Wireless Peer-to-Peer Scheduling in Mobile Networks

TL;DR

This paper introduces a new peer-to-peer scheduling algorithm for mobile wireless networks that reduces delay by opportunistically downloading from neighbors, and also develops a throughput-utility optimization algorithm for single-file requests.

Contribution

It presents a novel delay-reducing scheduling algorithm for mobile networks and a throughput-utility optimization method for single-file requests with incentives.

Findings

The new algorithm significantly reduces delay in mobile peer-to-peer networks.

The throughput-utility algorithm provably incentivizes participation.

Heuristic extensions perform well in more complex scenarios.

Abstract

This paper considers peer-to-peer scheduling for a network with multiple wireless devices. A subset of the devices are mobile users that desire specific files. Each user may already have certain popular files in its cache. The remaining devices are access points that typically have access to a larger set of files. Users can download packets of their requested file from an access point or from a nearby user. Our prior work optimizes peer scheduling in a general setting, but the resulting delay can be large when applied to mobile networks. This paper focuses on the mobile case, and develops a new algorithm that reduces delay by opportunistically grabbing packets from current neighbors. However, it treats a simpler model where each user desires a single file with infinite length. An algorithm that provably optimizes throughput utility while incentivizing participation is developed for this case. The algorithm extends as a simple heuristic in more general cases with finite file sizes and random active and idle periods.