Adjustment of Cluster-Then-Predict Framework for Multiport Scatterer Load Prediction

TL;DR

The paper tackles the challenge of predicting interdependent load impedances in high-dimensional multiport scatterers by proposing a two-stage cluster-then-predict framework that first clusters data and then trains cluster-specific regressors. It compares K-means and OT-based clustering coupled with GB and KNN regression, and introduces the Real-world Unified Index (RUI) to balance competing objectives such as accuracy, cluster quality, and inference time. A key finding is that GB benefits from clustering, achieving up to about a 46% RMSE reduction, while K-means + KNN emerges as the optimal setup under RUI for practical deployment. The work advances real-time design and measurement workflows for RIS-supported channels and virtual VNAs by enabling accurate, fast multi-load impedance prediction on large, realistic datasets.

Abstract

Predicting interdependent load values in multiport scatterers is challenging due to high dimensionality and complex dependence between impedance and scattering ability, yet this prediction remains crucial for the design of communication and measurement systems. In this paper, we propose a two-stage cluster-then-predict framework for multiple load values prediction task in multiport scatterers. The proposed cluster-then-predict approach effectively captures the underlying functional relation between S-parameters and corresponding load impedances, achieving up to a 46% reduction in Root Mean Square Error (RMSE) compared to the baseline when applied to gradient boosting (GB). This improvement is consistent across various clustering and regression methods. Furthermore, we introduce the Real-world Unified Index (RUI), a metric for quantitative analysis of trade-offs among multiple metrics with conflicting objectives and different scales, suitable for performance assessment in realistic scenarios. Based on RUI, the combination of K-means clustering and k-nearest neighbors (KNN) is identified as the optimal setup for the analyzed multiport scatterer.

Figures (1)

• Figure 1: Metric values per cluster size for 7 methods for K-means clustering (red-toned lines) and OT K-means clustering (blue-toned lines). GB and KNN are distinguished using circle and triangle markers, respectively.