Caching at the Edge with LT codes

TL;DR

This paper analyzes the performance of LT code-based caching schemes in a multi-transmitter network, deriving optimal placement strategies and comparing them to MDS codes, showing comparable effectiveness.

Contribution

It introduces a caching scheme using LT codes with optimized placement to minimize backhaul transmission, providing a closed-form expression for the rate and comparison with MDS codes.

Findings

LT code caching performs as well as MDS code caching

Closed-form expression for backhaul transmission rate derived

Optimized cache placement reduces backhaul load

Abstract

We study the performance of caching schemes based on LT under peeling (iterative) decoding algorithm. We assume that users ask for downloading content to multiple cache-aided transmitters. Transmitters are connected through a backhaul link to a master node while no direct link exists between users and the master node. Each content is fragmented and coded with LT code. Cache placement at each transmitter is optimized such that transmissions over the backhaul link is minimized. We derive a closed form expression for the calculation of the backhaul transmission rate. We compare the performance of a caching scheme based on LT with respect to a caching scheme based on maximum distance separable codes. Finally, we show that caching with \acl{LT} codes behave as good as caching with maximum distance separable codes.