Latency Reduction for Mobile Edge Computing in HetNets by Uplink and Downlink Decoupled Access

TL;DR

This paper investigates how decoupled uplink and downlink access in heterogeneous networks can reduce latency in mobile edge computing, despite backhaul delays, by formulating and comparing latency models.

Contribution

It introduces a new MEC offloading scheme with decoupled UL/DL access, demonstrating potential latency improvements over traditional coupled schemes using stochastic geometry.

Findings

Decoupled access reduces offloading latency compared to coupled access.

Backhaul delay has limited impact on the benefits of decoupled access.

Decoupled UL/DL association improves network throughput and latency performance.

Abstract

Achieving an end-to-end low-latency for computations offloading, in Mobile Edge Computing (MEC) systems, is still a critical design problem. This is because the offloading of computational tasks via the MEC servers entails the use of uplink (UL) and downlink (DL) radio links that are usually assumed to be coupled to a single base station (BS). However, for heterogeneous networks, a new architectural paradigm whereby UL and DL are not associated with the same BS is proposed and seen to provide gains in network throughput due to the improved UL performance. Motivated by such gains, and by using typical results from stochastic geometry, we formulate the offloading latency for the MEC-based scheme with decoupled UL/DL association, or decoupled access, and compare its performance to the conventional coupled access scheme. Despite the backhaul delay necessary for the communication between the two serving BSs in UL and DL, the offloading scheme with decoupled access is still capable of providing a fairly lower offloading latency compared to the conventional offloading scheme with coupled access.