DRL-Based Resource Allocation for Motion Blur Resistant Federated Self-Supervised Learning in IoV

TL;DR

This paper introduces a DRL-based resource allocation scheme for federated self-supervised learning in IoV, enhancing privacy, reducing motion blur effects, and optimizing energy and latency in model aggregation.

Contribution

It proposes a novel DRL-based resource allocation method for motion blur-resistant federated self-supervised learning in IoV, addressing privacy, energy, and latency challenges.

Findings

The DRL-based scheme effectively reduces energy consumption and latency.

The motion blur-resistant method improves model aggregation accuracy.

Simulation results validate the proposed methods' effectiveness.

Abstract

In the Internet of Vehicles (IoV), Federated Learning (FL) provides a privacy-preserving solution by aggregating local models without sharing data. Traditional supervised learning requires image data with labels, but data labeling involves significant manual effort. Federated Self-Supervised Learning (FSSL) utilizes Self-Supervised Learning (SSL) for local training in FL, eliminating the need for labels while protecting privacy. Compared to other SSL methods, Momentum Contrast (MoCo) reduces the demand for computing resources and storage space by creating a dictionary. However, using MoCo in FSSL requires uploading the local dictionary from vehicles to Base Station (BS), which poses a risk of privacy leakage. Simplified Contrast (SimCo) addresses the privacy leakage issue in MoCo-based FSSL by using dual temperature instead of a dictionary to control sample distribution. Additionally, considering the negative impact of motion blur on model aggregation, and based on SimCo, we propose a motion blur-resistant FSSL method, referred to as BFSSL. Furthermore, we address energy consumption and delay in the BFSSL process by proposing a Deep Reinforcement Learning (DRL)-based resource allocation scheme, called DRL-BFSSL. In this scheme, BS allocates the Central Processing Unit (CPU) frequency and transmission power of vehicles to minimize energy consumption and latency, while aggregating received models based on the motion blur level. Simulation results validate the effectiveness of our proposed aggregation and resource allocation methods.