Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep Learning-based Design of Uplink Integrated Sensing and Communication

Qiao Qi, Xiaoming Chen, Caijun Zhong, Chau Yuen, Zhaoyang Zhang

TL;DR

This work tackles uplink ISAC in 6G, where sensing echoes and uplink communications collide in the BS receiver. It introduces a DL-based joint design (ISACNN) to jointly optimize sensing waveform and receive beamforming by transforming a challenging MOOP into tractable SOOPs and an unsupervised learning objective. The approach leverages MI-based sensing metrics and SINR-based communications metrics, uses DL to predict sensing-singular values and beams, and recovers the full solution with a principled post-processing step. Simulations show that ISACNN achieves higher performance and robustness than traditional optimization methods, with reduced computational complexity and resilience to imperfect CSI.

Abstract

In this paper, we investigate the issue of uplink integrated sensing and communication (ISAC) in 6G wireless networks where the sensing echo signal and the communication signal are received simultaneously at the base station (BS). To effectively mitigate the mutual interference between sensing and communication caused by the sharing of spectrum and hardware resources, we provide a joint sensing transmit waveform and communication receive beamforming design with the objective of maximizing the weighted sum of normalized sensing rate and normalized communication rate. It is formulated as a computationally complicated non-convex optimization problem, which is quite difficult to be solved by conventional optimization methods. To this end, we first make a series of equivalent transformation on the optimization problem to reduce the design complexity, and then develop a deep learning (DL)-based scheme to enhance the overall performance of ISAC. Both theoretical analysis and simulation results confirm the effectiveness and robustness of the proposed DL-based scheme for ISAC in 6G wireless networks.

Deep Learning-based Design of Uplink Integrated Sensing and Communication

TL;DR

This work tackles uplink ISAC in 6G, where sensing echoes and uplink communications collide in the BS receiver. It introduces a DL-based joint design (ISACNN) to jointly optimize sensing waveform and receive beamforming by transforming a challenging MOOP into tractable SOOPs and an unsupervised learning objective. The approach leverages MI-based sensing metrics and SINR-based communications metrics, uses DL to predict sensing-singular values and beams, and recovers the full solution with a principled post-processing step. Simulations show that ISACNN achieves higher performance and robustness than traditional optimization methods, with reduced computational complexity and resilience to imperfect CSI.

Abstract

In this paper, we investigate the issue of uplink integrated sensing and communication (ISAC) in 6G wireless networks where the sensing echo signal and the communication signal are received simultaneously at the base station (BS). To effectively mitigate the mutual interference between sensing and communication caused by the sharing of spectrum and hardware resources, we provide a joint sensing transmit waveform and communication receive beamforming design with the objective of maximizing the weighted sum of normalized sensing rate and normalized communication rate. It is formulated as a computationally complicated non-convex optimization problem, which is quite difficult to be solved by conventional optimization methods. To this end, we first make a series of equivalent transformation on the optimization problem to reduce the design complexity, and then develop a deep learning (DL)-based scheme to enhance the overall performance of ISAC. Both theoretical analysis and simulation results confirm the effectiveness and robustness of the proposed DL-based scheme for ISAC in 6G wireless networks.
Paper Structure (19 sections, 2 theorems, 41 equations, 11 figures, 2 tables)

This paper contains 19 sections, 2 theorems, 41 equations, 11 figures, 2 tables.

Table of Contents

  1. Introduction
  2. System Model
  3. Sensing Information Extraction
  4. Communication Signal Decoding
  5. Problem Formulation and Transformation
  6. Sensing Rate Maximization
  7. Communication Rate Maximization
  8. Problem Transformation
  9. DL-based Scheme Design
  10. Pre-processing
  11. Post-processing
  12. Network Structure
  13. Recover Module
  14. Complexity Analysis
  15. Simulation Results
  16. ...and 4 more sections

Key Result

Lemma 1

For arbitrary matrice $\mathbf{A}\in\mathbb{C}^{m\times n}$, $\mathbf{B}\in\mathbb{C}^{n\times p}$ and $\mathbf{C}\in\mathbb{C}^{p\times q}$, we have $\text{vec}\left( \mathbf{ABC} \right)=({{\mathbf{C}}^{T}}\otimes \mathbf{A})\text{vec}\left( \mathbf{B} \right)$.

Figures (11)

  • Figure 1: An uplink ISAC system.
  • Figure 2: System flowchart, where "Info." denotes information.
  • Figure 3: Process of problem transformation.
  • Figure 4: The Proposed DL-based scheme for joint waveform and beamforming design.
  • Figure 5: Framework of "ISACNN".
  • ...and 6 more figures

Theorems & Definitions (4)

  • Remark 1
  • Lemma 1
  • Remark 2
  • Lemma 2