search query: @instructor Lassila, Pasi / total: 14
reference: 6 / 14
Author: | Osti, Prajwal |
Title: | Optimal flow-level performance of opportunistic scheduling with size information |
Publication type: | Master's thesis |
Publication year: | 2011 |
Pages: | [6] + 42 s. + liitt. 6 Language: eng |
Department/School: | Tietoliikenne- ja tietoverkkotekniikan laitos |
Main subject: | Tietoverkkotekniikka (S-38) |
Supervisor: | Aalto, Samuli |
Instructor: | Lassila, Pasi |
Electronic version URL: | http://urn.fi/URN:NBN:fi:aalto-201207022680 |
OEVS: | Electronic archive copy is available via Aalto Thesis Database.
Instructions Reading digital theses in the closed network of the Aalto University Harald Herlin Learning CentreIn the closed network of Learning Centre you can read digital and digitized theses not available in the open network. The Learning Centre contact details and opening hours: https://learningcentre.aalto.fi/en/harald-herlin-learning-centre/ You can read theses on the Learning Centre customer computers, which are available on all floors.
Logging on to the customer computers
Opening a thesis
Reading the thesis
Printing the thesis
|
Location: | P1 Ark Aalto 1233 | Archive |
Keywords: | proportional fairness optimality opportunistic scheduling SRPT SRPT-FM sized-based scheduling flow-level performance |
Abstract (eng): | Proportional fair (PF) is well known as the algorithm of choice for scheduling data flow in a wireless setting. Although considered as having good performance characteristics with nice fairness properties, PF is far from optimal when it comes to minimizing the flow-level delays. In this thesis, we study the optimal scheduling policy in an opportunistic wireless environment. The analysis can be applied to any setting where some kind of rate region can be constructed. As the main result we prove that when a fixed number of flows can be served at an average rate taken from a compact and symmetric rate region, the optimal policy is consistent with the SRPT-FM scheduling policy used in queuing systems with multiple servers. Moreover, we see that for some specially crafted rate region we can also get the explicit analytic expressions of the optimal long term service rates that minimize the cumulative delay of the flows. Through simulation, we observe that in a dynamic setting the performance of this static-case optimal policy is no worse than the PF policy in this kind of rate region. For a case with asymmetric service requirements, we consider a case with only two flows where we observe that if the rate region has certain characteristics met in practice, there are only two possible optimal rate pairs. The choice between them can be made based on the size of flows at hand. We also study a numerical example where the optimal policy still follows the spirit of the SRPT-FM policy even in the asymmetric case. |
ED: | 2011-09-08 |
INSSI record number: 42721
+ add basket
INSSI