Search

The growing network density and unprecedented increase in network traffic, caused by the massively expanding number of connected devices and online services, require intelligent network operations. Machine Learning (ML) has been applied in this regard in different types of networks and networking technologies to meet the requirements of future communicating devices and services. In this article, we provide a detailed account of current research on the application of ML in communication networks and shed light on future research challenges. Research on the application of ML in communication networks is described in: i) the three layers, i.e., physical, access, and network layers; and ii) novel computing and networking concepts such as Multi-access Edge Computing (MEC), Software Defined Networking (SDN), Network Functions Virtualization (NFV), and a brief overview of ML-based network security. Important future research challenges are identified and presented to help stir further research in key areas in this direction.

Multi-access edge computing has been proposed as a promising approach to localize the access of mobile clients to the network edges, therefore, reducing significantly the traffic congestion on the backhaul network. Due to time-varying wireless channel condition, the video caching at the mobile edges for dynamic adaptive video streaming over HTTP (DASH) needs to be efficiently handled to alleviate the high bandwidth demand on the backhaul network and improve the quality of experience (QoE) of end users. We investigate the impact of collaborative mobile edge caching on joint QoE and backhaul data traffic by proposing the joint QoE-traffic optimization with collaborative edge caching which introduces the BFTR (backhaul/fronthaul traffic ratio) parameter adjustable by the mobile network operator. We then design a self-tuned bitrate selection algorithm with low complexity to solve the optimization problem and further propose an efficient cache replacement strategy called retention-based collaborative caching. Through simulation-based evaluations, we show a noticeable gain in the percentage of cache miss and specify some threshold for BFTR parameter after which the significant reduction in the data traffic with further improvement in average video bitrate is obtained using collaborative caching. Our findings help mobile edge system developers design an efficient collaborative caching mechanism for 5G networks.

Content-centric networking (CCN) introduces a paradigm shift from a host centric to an information centric communication model for future Internet architectures. It supports the retrieval of a particular content regardless of the physical location of the content. Content caching and content delivery networks are the most popular approaches to deal with the inherent issues of content delivery on the Internet that are caused by its design. Moreover, intermittently connected mobile environments or disruptive networks present a significant challenge to CCN deployment. In this paper, we consider the possibility of using mobile users in improving the efficiency of content delivery. Mobile users are producing a significant fraction of the total Internet traffic, and modern mobile devices have enough storage to cache the downloaded content that may interest other mobile users for a short period too. We present an analytical model of the CCN framework that integrates a delay tolerant networking architecture into the native CCN, and we present large-scale simulation results. Caching on mobile devices can improve the content retrieval time by more than 50%, while the fraction of the requests that are delivered from other mobile devices can be more than 75% in many cases.