Electromagnetic Property Sensing Based on Diffusion Model in ISAC System

Yuhua Jiang,Feifei Gao,Shi Jin,Tie Jun Cui

2024-10-20

Abstract:Integrated sensing and communications (ISAC) has opened up numerous game-changing opportunities for future wireless systems. In this paper, we develop a novel ISAC scheme that utilizes the diffusion model to sense the electromagnetic (EM) property of the target in a predetermined sensing area. Specifically, we first estimate the sensing channel by using both the communications and the sensing signals echoed back from the target. Then we employ the diffusion model to generate the point cloud that represents the target and thus enables 3D visualization of the target's EM property distribution. In order to minimize the mean Chamfer distance (MCD) between the ground truth and the estimated point clouds, we further design the communications and sensing beamforming matrices under the constraint of a maximum transmit power and a minimum communications achievable rate for each user equipment (UE). Simulation results demonstrate the efficacy of the proposed method in achieving high-quality reconstruction of the target's shape, relative permittivity, and conductivity. Besides, the proposed method can sense the EM property of the target effectively in any position of the sensing area.

Signal Processing

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to accurately sense the electromagnetic (EM) characteristics of targets using diffusion models in the Integrated Sensing and Communication (ISAC) system. Specifically, the research aims to estimate the sensing channel by simultaneously using communication and sensing signals and generate a point cloud representing the target, thereby achieving three - dimensional visualization of electromagnetic characteristics such as target shape, relative permittivity, and conductivity.

To achieve this goal, the paper proposes the following key steps:

1. **Sensing Channel Estimation**: Estimate the sensing channel by receiving communication and sensing signals reflected from the target.
2. **Point Cloud Generation**: Use the diffusion model to generate a point cloud representing the target to achieve 3D visualization.
3. **Optimization Design**: Design communication and sensing beamforming matrices to minimize the Mean Chamfer Distance (MCD) between the real point cloud and the estimated point cloud, and optimize under the constraints of maximum transmit power and minimum communication rate.

### Formula Summary

- **Sensing Channel Estimation**:
\[
x(l) = W_s s(l) + W_c c(l) = \begin{bmatrix} W_s & W_c \end{bmatrix} \begin{bmatrix} s(l)^T & c(l)^T \end{bmatrix}^T = \hat{W} x(l)
\]
where \( s(l) \in \mathbb{C}^{N_t \times 1} \) is the dedicated sensing signal, and \( c(l) \in \mathbb{C}^{K \times 1} \) is the symbol sent to the user.

- **Point Cloud Generation**:
\[
\tilde{y} = (X^T \otimes I_{N_r}) \tilde{h}_s + \tilde{z}
\]
where \( \tilde{y} = \text{vec}(Y) \), \( \tilde{h}_s = \text{vec}(H_s) \), and \( \tilde{z} = \text{vec}(Z) \).

- **Forward Diffusion Process in the Diffusion Model**:
\[
q(x_i^{(1:T)} | x_i^{(0)}) = \prod_{t = 1}^T q(x_i^{(t)} | x_i^{(t - 1)})
\]
\[
q(x_i^{(t)} | x_i^{(t - 1)}) = \mathcal{N}\left(x_i^{(t)} \mid \sqrt{1 - \beta_t} x_i^{(t - 1)}, \beta_t I\right)
\]

- **Reverse Diffusion Process in the Diffusion Model**:
\[
p_\theta(x_i^{(0:T)} | \hat{H}_s) = p(x_i^{(T)}) \prod_{t = 1}^T p_\theta(x_i^{(t - 1)} | x_i^{(t)}, \hat{H}_s)
\]
\[
p_\theta(x_i^{(t - 1)} | x_i^{(t)}, \hat{H}_s) = \mathcal{N}\left(x_i^{(t - 1)} \mid \mu_\theta(x_i^{(t)}, t, \hat{H}_s), \beta_t I\right)
\]

- **Maximization of the Variational Lower Bound**:
\[
\text{VLB}(\hat{H}_s) = \mathbb{E}_{q(x_i^{(0:T)})} \left[ \ln \frac{p_\theta(x_i^{(0:T)} | \hat{H}_s)}{q(x_i^{(1:T)} | x_i^{(0)})} \right]
\]

Through these steps, the paper demonstrates the effectiveness of the proposed method in high - quality reconstruction of target shape, relative permittivity, and conductivity.