Scheduling Algorithms in Wireless Networking Using Machine Learning

Master Thesis

open


Motivation

Carrier-sense multiple access (CSMA)-type random access algorithms and their variants such as the queue-length based CSMA (Q-CSMA) [1] are widely adopted in the ad-hoc wireless networking community. Their performance with respect to various metrics, such as throughput and delay have been studied in the literature under different sets of assumptions. Some of these link scheduling algorithms or medium access control (MAC) algorithms have been shown to possess good theoretical guarantees such as throughput optimality. However, there is a need for scheduling algorithms that optimize a combination of two or more performance metrics. An example would be to optimize delay while keeping the queue lengths below a pre-specified threshold. The objective of this master thesis is to design scheduling algorithms to enhance the performance of wireless multi-hop networks based on machine learning techniques. A possible approach to solve this problem in a decentralised fashion is to formulate a suitable decentralised partially observable Markov decision process (Dec-POMDP). Since solving a Dec-POMDP is computationally infeasible for large state spaces, an important aspect of the master thesis will be to devise an approximate solution method for the associated Dec-POMDP tailored to the wireless multi-hop networking setup. The thesis will be supervised by Wasiur Rahman Khuda Bukhsh from the Bioinspired Communication Systems Lab and co-supervised by Robin Klose from the Secure Mobile Networking Lab, department of Computer Science, Technische Universität Darmstadt. The candidate will get an opportunity to test his/her ideas on full-stack simulation environments. Good programming skills are a prerequisite. The candidate is expected to be familiar with Markov chains. Willingness to learn some machine learning topics as and when required is also necessary for this project.

Contact person: Wasiur Rahman Khuda Bukhsh


[1] J. Ni, B. Tan, R. Srikant, Q-CSMA: Queue-Length-Based CSMA/CA Algorithms for Achieving Maximum Throughput and Low Delay in Wireless Networks, IEEE/ACM Transactions on Networking, 10.1109/TNET.2011.2177101, 2012.
[2] M. Abu Alsheikh, D. T. Hoang, D. Niyato, H. P. Tan, S. Lin, Markov Decision Processes With Applications in Wireless Sensor Networks: A Survey, IEEE Communications Surveys Tutorials, 10.1109/COMST.2015.2420686, 2015.


Supervisor:

Research Areas: Sichere Mobile Netze



Back


Contact

Prof. Dr.-Ing. Matthias Hollick

Technische Universität Darmstadt
Department of Computer Science
Secure Mobile Networking Lab 

Mornewegstr. 32 (S4/14)
64293 Darmstadt, Germany

Phone: +49 6151 16-25472
Fax: +49 6151 16-25471
office@seemoo.tu-darmstadt.de

Affiliations

A A A | Drucken Drucken | Impressum Impressum | Sitemap Sitemap | Suche Suche | Kontakt Kontakt | Webseitenanalyse: Mehr Informationen
zum Seitenanfangzum Seitenanfang