Cloud Radio Access Networks with Coded Caching

TL;DR

This paper explores the use of coded caching in cloud radio access networks to reduce backhaul load and power consumption, demonstrating its advantages over uncoded caching through optimization techniques.

Contribution

It introduces a convex optimization framework for coded caching in C-RANs, showing how it effectively reduces backhaul and transmission costs.

Findings

Coded caching decreases backhaul load compared to uncoded caching.

Optimization techniques improve network cost efficiency.

C-RAN with coded caching enhances data delivery performance.

Abstract

A cloud radio access network (C-RAN) is considered as a candidate to meet the expectations of higher data rate de- mands in wireless networks. In C-RAN, low energy base stations (BSs) are deployed over a small geography and are allowed to connect to the cloud via finite capacity backhaul links where the information is processed. A conventional C-RAN, however, requires high capacity backhaul links, since the requested files need to be transferred first from the cloud to the BS before conveying them to the users. One approach to overcome the limitations of the backhaul links is to introduce local storage caches at the BSs, in which the popular files are stored locally in order to reduce the load of the backhaul links. Furthermore, we utilize coded caching with the goal to minimize the total network cost, i.e., the transmit power and the cost associated with the backhaul links. The initial formulation of the optimization problem for this model is non-convex. We first reformulate and then convexify the problem through some relaxation techniques. In comparison to the uncoded caching at the BSs, our results highlight the benefits associated with coded caching and show that it decreases the backhaul cost.