# Parameter-Free Statistical Generator-Based Class Incremental Learning for Multi-User Physical Layer Authentication in the Industrial Internet of Things

## TL;DR

This paper introduces a lightweight method for updating user authentication systems in industrial IoT as new users join, without requiring heavy computations.

## Contribution

A parameter-free statistical generator-based class incremental learning framework is proposed for efficient multi-user physical layer authentication.

## Key findings

- PSG-CIL achieves superior accuracy with lower computational overhead compared to existing methods.
- In the AAP outer loop scenario, PSG-CIL outperforms retraining from scratch with 70.68% accuracy.
- The method maintains performance while adapting to new users without forgetting old ones.

## Abstract

Deep learning (DL)-based multi-user physical layer authentication (PLA) in the Industrial Internet of Things (IIoT) requires frequent updates as new users join. Class incremental learning (CIL) addresses this challenge, but existing generative replay approaches depend on heavy parameterized models, causing high computational overhead and limiting deployment in resource-constrained environments. To address these challenges, we propose a parameter-free statistical generator-based CIL framework, PSG-CIL, for DL-based multi-user PLA in the IIoT. The parameter-free statistical generator (PSG) produces Gaussian sampling on user-specific means and variances to generate pseudo-data without training extra models, greatly reducing computational overhead. A confidence-based pseudo-data selection ensures pseudo-data reliability, while a dynamic adjustment mechanism for the loss weight balances the retention of old users’ knowledge and the adaptation to new users. Experiments on real industrial datasets show that PSG-CIL consistently achieves superior accuracy while maintaining a lightweight scale; for example, in the AAP outer loop scenario, PSG-CIL reaches 70.68%, outperforming retraining from scratch (58.57%) and other CIL methods.