Distributed Deep Learning at the Edge: A Novel Proactive and Cooperative Caching Framework for Mobile Edge Networks

Abstract

This letter proposes two novel proactive cooperative caching approaches using deep learning (DL) to predict users' content demand in a mobile edge caching network. In the first approach, a (central) content server takes responsibilities to collect information from all mobile edge nodes (MENs) in the network and then performs our proposed deep learning (DL) algorithm to predict the content demand for the whole network. However, such a centralized approach may disclose the private information because MENs have to share their local users' data with the content server. Thus, in the second approach, we propose a novel distributed deep learning (DDL) based framework. The DDL allows MENs in the network to collaborate and exchange information to reduce the error of content demand prediction without revealing the private information of mobile users. Through simulation results, we show that our proposed approaches can enhance the accuracy by reducing the root mean squared error (RMSE) up to 33.7% and reduce the service delay by 36.1% compared with other machine learning algorithms.Index Terms-Mobile edge caching, deep learning, distributed deep learning, proactive and cooperative caching.

Machine Learning: A Catalyst for THz Wireless Networks

“…The main idea is to proactively transfer the content to be requested by a single of a cluster of UEs to the nearest possible BS/AP. Towards this direction, several researchers presented ML-based caching policies (Cheng et al, 2019;Jiang et al, 2019;Saputra et al, 2019;Wang X. et al, 2020;Kirilin et al, 2020;Ye et al, 2020). Specifically, in (Ye et al, 2020), the authors reported a device to primary and secondary BS clustering approach based on the requested content location in mmW ultra-dense wireless networks.…”

A Novel Predictive-Collaborative-Replacement (PCR) Intelligent Caching Scheme for Multi-Access Edge Computing

“…Popular preferences are then cached to improve the global cache hits. Saputra et al [30] have proposed two proactive and cooperative caching framework for mobile edge network. In the first approach, the edge nodes send data to a central server, which then creates a deep learning model based on the data popularity and sends it to the edge servers.…”

Dueling Deep-Q-Network Based Delay-Aware Cache Update Policy for Mobile Users in Fog Radio Access Networks

“…Besides, two popularity prediction algorithms are developed for two noise models. By using deep learning, authors in [11] proposed two proactive cooperative caching algorithms to predict user preferences in a centralized way and a distributed way, respectively. By learning user preference, two edge caching architectures are proposed to predict content popularity in [12].…”

“…Besides, cooperative caching [11], [18]- [22] is an effective approach to improve the utilization of storage resource. Researchers in [18] focused on a cooperative edge caching architecture for content-centric 5G networks, and proposed a mobility-aware caching framework for MUs.…”

“…In addition, cooperative caching [11], [18]- [22] is an effective way to improve the cache space utilization. The cooperative caching means that the requested contents can be obtained from multiple content providers via content sharing and other manners.…”