PaperlayerView exact revision on arXiv

Source-linked AI summary

Joint Communication, Sensing and Computation enabled 6G Intelligent Machine System

Zhiyong Feng, Zhiqing Wei, Xu Chen, Heng Yang, Qixun Zhang, Ping Zhang

arXiv:2302.10416v1cs.NIeess.SP

TL;DR

Separately designed communication, sensing, and computation functions can create inconsistency, protocol translation overhead, and interaction costs, making close-loop control difficult; 6G intelligent machines require accurate sensing and distributed collaborative computing. The paper proposes a unified JCSC framework combining joint communication and sensing, JCSC networking, intelligent computation, and hierarchical cloud-edge-terminal decision-making. With 30 neighbor nodes, sensing-enhanced RL-CRA reduces neighbor-discovery delay by 31.7%, while the JCSC network reduces average data-frame transmission delay by around 50%.

  • Problem

    Separately designed communication, sensing, and computation functions can create inconsistency, protocol translation overhead, and interaction costs, making close-loop control difficult; 6G intelligent machines require accurate sensing and distributed collaborative computing.

  • Method

    The paper proposes a unified JCSC framework combining joint communication and sensing, JCSC networking, intelligent computation, and hierarchical cloud-edge-terminal decision-making.

  • Results

    With 30 neighbor nodes, sensing-enhanced RL-CRA reduces neighbor-discovery delay by 31.7%, while the JCSC network reduces average data-frame transmission delay by around 50%.

  • Takeaways & Limitations

    The proposed framework and evaluated technologies verify the feasibility of using integrated communication, sensing, and computation for 6G intelligent machine systems.

  • Takeaways & Limitations

    The relationship, theoretical bounds, and performance trade-offs among communication, sensing, and computing remain fundamental problems for implementation.

Abstract

from arXiv · show

With the rapid development of the smart city, high-level autonomous driving, intelligent manufacturing, and etc., the stringent industrial-level requirements of the extremely low latency and high reliability for communication and new trends for sub-centimeter sensing have transcended the abilities of 5G and call for the development of 6G. Based on analyzing the function design of the communication, sensing and the emerging intelligent computation systems, we propose the joint communication, sensing and computation (JCSC) framework for 6G intelligent machine-type communication (IMTC) network to realize low latency and high reliability of communication, highly accurate sensing and fast environment adaption. In the proposed JCSC framework, the communication, sensing and computation abilities cooperate to benefit each other by utilizing the unified hardware, resource and protocol design. Sensing information is exploited as priori information to enhance the reliability and latency performance of wireless communication and to optimize the resource utilization of the communication network, which further improves the distributed computation and cooperative sensing ability. We propose the promising enabling technologies such as joint communication and sensing (JCS) technique, JCSC wireless networking techniques and intelligent computation techniques. We also summarize the challenges to achieve the JCSC framework. Then, we introduce the intelligent flexible manufacturing as a typical use case of the IMTC with JCSC framework, where the enabling technologies are deployed. Finally, we present the simulation results to prove the feasibility of the JCSC framework by evaluating the JCS waveform, the JCSC enabled neighbor discovery (ND) and medium access control (MAC).

I. INTRODUCTION

6G intelligent machine-type communication requires real-time sensing, autonomous cooperation, distributed computation, and close-loop control beyond current 5G capabilities. The paper motivates integrating communication, sensing, and computation to address these requirements.

  • I. INTRODUCTION: 6G intelligent machines require real-time sensing, autonomous cooperation, distributed collaborative computing, and close-loop control in dynamic environments.The close-loop flow includes sensing-data acquisition and sharing, intelligent processing and decision-making, control-instruction transmission, and execution.
  • I. INTRODUCTION: Current 5G systems cannot adequately support accurate sensing, close-loop control information flow, and distributed collaborative computing.The paper attributes this limitation to the separate design of communication, sensing, and computation functions.
  • I. INTRODUCTION: Separate function designs create protocol translation overhead, time-energy consumption, and inconsistent optimization across communication, sensing, and computation.These effects make close-loop control performance difficult to guarantee.
  • I. INTRODUCTION: Separating sensing and communication degrades space-time-frequency resource scheduling and limits efficient cooperative sensing among multiple nodes.
  • I. INTRODUCTION: The paper proposes JCSC to support low-latency, high-reliability communication, accurate sensing, and collaborative computation for 6G intelligent machine networks.The framework is presented as a response to the limitations of traditional architectures.

II. JCSC FRAMEWORK

The JCSC framework unifies communication, sensing, and computation so their information and resources can mutually improve close-loop industrial control, sensing, and computation. It combines unified protocols with hierarchical cloud-edge-terminal decision-making and customized uplink/downlink optimization.

  • II. JCSC FRAMEWORK: JCSC jointly optimizes communication, sensing, and computation to improve close-loop control, sensing performance, and computing efficiency.It also supports customized uplink and downlink optimization because sensing-data upload and instruction transmission have different requirements.
  • II. JCSC FRAMEWORK: Sensing information provides prior information for communication, while communication supports distributed computation and cooperative sensing.Intelligent computation reciprocally enhances sensing and communication performance.
  • II. JCSC FRAMEWORK: The framework treats sensing, communication, and intelligent information processing as one process for close-loop industrial optimization and fast environment adaptation.
  • II. JCSC FRAMEWORK: A cloud-edge-terminal hierarchy assigns global control to the cloud, local control to the edge, and autonomous sensing and decision-making to terminals.This organization supports cloud and edge computing, collaborative sensing, and close-loop optimization.
  • II. JCSC FRAMEWORK: Using a unified protocol for sensing information and communication data reduces protocol interaction overhead and lets location and motion information aid beam alignment, channel estimation, and MAC.

III. ENABLING TECHNOLOGIES

The paper identifies three mutually reinforcing enabling-technology directions for JCSC: joint communication and sensing, JCSC networking, and intelligent collaborative computation.

  • III. ENABLING TECHNOLOGIES: JCS provides prior sensing information for communication networking, while networking techniques support intelligent collaborative computation in complex IMTC scenarios.
  • III. ENABLING TECHNOLOGIES: Intelligent computation reciprocally improves JCS and JCSC networking through intelligent optimization and decision-making.
  • III. ENABLING TECHNOLOGIES: The three enabling technologies are intended to support accurate sensing and close-loop control information flow in complex intelligent machine networks.

A. Joint Communication and Sensing Technique

The paper focuses on firmly integrated joint communication and sensing that shares hardware, spectrum, time, spatial resources, and waveform. It contrasts this design with separate-resource coexistence and discusses sensing-centric and communication-centric waveform paradigms.

  • A. Joint Communication and Sensing Technique: JCS can either coexist communication and sensing with separate resources or firmly integrate them within one system.Separate resources avoid mutual interference, whereas integrated operation uses shared resources.
  • A. Joint Communication and Sensing Technique: Integrated JCS shares hardware, spectrum, time slot, spatial resource, and waveform to obtain sensing from communication-signal reflections.This second class is the article’s focus and is intended to improve energy and spectrum utilization.
  • A. Joint Communication and Sensing Technique: Sensing-centric waveforms maximize sensing with low communication performance, while communication-centric waveforms preserve communication performance with weakened sensing accuracy.

2) JCS Intelligent Signal Processing:

The paper proposes intelligent processing for joint communication and sensing signals, while identifying implementation and theoretical challenges for JCS systems.

  • 2) JCS Intelligent Signal Processing:: AI-based processing can combine linear transforms and nonlinear signal classification to obtain communication and sensing information simultaneously.The approach builds on shared processing structures in OFDM communication, FMCW, and OFDM radar signal processing.
  • 2) JCS Intelligent Signal Processing:: Correlations between communication CSI and sensing features such as range and Doppler may reduce the computation load and processing delay of acquiring both types of information.
  • 2) JCS Intelligent Signal Processing:: JCS implementation faces challenges in hardware, signal processing, information theory, and performance metrics.
  • 2) JCS Intelligent Signal Processing:: High equipment density and complex device functions intensify spectrum congestion and interference among JCS devices, motivating intelligent interference mitigation.
  • 2) JCS Intelligent Signal Processing:: Highly dynamic environments require rapidly adaptive JCS waveforms and a close-loop information-flow theory with corresponding performance metrics.Accurate, frequently refreshed sensing information is needed to provide better priori information for waveform adaptation.

B. JCSC Wireless Networking Techniques

JCSC wireless networking uses sensing information from JCS-enabled handshake frames and edge devices to accelerate neighbor discovery and improve cooperative networking decisions.

  • B. JCSC Wireless Networking Techniques: JCSC environment awareness improves neighbor discovery, MAC, routing, and resource allocation to support fast cooperation among IM nodes.
  • B. JCSC Wireless Networking Techniques: JCS sensing enables neighbor discovery handshake frames to obtain neighbor positions in an IM's beam direction.
  • B. JCSC Wireless Networking Techniques: Sensing-based neighbor information reduces handshaking overload and network delay, while reinforcement learning improves successful neighbor-discovery probability by adapting detection-direction selection.
  • B. JCSC Wireless Networking Techniques: Wireless and physical-environment sensing helps IMs select spectrum and spatial resource blocks more efficiently, increasing successful channel access rate.Communication and sensing share resources, and deep reinforcement learning can control wireless access for JCS operation.
  • B. JCSC Wireless Networking Techniques: Location, motion state, and node-density information provides priori information for relay selection and sensing-refresh-rate adjustment.

4) JCSC Enabled Resource Allocation Method:

The proposed resource-allocation direction uses JCS-derived predictions and must address heterogeneous, rapidly changing networks, routing identity, spectrum diversity, and optimization complexity.

  • 4) JCSC Enabled Resource Allocation Method:: CUs can predict future service changes, spectrum occupancy, and IM access probabilities from motion trajectories and spectrum-use histories, enabling efficient pre-allocation of transmit power and spectra.The stated goal is to reduce access delay through predictive resource allocation.
  • 4) JCSC Enabled Resource Allocation Method:: Optimizing the ND handshake-frame structure is challenging because it affects cross-correlation ranging performance and positioning accuracy.
  • 4) JCSC Enabled Resource Allocation Method:: Rapidly changing topology and heterogeneity in functions, resources, and access methods make efficient JCSC MAC frames and intelligent random access mechanisms challenging.
  • 4) JCSC Enabled Resource Allocation Method:: Routing requires reliable association between an IM's MAC address and its sensing data to confirm relay-node identity.
  • 4) JCSC Enabled Resource Allocation Method:: Different propagation properties across wider frequency bands complicate intelligent spectrum allocation between sensing and communication according to environmental conditions.
  • 4) JCSC Enabled Resource Allocation Method:: Because JCSC networks combine centralized and distributed organizations, low-complexity optimization algorithms are needed to balance sensing performance, communication latency, and energy consumption.

C. Intelligent Computation Enhanced Communication and Sensing Techniques

Intelligent computation in the JCSC framework fuses distributed sensing information and uses learned environmental context to enhance communication and sensing across an intelligent flexible manufacturing architecture.

  • C. Intelligent Computation Enhanced Communication and Sensing Techniques: Distributed and centralized computation can fuse and match sensing data to improve sensing vision and environment-sensing accuracy, then enhance sensing and communication resource pre-allocation.
  • C. Intelligent Computation Enhanced Communication and Sensing Techniques: IMs can share locally processed information from multi-source sensing data, fuse it at central IMs or cloud servers, and use federated learning to train distributed models.
  • C. Intelligent Computation Enhanced Communication and Sensing Techniques: Position and velocity information can support deep-learning channel estimation and prediction, improving beamforming performance.
  • C. Intelligent Computation Enhanced Communication and Sensing Techniques: Reinforcement learning can assist beam alignment, while sensed position information supports SINR prediction, power allocation, and beam-user association.
  • C. Intelligent Computation Enhanced Communication and Sensing Techniques: In the flexible production scenario, JCS is implemented at base stations and IMs, sensors form a network, IMs perform distributed computation and sensing, and clouds integrate information and control the network.

3) JCSC Cloud-Edge-Terminal Collaborative Hierarchical Decision:

The JCSC framework coordinates sensing, communication, and computation across terminals, edge servers, and cloud systems for intelligent machine cooperation. Its intelligent flexible manufacturing use case combines distributed sensing, unified resource management, and collaborative computing.

  • 3) JCSC Cloud-Edge-Terminal Collaborative Hierarchical Decision:: JCSC distributes sensing and computation across intelligent machines, edge servers, and cloud systems for multi-level sensing fusion and autonomous decisions.Machines preprocess sensing data, while MEC servers perform multi-level fusion and local or global decision-making.
  • 3) JCSC Cloud-Edge-Terminal Collaborative Hierarchical Decision:: The framework remains challenged by dynamic channels, imperfect pre-convergence sensing predictions, limited edge computation, and the need for efficient fusion with low decision delay.These constraints require adaptive preprocessing and learning strategies that preserve detection accuracy and reliability.
  • 3) JCSC Cloud-Edge-Terminal Collaborative Hierarchical Decision:: Intelligent flexible manufacturing uses autonomous vehicles and robots that require highly reliable, low-latency control, accurate sensing, and efficient computation.These capabilities support autonomous precision production through cooperation among intelligent machines.
  • 3) JCSC Cloud-Edge-Terminal Collaborative Hierarchical Decision:: JCSC virtualizes radio, computation, and hardware resources into an integrated pool that jointly manages communication, sensing, and computation resources.The pool stores information about available spectrum, beams, and slots for close-loop information interaction and optimization.
  • 3) JCSC Cloud-Edge-Terminal Collaborative Hierarchical Decision:: Sensing information provides prior information for beam alignment, neighbor discovery, MAC, routing, and resource allocation, improving communication reliability, throughput, and latency.Improved communication performance further supports collaborative sensing and computing.

V. EVALUATION

The evaluation examines the JCS waveform, neighbor discovery, and MAC enabled by JCSC. Results report improved waveform BER and RMSE, 31.7% lower neighbor-discovery delay, and roughly 50% lower average data-frame transmission delay.

  • V. EVALUATION: The proposed CD-OFDM JCS waveform achieves better BER and RMSE than the conventional OFDM JCS waveform.The stated explanation is that CDM gain improves communication demodulation reliability and the echo-signal-to-interference-plus-noise ratio.
  • V. EVALUATION: Prior sensing information accelerates the JCSC neighbor-discovery process by supporting the sensing-enhanced RL-CRA algorithm.The evaluation compares RL-CRA with conventional complete random scanning.
  • V. EVALUATION: 31.7% lower neighbor-discovery process delay is achieved by RL-CRA than CRA when 30 neighbor nodes are present.The comparison uses a sensing-distance to communication-distance ratio of 1/2 and a 10-degree JCS transceiver beamwidth.
  • V. EVALUATION: JCSC neighbor discovery can detect hidden nodes, helping avoid potential data collisions and reduce transmission delay in the MAC process.The MAC evaluation connects hidden-node detection from neighbor discovery with lower network data-frame latency.
  • V. EVALUATION: Around 50% lower average data transmission delay is achieved by the JCSC network than the conventional method for a 10-node network.Figure 6 evaluates average data-frame transmission delay as transmitted-frame length varies.

VI. CONCLUSION

The paper proposes JCSC for 6G intelligent machine systems, summarizes enabling technologies and challenges, and uses numerical results to verify feasibility. It identifies deeper integration and fundamental communication-sensing-computation trade-offs as future challenges.

  • VI. CONCLUSION: The proposed JCSC framework targets extremely low latency and high reliability for 6G intelligent machine systems.The conclusion presents JCSC as an integrated framework for these requirements.
  • VI. CONCLUSION: Numerical results verify the feasibility of the proposed JCSC technologies after introducing intelligent flexible manufacturing as a typical use case.The conclusion covers the framework, enabling technologies, challenges, use case, and numerical evaluation.
  • VI. CONCLUSION: Future work must clarify relationships, theoretical bounds, and performance trade-offs among communication, sensing, and computing functions.The conclusion also calls for deeper integration across the physical, MAC, and network layers.
  • VI. CONCLUSION: JCSC implementation remains challenging because integration must advance across the physical, MAC, and network layers.This scope boundary is stated as a future challenge rather than a resolved capability.
Loading 2302.10416v1…