Lightweight Foundation Model for Wireless Time Series Downstream Tasks on Edge Devices

TL;DR

This paper introduces a lightweight MLP-based foundation model for wireless time series tasks on edge devices, achieving high accuracy with minimal parameters and fast inference, suitable for real-time deployment.

Contribution

The paper proposes a simple MLP-based foundation model that maintains robust performance across multiple wireless tasks, unlike transformer-based models, with significantly fewer parameters.

Findings

Achieves over 97% accuracy on unseen data classes.

Contains only 21K trainable parameters, enabling fast inference.

Supports multiple wireless tasks and input types with consistent performance.

Abstract

While machine learning is widely used to optimize wireless networks, training a separate model for each task in communication and localization is becoming increasingly unsustainable due to the significant costs associated with training and deployment. Foundation models offer a more scalable alternative by enabling a single model to be adapted across multiple tasks through fine-tuning with limited samples. However, current foundation models mostly rely on large-scale Transformer architectures, resulting in computationally intensive models unsuitable for deployment on typical edge devices. This paper presents a lightweight foundation model based on simple Multi-Layer-Perceptron (MLP) encoders that independently process input patches. Our model supports 4 types of downstream tasks (long-range technology recognition, short-range technology recognition, modulation recognition and line-of-sight-detection) from multiple input types (IQ and CIR) and different sampling rates. We show that, unlike Transformers, which can exhibit performance drops as downstream tasks are added, our MLP model maintains robust generalization performance, achieving over 97% accurate fine-tuning results for previously unseen data classes. These results are achieved despite having only 21K trainable parameters, allowing an inference time of 0.33 ms on common edge devices, making the model suitable for constrained real-time deployments.