Advancing 5G Networks with Container-Based Multi-Access Edge Computing.

Advancing 5G Networks with Container-Based Multi-Access Edge Computing.

Abstract:

The telecommunication industry has witnessed rapid growth and remarkable engineering advancements. However, the widespread use of portable mobile handheld devices has led to constraints in computational power, storage, and energy capacity due to the need for portability, compact form factors, ergonomic design, and style. Various solutions, such as cloudlets, cyber foraging, mobile cloud computing (MCC), and more recently, multi-access edge computing (MEC), have been proposed to address these limitations. The deployment of 5G and its associated latency-sensitive and resource-intensive applications present new and emerging use cases.

To tackle these challenges, this study introduces the use of secure containerization for MEC applications and proposes locating the MEC host at the 5G centralized unit within the radio access network (RAN). This setup facilitates offloading computational, storage, and analytics requirements closer to the user equipment (UE) at the network’s edge, where data is generated and results are applied. The key contribution of this thesis is the use of secured containerization technology to replace virtual machines, enabling the adoption of containers for MEC applications. This not only reduces application overhead but also ensures the necessary isolation of MEC infrastructure, meeting the security requirements outlined by the European Technology Standard Institute (ETSI).

The study evaluates 5G end-to-end transport specifications to determine the optimal location of the MEC server within the RAN, achieving theoretical latency values ranging from 4.1ms to 14.1ms. These latency figures satisfy the demands of latency-sensitive applications, such as virtual reality/augmented reality (VR/AR), tactile Internet, Vehicle-to-Vehicle communication, and Manufacturing & Robotic Control/Safety Systems. The results confirm that edge computing offers lower user plane latency and reduced backhaul traffic, resulting in a lower application failure rate.

Secured containerized multi-access computing infrastructures offer significant advantages over mobile cloud computing, particularly in addressing computational power and energy capacity deficiencies for asymmetric UE applications. Hosting applications within the RAN provides robust support for emerging application requirements, offering ample computational, storage, and analytics capabilities at low latency figures. Edge deployments alleviate the pressure on network operators’ backhaul links, safeguarding the end-to-end ecosystem from potential collapse due to heavy backhaul traffic from billions of 5G UEs.

Regardless of the speed of 5G networks, MEC ensures that extensive data transportation for cloud-based processing and result retrieval is unnecessary. This approach enhances privacy and security while conserving precious bandwidth resources.

Advancing 5G Networks with Container-Based Multi-Access Edge Computing.    GET MORE, ACTUARIAL SCIENCE PROJECT TOPICS AND MATERIALS