computing functions, thereby providing efficient services with real-time capabilities for intelligent applications.
Given the scientific value and application prospects of integrated sensing and computing, the industry has shown great interest in it. In the latest developments in this field, China Mobile first proposed the integrated sensing and computing framework for 6G networks at the World 5G Conference in 2020[8]. In 2021, Yan Shi and others from Beijing University of Posts and Telecommunications[9] proposed the theory and vision of integrated sensing and computing, emphasizing the key role of integrated sensing and computing technology in supporting the intelligent interconnection of humans, machines, and things in 6G networks. Subsequently, the China Communications Society launched the “Frontier Report on Integrated Sensing and Computing Networks” in 2022[10], clearly stating that integrated sensing and computing is an important cornerstone for realizing the interaction and deep integration of the physical and digital worlds. In the same year, the China Computing Power Conference held a sub-forum on the innovative development of integrated sensing and computing and jointly established an integrated sensing and computing innovation laboratory. At the 2023 Global 6G Technology Conference, experts discussed the key technologies and potential issues and challenges in achieving integrated sensing and computing. In addition, academia has also explored integrated sensing and computing in specific scenarios, such as extended reality[11], the Internet of Vehicles[12], drone networks[13], and satellite communication networks[14]. Although integrated sensing and computing has attracted widespread attention, it is still in the early stages of exploration and development. Current research on integrated sensing and computing mainly focuses on the coexistence of communication, sensing, and computing parts or the design of functional integration, such as integrated sensing and communication[15-16], integrated sensing and computing[17-18], and integrated communication and computing[19-20]. This preliminary exploration has provided academia and industry with an initial understanding of the potential and future development directions of integrated sensing and computing. However, in practical applications, achieving deep integration of sensing, communication, and computing still faces multiple challenges, necessitating further in-depth research and practical exploration. Based on this, this paper aims to propose an integrated sensing and computing architecture for edge intelligent networks based on their characteristics, discuss its application scenarios, key challenges, and solutions, and point out future research and development directions.
1 Integrated Sensing and Computing Empowered Edge Intelligent Network Architecture
In edge intelligent networks, communication, sensing, and computing are three key functions that are interdependent and mutually reinforcing, forming the foundation for network intelligence, real-time capabilities, and efficiency. Specifically, communication refers to cellular mobile communication that promotes information exchange and functional collaboration between nodes in the edge intelligent network through real-time connections and efficient data transmission. Sensing is an important means for edge intelligent networks to acquire environmental and status information, where endogenous sensing (e.g., perceiving network status, service characteristics, terminal capabilities through information interaction) obtains detailed information from within the network, while external sensing (e.g., collecting surrounding environmental information through sensors or achieving target detection, positioning, tracking through radio frequency signals) provides important data about the external environment of the network. Computing, as the core link for processing and analyzing data, plays an indispensable role in edge intelligent networks. Pushing computing functions to the network edge allows for real-time processing and analysis of acquired data, extracting useful information and making corresponding decisions, enabling edge intelligent networks to better adapt to complex and dynamic environmental changes. By efficiently integrating these three functions, edge intelligent networks can provide faster, more efficient, flexible, and secure services, promoting the development of IoT, industrial IoT, and the Internet of Vehicles. The following sections will detail the network architecture and system architecture empowered by integrated sensing and computing, as well as typical application scenarios.
1.1 Network Architecture
As shown in Figure 2, the edge intelligent network empowered by integrated sensing and computing consists of an edge control center, integrated network elements, and edge nodes, which work together to achieve efficient operation and intelligent decision-making of the edge intelligent network, providing support for edge intelligent applications.
(1) Edge Control Center: The edge control center, as the core part, is typically located in data centers or server clusters at the network edge. Its functions include data analysis and decision-making, computing power scheduling and resource management, as well as task issuance and service orchestration. By managing the status and information of numerous edge nodes and integrated network elements in the edge intelligent network, it enables monitoring, scheduling, and management of the edge intelligent network to provide efficient and reliable edge intelligent services.
(2) Integrated Network Elements: Integrated network elements are devices that integrate communication, sensing, computing, and storage functions, mainly consisting of integrated sensing and computing base stations and edge servers. They are usually deployed at key locations in the edge network to process large-scale data and complex algorithms. Through data interaction and functional collaboration with the edge control center and edge nodes, they can support intelligent applications.
(3) Edge Nodes: Edge nodes are typically deployed close to data sources, such as smart terminal devices, sensors, or IoT nodes. Edge nodes not only have communication functions but also integrate some or all of the sensing and computing functions. By deploying edge nodes, the edge intelligent network can achieve real-time processing of data and rapid response, reducing data transmission delays and bandwidth requirements. Additionally, the multifunctionality of edge nodes allows edge intelligent applications to operate more flexibly and efficiently, adapting to changes in different scenarios and requirements.
1.2 System Architecture
As shown in Figure 3, the system architecture of integrated sensing and computing in edge intelligent networks consists of a resource layer, capability layer, and application layer. The collaborative work between different layers enables the system to adapt to various application scenarios and requirements, providing customized services.
(1) Resource Layer: The resource layer is the foundational layer of the integrated sensing and computing system, including infrastructure, signal processing, wireless, and hardware/software resources. The resource layer provides the necessary infrastructure and resource support for the integrated sensing and computing system, offering an execution environment and available resources for the upper capability layer and application layer.
(2) Capability Layer: The capability layer is built on top of the resource layer, mainly providing data processing and analysis capabilities. It includes functions such as data processing, data transmission, sensing detection, target tracking, algorithm execution, and model training. The capability layer can utilize the communication, sensing, and computing resources provided by the resource layer to execute corresponding data processing and model algorithms, providing valuable services and functions for users and the application layer.
(3) Application Layer: The application layer is the top layer of the integrated sensing and computing system, used to implement specific application scenarios and functions. It includes various edge intelligent applications, such as smart transportation, smart factories, and smart cities. The application layer utilizes the data processing capabilities and algorithm models provided by the capability layer to achieve real-time analysis and intelligent decision-making of data, thereby meeting the high-performance demands of intelligent applications.
1.3 Application Scenarios
The edge intelligent network based on integrated sensing and computing has a wide range of application scenarios. Below are some typical application scenarios.
(1) Smart Transportation: In smart transportation systems, integrated sensing and computing can achieve real-time traffic monitoring, intelligent traffic signal control, and smart route planning. For example, by using various traffic sensing devices on the road, real-time collection of traffic flow, speed, and other parameters can be achieved for comprehensive traffic monitoring. The collected real-time flow data can be used for machine learning model training at edge nodes to predict potential congestion areas. Based on the congestion prediction results, intelligent control adjustments of traffic signals at intersections can be made to reduce vehicle waiting times and alleviate traffic congestion. At the same time, real-time environmental sensing can provide optimal route selection based on real-time traffic for buses, taxis, etc., thereby improving operational efficiency.
(2) Low-altitude Security: With the rapid development of drone technology, low-altitude safety and monitoring have become increasingly important. In the field of low-altitude security, integrated sensing and computing can achieve real-time environmental monitoring and early warning. By using various sensing modalities, such as radar, cameras, and sensors, comprehensive perception of low-altitude areas can be achieved. At the same time, intelligent algorithms on edge nodes can perform real-time target detection, tracking, and recognition to identify potential threats or abnormal behaviors. Through information interaction and functional collaboration, edge nodes can timely transmit the processed sensing results to central control systems or other related devices to support timely decision-making and response.
(3) Smart Industry: In industrial IoT environments, integrated sensing and computing can be used to achieve real-time monitoring of production lines and equipment, predictive maintenance, and refined production scheduling. By deploying various multifunctional edge nodes at the factory edge, real-time monitoring of the operational status of various industrial devices and production line processes can be achieved. Edge nodes can collect and analyze operational data in real-time to predict potential failure risks. Additionally, based on the operational status of equipment and production capacity requirements, the control center can make refined scheduling arrangements for production tasks, optimizing production management.
2 Key Challenges and Solutions for Integrated Sensing and Computing
Implementing integrated sensing and computing in edge intelligent networks is not an easy task, as it faces several significant challenges. These challenges mainly include massive data processing, complex interference coordination, and joint resource management. Notably, these challenges cannot be solved using traditional methods and require the design of corresponding solutions tailored to the characteristics of edge intelligent networks. Below, these key challenges and their feasible solutions are introduced one by one.
2.1 Massive Data Processing
The scale of sensing data collected by large-scale edge nodes is enormous and heterogeneous, making it crucial to process this massive data in real-time and effectively within edge intelligent networks. Traditional centralized data processing methods consume a large amount of bandwidth resources, leading to network congestion and delays, which cannot meet real-time requirements. Furthermore, since raw data needs to be uniformly uploaded to servers for processing, centralized data processing methods also fail to protect user data privacy and security. To address this issue, edge federated learning based on aerial computing provides a feasible solution.
Edge federated learning is a distributed machine learning method with privacy protection features that allows edge nodes to locally train models on datasets and iteratively exchange updated model parameters with global model parameters on edge servers via wireless links[21]. To further improve spectrum utilization and reduce model parameter aggregation latency, aerial computing technology based on communication and computing integration can be effectively applied in edge federated learning. As shown in Figure 4, it utilizes the natural superposition characteristics of wireless multiple access channels to achieve rapid aggregation of model parameters on edge servers through concurrent transmission of model parameters[22].
In other words, the aerial computing-based edge federated learning technology transforms the traditional data processing model of “perceive first, then communicate, and finally compute” into a new data processing model of “integrated sensing, communication, and computing,” effectively reducing data transmission latency while improving information perception and model computation efficiency.
2.2 Complex Interference Coordination
Sharing wireless and hardware/software resources to achieve integrated sensing and computing can lead to various complex interferences. Effectively coordinating these interferences in edge intelligent networks to improve system performance is a highly challenging task. Specifically, simultaneously transmitting communication, sensing, and computing signals under the same spectrum resources can lead to severe self-interference and mutual interference, thereby degrading the performance of the integrated system. In this case, full-duplex technology, massive MIMO (Multiple-Input Multiple-Output), beamforming technology, and integrated waveform design provide technical pathways to address complex interference coordination issues.
Full-duplex technology allows devices to simultaneously transmit and receive on the same spectrum resources, effectively solving the antenna self-interference problem in integrated sensing and computing systems[23]. Massive MIMO technology utilizes the spatial degrees of freedom between base stations and terminal devices with multiple antennas to transmit and receive signals, enhancing capacity and coverage. Beamforming technology adjusts the weights and phases of antenna arrays to form directional beams, enhancing the reception strength of target signals. By employing spatial beamforming technology based on massive antenna arrays, user-to-user interference and functional mutual interference in integrated sensing and computing systems can be effectively reduced[24]. Additionally, by comprehensively considering the functional characteristics and target requirements of different signals, analyzing the coupling relationships between communication, sensing, and computing signals, and designing compatible integrated sensing and computing waveforms, mutual interference between system functions can also be effectively reduced[25]. As shown in Figure 5, the comprehensive application of full-duplex technology, massive MIMO, beamforming technology, and integrated waveform design can effectively coordinate and manage various complex interferences in edge intelligent networks, thereby improving the overall performance and efficiency of the system.
2.3 Joint Resource Management
Different application scenarios present differentiated performance requirements for communication, sensing, and computing, making joint resource management for integrated sensing and computing a significant challenge in edge intelligent networks. To support various new intelligent services, effective management and allocation of multidimensional resources are required. However, the complex relationships of competition, collaboration, and coupling among communication, sensing, and computing services make resource allocation more challenging. To address this challenge, intelligent computing power scheduling and adaptive resource management can be employed to effectively manage multidimensional resources for integrated sensing and computing to meet different application needs.
Through intelligent computing power scheduling, computing resources can be dynamically scheduled based on the characteristics and priorities of tasks, improving data processing efficiency and task execution speed[26]. Adaptive resource management can jointly allocate and manage resources based on system demands and the availability of spectrum resources, achieving a balance in integrated sensing and computing performance and maximizing resource utilization[27]. Intelligent computing power scheduling and adaptive resource management schemes can utilize optimization methods such as machine learning, heuristic algorithms, and game theory to adaptively achieve multidimensional resource allocation and management.As shown in Figure 6, through the combined effects of intelligent computing power scheduling and adaptive resource management, the efficiency, performance, and flexibility of edge intelligent networks can be effectively improved, thereby supporting the deployment and operation of various new intelligent services.
3 Future Research Directions
Although integrated sensing and computing has received widespread attention and research both domestically and internationally, many urgent issues remain to be addressed. Based on this, future research can focus on the following directions.
3.1 Integrated Sensing and Computing in High-Dynamic Environments
Edge intelligent networks face some unique challenges in high-dynamic environments. Specifically, in high-speed moving scenarios, channels change rapidly and are significantly affected by multipath effects and Doppler effects, leading to difficulties in real-time acquisition of environmental information, limited data transmission rates, and deviations in decision-making. In terms of sensing, high-speed sampling capabilities are required to quickly and accurately perceive environmental changes and provide reliable sensing data in a timely manner. For communication, adaptive modulation and coding techniques need to be employed to enhance signal reliability and transmission efficiency. In terms of computing, efficient algorithms and models need to be designed to balance real-time and performance requirements to adapt to rapidly changing scenarios, achieving real-time data processing and decision-making. Addressing the challenges of edge intelligent networks in high-dynamic environments requires further research and innovation in the design of integrated sensing and computing to enhance adaptability, performance, and reliability, achieving intelligent perception and decision-making capabilities in high-dynamic environments.
3.2 Information Security Mechanisms for Integrated Sensing and Computing
Integrated sensing and computing also faces some challenges in information security. In edge intelligent networks, the process of integrated sensing and computing involves collecting and forwarding data from multiple distributed edge nodes, followed by joint processing and decision-making. This mode of data interaction and functional collaboration brings about challenges in information security, making traditional information security mechanisms often inadequate to cope with the scale and complexity of data in edge intelligent networks. Firstly, some edge nodes in the network have simple structures, making their data vulnerable to eavesdropping, attacks, or tampering, which can lead to penetration and destruction of the entire network, highlighting the shortcomings in information security. Secondly, edge nodes may collect users’ private data, such as location information and personal health data. Ensuring the privacy and confidentiality of this data during the integration process is crucial. Therefore, integrated sensing and computing for edge intelligent networks must fully consider information security issues and adopt appropriate security mechanisms to protect data privacy, ensure data integrity, safeguard data security, and detect and prevent malicious attacks, thereby enhancing the security and trustworthiness of integrated sensing and computing systems.
3.3 Hardware Implementation of Integrated Sensing and Computing
In terms of hardware design, integrating communication, sensing, and computing functions into edge devices faces multiple challenges, including resource limitations, conflicting demands, cost and size constraints, and complexity. It is not simply a matter of introducing sensing and computing modules into traditional communication devices. For example, sensing devices require high-precision and high-resolution sensors, while computing devices need high-performance processors and memory. Under limited hardware resources and costs, these different functional demands conflict with each other. At the same time, integrating communication, sensing, and computing functions into edge devices brings significant complexity, involving challenges in collaboration between different modules, data transmission and processing, and algorithm optimization. Therefore, how to balance these functional demands within limited cost and size constraints, effectively achieve collaboration between different functional modules, and design hardware devices that match communication, sensing, and computing capabilities and requirements is a key issue for achieving integrated sensing and computing.
4 Conclusion
Integrated sensing and computing provides a new approach for enhancing resource utilization and system performance in edge intelligent networks. However, implementing integrated sensing and computing in edge intelligent networks still faces numerous challenges, such as massive data processing, complex interference coordination, and joint resource management. This paper proposes some solutions but further research and refinement are needed. Based on this, future research directions include but are not limited to system design in high-dynamic environments, information security mechanisms, and hardware implementation. Overall, integrated sensing and computing is a complex and promising research field with vast research space yet to be explored. Academia and industry need to work closely together to promote the exploration and validation of related technologies, continuously improve and optimize the technical framework of integrated sensing and computing, and jointly contribute to realizing the beautiful vision of 6G’s intelligent connectivity for all things.
References: (Scroll to browse)
[1] IMT-2030 (6G) Promotion Group. White Paper on the Overall Vision and Potential Key Technologies of 6G[R]. 2021.
[2] Statista. Number of IoT connected devices worldwide 2019-2021, with forecasts to 2030[R]. 2022.
[3] Wang W, Kumar N, Chen J, et al. Realizing the potential of the internet of things for smart tourism with 5G and AI[J]. IEEE Network, 2020, 34(6): 295-301.
[4] You X, Zhang C, Tan X, et al. AI for 5G: research directions and paradigms[J]. Science China Information Sciences, 2019,62 : 1-13.
[5] Deng S, Zhao H, Fang W, et al. Edge intelligence: The confluence of edge computing and artificial intelligence[J]. IEEE Internet of Things Journal, 2020,7(8) : 7457-7469.
[6] Saad W, Bennis M, Chen M. A vision of 6G wireless systems: Applications, trends, technologies, and open research problems[J]. IEEE Network, 2019,34(3) : 134-142.
[7] Liu M, Fang S, Dong H, et al. Review of digital twin about concepts, technologies, and industrial applications[J]. Journal of Manufacturing Systems, 2021,58 : 346-361.
[8] World 5G Conference. Integration of sensing, communication and computing toward 6G[R]. 2020
[9] Yan Shi, Peng Mugeng, Wang Wenbo. Integrated Sensing, Communication, and Computing: 6G Vision and Key Technologies[J]. Journal of Beijing University of Posts and Telecommunications, 2021,44(4): 1-10.
[10] China Communications Society. Frontier Report on Integrated Sensing and Computing Networks[R]. 2022.
[11] Wang Wenbo, Hou Yuefeng, Tian Feng. Integrated Video Transmission and QoE Evaluation for Extended Reality[J]. Mobile Communications, 2022,46(5): 84-88.
[12] Ma Zhonggui, Li Zhuo, Liang Yanpeng. Overview and Outlook of Integrated Sensing and Computing in Autonomous Driving Vehicle Networks[J]. Journal of Engineering Science, 2023,45(1): 137-149.
[13] Liao Jiajia, Xiang Luping, Hu Jie, et al. Overview of 6G Integrated Sensing and Computing Based on Drones[J]. Mobile Communications, 2023,47(3): 92-100.
[14] Jing Yi, Jiang Chunxiao, Zhan Yafeng. Integrated Sensing and Computing Architecture, Technologies, and Challenges for Satellite Communication in 6G[J]. Wireless Communication Technology, 2023,49(1) : 12-20.
[15] Cui Y, Liu F, Jing X, et al. Integrating sensing and communications for ubiquitous IoT: Applications, trends, and challenges[J]. IEEE Network, 2021,35(5) : 158-167.
[16] Liu A, Huang Z, Li M, et al. A survey on fundamental limits of integrated sensing and communication[J]. IEEE Communications Surveys & Tutorials, 2022,24(2): 994-1034.
[17] Hu L, Miao Y, Wu G, et al. iRobot-Factory: An intelligent robot factory based on cognitive manufacturing and edge computing[J]. Future Generation Computer Systems, 2019,90 : 569-577.
[18] Nie L, Ning Z, Wang X, et al. Data-driven intrusion detection for intelligent internet of vehicles: A deep convolutional neural network-based method[J]. IEEE Transactions on Network Science and Engineering, 2020,7(4) : 2219-2230.
[19] Wen D, Zhu G, Huang K. Reduced-dimension design of MIMO over-the-air computing for data aggregation in clustered IoT networks[J]. IEEE Transactions on Wireless Communications, 2019,18(11) : 5255-5268.
[20] Dai Y, Xu D, Maharjan S, et al. Joint computation offloading and user association in multi-task mobile edge computing[J]. IEEE Transactions on Vehicular Technology, 2018,67(12) : 12313-12325.
[21] Niknam S, Dhillon H S, Reed J H. Federated learning for wireless communications: Motivation, opportunities, and challenges[J]. IEEE Communications Magazine, 2020,58(6): 46-51.
[22] Qi Q, Chen X. Robust design of federated learning for edge-intelligent networks[J]. IEEE Transactions on Communications, 2022,70(7): 4469-4481.
[23] Zhang Z, Chai X, Long K, et al. Full duplex techniques for 5G networks: self-interference cancellation, protocol design, and relay selection[J]. IEEE Communications Magazine, 2015, 53(5) : 128-137.
[24] Qi Q, Chen X, Zhong C, et al. Integrated sensing, computation and communication in B5G cellular Internet of Things[J]. IEEE Transactions on Wireless Communications, 2020,20(1): 332-344.
[25] Liu F, Zhou L, Masouros C, et al. Toward dual-functional radar-communication systems: Optimal waveform design[J]. IEEE Transactions on Signal Processing, 2018,66(16) : 4264- 4279.
[26] Zhou T, Tang D, Zhu H, et al. Reinforcement learning with composite rewards for production scheduling in a smart factory[J]. IEEE Access, 2020,9 : 752-766.
[27] Qi Q, Chen X, Khalili A, et al. Integrating sensing, computing, and communication in 6G wireless networks: Design and optimization[J]. IEEE Transactions on Communications, 2022,70(9): 6212-6227. ★
Scan the QR code to read and download this paper on Zhihu
★Original article published in “Mobile Communications”2024 Issue 3★
doi:10.3969/j.issn.1006-1010.20240110-0002
Classification Number: TN929.5 Document Code: A
Article Number: 1006-1010(2024)03-0040-07
Citation Format: Qi Qiao, Chen Xiaoming. Integrated Sensing, Communication, and Computing for Edge Intelligent Networks: Architecture, Challenges, and Outlook[J]. Mobile Communications, 2024,48(3): 40-46.
QI Qiao, CHEN Xiaoming. Integrated Sensing, Communication, and Computing for Edge Intelligent Networks: Architecture, Challenges, and Outlook[J]. Mobile Communications, 2024,48(3): 40-46.
Author InformationQi Qiao: PhD in Engineering from Zhejiang University, currently a lecturer at the College of Information Science and Technology, Hangzhou Normal University, with research interests in cellular IoT, edge intelligence, and integrated sensing and computing.Chen Xiaoming: PhD, currently a researcher and doctoral supervisor at the College of Information and Electronic Engineering, Zhejiang University, with research interests including key technologies for 5G/6G, low-orbit satellite constellation communication, and intelligent communication.
Submission method for “Mobile Communications” is online submission
Please log in to the web submission system
Link address:http://ydtx.cbpt.cnki.net
Highlights【2024 Special Call for Papers–Issue 7】6G Endogenous Intelligent Technology【2024 Special Call for Papers–Issue 8】Integration of 6G and AI【2024 Special Call for Papers–Issue 9】Integrated Networks of Air, Land, and SeaTable of Contents | 2024 Issue 4 Special Topic: RIS-Assisted Integrated Sensing and ComputingTable of Contents | 2024 Issue 3 Special Topic:6G Integrated Sensing and ComputingTable of Contents | 2024 Issue 2 Special Topic: Semantic CommunicationTable of Contents | 2024 Issue 1 Special Topic: Technologies for 6G Satellite-Ground Integrated Networks
#Scan to follow us#“Mobile Communications”Interpreting Communication through Papers
“Mobile Communications” magazine is supervised by China Electronics Technology Group Corporation and hosted by the Seventh Research Institute of China Electronics Technology Group Corporation. It is a “Double Effect Journal” in the Chinese journal matrix, a high-quality electronic journal of the Ministry of Industry and Information Technology, a source journal for Chinese scientific and technological papers, and included in the high-quality scientific journal grading directory of the China Communications Society and the China Electronics Society. Domestic continuous publication number: CN44-1301/TN, International continuous publication number: ISSN1006-1010, Postal distribution code: 46-181.