ACM Home Page
Please provide us with feedback. Feedback
On stability and maximum throughput of exponential backoff mechanisms with two users
Full text PdfPdf (202 KB)
Source
International Workshop on Modeling Analysis and Simulation of Wireless and Mobile Systems archive
Proceedings of the 3nd ACM workshop on Performance monitoring and measurement of heterogeneous wireless and wired networks table of contents
Vancouver, British Columbia, Canada
Pages 72-76  
Year of Publication: 2008
ISBN:978-1-60558-239-9
Authors
Flaminio Borgonovo  Politecnico di Milano, Milan, Italy
Matteo Cesana  Politecnico di Milano, Milan, Italy
Sponsors
SIGSIM: ACM Special Interest Group on Simulation and Modeling
ACM: Association for Computing Machinery
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 5,   Downloads (12 Months): 43,   Citation Count: 0
Additional Information:

abstract   references   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/1454630.1454641
What is a DOI?

ABSTRACT

The Binary Exponential Backoff (BEB) mechanism plays a key role in determining stability and efficiency of popular MAC protocols, such as IEEE 802.3 and IEEE 802.11. Unfortunately, the maximum throughput allowed by this protocol under different system parameters and user's multiplicity is largely unknown. In this paper we evaluate the capacity of a two-station system under Bernoulli arrivals, which is proved in the literature to be lower bounded by 0:3 under balanced arrivals. We refer to the general exponential backoff law b-i and show that, when close to capacity, the system behaves in a simple manner, which allows us to assess capacity without resorting to the general Markov Chain approach. Our approach shows that the capacity of BEB under balanced load is equal to 0.6096. We also provide the stability region when arrivals are unbalanced in the two queues. Simulation results support our finding.



Collaborative Colleagues:
Flaminio Borgonovo: colleagues
Matteo Cesana: colleagues