Enabling MAC protocol implementations on software-defined radios

  • Authors:
  • George Nychis;Thibaud Hottelier;Zhuocheng Yang;Srinivasan Seshan;Peter Steenkiste

  • Affiliations:
  • Carnegie Mellon University;Carnegie Mellon University;Carnegie Mellon University;Carnegie Mellon University;Carnegie Mellon University

  • Venue:
  • NSDI'09 Proceedings of the 6th USENIX symposium on Networked systems design and implementation
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Over the past few years a range of new Media Access Control (MAC) protocols have been proposed for wireless networks. This research has been driven by the observation that a single one-size-fits-all MAC protocol cannot meet the needs of diverse wireless deployments and applications. Unfortunately, most MAC functionality has traditionally been implemented on the wireless card for performance reasons, thus, limiting the opportunities for MAC customization. Software-defined radios (SDRs) promise unprecedented flexibility, but their architecture has proven to be a challenge for MAC protocols. In this paper, we identify a minimum set of core MAC functions that must be implemented close to the radio in a high-latency SDR architecture to enable high performance and efficient MAC implementations. These functions include: precise scheduling in time, carrier sense, backoff, dependent packets, packet recognition, fine-grained radio control, and access to physical layer information. While we focus on an architecture where the bus latency exceeds common MAC interaction times (tens to hundreds of microseconds), other SDR architectures with lower latencies can also benefit from implementing a subset of these functions closer to the radio. We also define an API applicable to all SDR architectures that allows the host to control these functions, providing the necessary flexibility to implement a diverse range of MAC protocols. We show the effectiveness of our split-functionality approach through an implementation on the GNU Radio and USRP platforms. Our evaluation based on microbenchmarks and end-to-end network measurements, shows that our design can simultaneously achieve high flexibility and high performance.