Paper: A Deep Generative Model for Synthesizing Labeled Wireless Signals (IIns-GAN)

Forum topic · 小凯 · 2026-09-08

Summary

A forum post on zhichai.net shares an arXiv paper (2609.05396) by Yuxiao Li, Keke Hu, Santiago Mazuelas, and Yuan Shen, introducing Inter-Instance Generative Adversarial Networks (IIns-GAN), a deep learning method for synthesizing realistic labeled wireless signals. Position-labeled wireless signals are crucial for performance evaluation and model training in wireless sensing, but real-world datasets are costly to measure and label. Traditional synthesis approaches depend on environmental models, requiring extensive hyper-parameter tuning and producing signals that lack realism for comprehensive model training. The proposed IIns-GAN generates realistic labeled wireless signals that adapt to different environment scenarios and suit various model training tasks. The post was auto-collected on 2026-09-08 and includes the original abstract; a Chinese translation is not yet available.

Paper Overview

Research area: ML Authors: Yuxiao Li, Keke Hu, Santiago Mazuelas, Yuan Shen Published: 2026-09-04 arXiv: 2609.05396

Chinese Abstract

Translation not yet available.

Original Abstract

Wireless signals with position-related labels are pivotal for both performance evaluation and model training in the realm of wireless sensing. However, acquiring real-world datasets is often challenged by significant measurement and labeling costs. Traditional methods for synthesizing labeled wireless signals typically rely on environmental models, leading to extensive hyper-parameter tuning and inadequate realism for comprehensive model training purposes. To address these limitations, we introduce a novel deep learning (DL)-based method, namely Inter-Instance Generative Adversarial Networks (IIns-GAN), to generate realistic labeled wireless signals. The generated signals are particularly adaptive to different environment scenarios and well-suited for various model training tasks.

--- *Auto-collected on 2026-09-08*