Delay based approach to support low priority users in preemptive wireless networks

At times of serious disasters (natural or man-made), wireless networks are quickly congested due to the sheer volume and stress on network resources, and, preferential treatment is necessary for National Security/Emergency Preparedness (NS/EP) users to combat the disaster by responding effectively and potentially save many lives. Under such circumstances, with scarce resources, the new request for sessions are denied and worse even, active sessions are dropped for general public whilst they have come to rely on these resources and depend on them especially during distressed times. Prior research has been conducted to examine upper limit (UL) and preemptive approaches to support emergency users but the traditional approach of blocking the capacity for emergency users is, from one perspective, restrictive to the general public. In this thesis, we propose the delay-based soft preemptive approach to support the low priority users and provide an alternative to several preemptive policies by further examining them. We provide a queuing algorithm in the network that warns the low priority users with an active session of scarce resources thereby giving them an opportunity to complete their session prior to reducing the quality of service (QoS) of their session and moving their bandwidth to emergency users, if blocked. The emergency users in turn wait for the resources to become available and are on hold until resources become available. By creating a queuing modeling system for this algorithm, we present simulation model in C with results of our delay-based soft preemptive approach and examine other preemptive approaches to provide a comparative analysis. The results demonstrate that increasing the warning time also increases the number of sessions blocked for emergency users as well as general public due to further constraining the resources, however, this reduces the inconvenience of preemption caused to the low priority users.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Electrical Engineering (UMKC)

Grantor dc:publisher

University of Missouri--Kansas City

Year dc:date.issued

2012

Author dc:creator

• Chandra, Aashish

Advisor dc:contributor.advisor

• Beard, Cory

Language dc:language.iso

en_US

Handle dc:identifier.uri

http://hdl.handle.net/10355/12566

OAI identifier oai:identifier

oai:mospace.umsystem.edu:10355/12566