Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Extreme Learning Machine-based Channel Estimation in IRS-Assisted Multi-User ISAC System

Yu Liu, Ibrahim Al-Nahhal, Octavia A. Dobre, Fanggang Wang, Hyundong Shin

TL;DR

This work addresses the challenge of estimating sensing and communication channels in IRS-assisted multi-user ISAC systems, where the IRS is passive and SAC signals interfere. It introduces a practical two-stage estimation approach that first captures direct channels with the IRS off and then reflected channels with the IRS on, coupled with an Extreme Learning Machine (ELM) based framework comprising DE-ELM for direct channels and RE-ELM for reflected channels. By designing two input-output training pair types and employing data augmentation, the method achieves substantial NMSE improvements over LS and CNN benchmarks while offering fast training and comparable online complexity. The results demonstrate strong generalization across SNRs and channel dimensions, highlighting the approach’s potential for low-cost, real-time SAC-channel estimation in IRS-enhanced ISAC deployments.

Abstract

Multi-user integrated sensing and communication (ISAC) assisted by intelligent reflecting surface (IRS) has been recently investigated to provide a high spectral and energy efficiency transmission. This paper proposes a practical channel estimation approach for the first time to an IRS-assisted multiuser ISAC system. The estimation problem in such a system is challenging since the sensing and communication (SAC) signals interfere with each other, and the passive IRS lacks signal processing ability. A two-stage approach is proposed to transfer the overall estimation problem into sub-ones, successively including the direct and reflected channels estimation. Based on this scheme, the ISAC base station (BS) estimates all the SAC channels associated with the target and uplink users, while each downlink user estimates the downlink communication channels individually. Considering a low-cost demand of the ISAC BS and downlink users, the proposed two-stage approach is realized by an efficient neural network (NN) framework that contains two different extreme learning machine (ELM) structures to estimate the above SAC channels. Moreover, two types of input-output pairs to train the ELMs are carefully devised, which impact the estimation accuracy and computational complexity under different system parameters. Simulation results reveal a substantial performance improvement achieved by the proposed ELM-based approach over the least-squares and NN-based benchmarks, with reduced training complexity and faster training speed.

Extreme Learning Machine-based Channel Estimation in IRS-Assisted Multi-User ISAC System

TL;DR

This work addresses the challenge of estimating sensing and communication channels in IRS-assisted multi-user ISAC systems, where the IRS is passive and SAC signals interfere. It introduces a practical two-stage estimation approach that first captures direct channels with the IRS off and then reflected channels with the IRS on, coupled with an Extreme Learning Machine (ELM) based framework comprising DE-ELM for direct channels and RE-ELM for reflected channels. By designing two input-output training pair types and employing data augmentation, the method achieves substantial NMSE improvements over LS and CNN benchmarks while offering fast training and comparable online complexity. The results demonstrate strong generalization across SNRs and channel dimensions, highlighting the approach’s potential for low-cost, real-time SAC-channel estimation in IRS-enhanced ISAC deployments.

Abstract

Multi-user integrated sensing and communication (ISAC) assisted by intelligent reflecting surface (IRS) has been recently investigated to provide a high spectral and energy efficiency transmission. This paper proposes a practical channel estimation approach for the first time to an IRS-assisted multiuser ISAC system. The estimation problem in such a system is challenging since the sensing and communication (SAC) signals interfere with each other, and the passive IRS lacks signal processing ability. A two-stage approach is proposed to transfer the overall estimation problem into sub-ones, successively including the direct and reflected channels estimation. Based on this scheme, the ISAC base station (BS) estimates all the SAC channels associated with the target and uplink users, while each downlink user estimates the downlink communication channels individually. Considering a low-cost demand of the ISAC BS and downlink users, the proposed two-stage approach is realized by an efficient neural network (NN) framework that contains two different extreme learning machine (ELM) structures to estimate the above SAC channels. Moreover, two types of input-output pairs to train the ELMs are carefully devised, which impact the estimation accuracy and computational complexity under different system parameters. Simulation results reveal a substantial performance improvement achieved by the proposed ELM-based approach over the least-squares and NN-based benchmarks, with reduced training complexity and faster training speed.
Paper Structure (27 sections, 56 equations, 6 figures, 2 tables, 1 algorithm)

This paper contains 27 sections, 56 equations, 6 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Proposed ELM-based Estimation Approach
  4. Two-stage Estimation Approach
  5. First Stage
  6. Second Stage
  7. Input-Output Pairs Design
  8. First Stage-Direct SAC Channels Estimation
  9. First Stage-Direct Downlink Communication Channel Estimation
  10. Second Stage-Reflected Uplink Communication Channel Estimation
  11. Second Stage-Reflected Downlink Communication Channel Estimation
  12. Proposed ELM-based Estimation Framework
  13. Training Dataset Generation
  14. Brief Introduction of ELM
  15. Offline Training
  16. ...and 12 more sections

Figures (6)

  • Figure 1: IRS-assisted multi-user ISAC system: (a) System model, (b) Pilot transmission policy.
  • Figure 2: The proposed ELM-based NN framework: (a) Offline training, (b) Online testing.
  • Figure 3: NMSE of SAC channels estimation versus SNR for $M=6$ and $L=30$: (a) SAC channels at the ISAC BS, (b) Communication channels at the downlink $D_j$.
  • Figure 4: NMSE of communication channels estimation versus $L$ for $M=6$ under $\text{SNR}=0\,\rm dB$ and $10\,\rm dB$: (a) Uplink reflected communication channel at the ISAC BS, (b) Downlink reflected communication channel at the downlink $D_j$.
  • Figure 5: NMSE of SAC channels estimation versus $M$ for $L=15$ under $\text{SNR}=0\,\rm dB$ and $10\,\rm dB$: (a) SAC channels at the ISAC BS, (b) Communication channels at the downlink $D_j$.
  • ...and 1 more figures