Quality of experience modeling and fair scheduling for instant messaging service

Instant Messaging (IM) service has drawn growing attention in current wireless networks. However, the Quality of Experience (QoE) of IM service users are instable, since the IM service is not served with dedicated channel or guaranteed bandwidth. In this paper, we adopt a Representative Service (RS) based QoE model for IM service, and propose a scheduling algorithm to provide fair QoE to IM service users. In the QoE model, the subservice that a user is currently focusing on is defined as the RS and the quality of RS, Representative Service Quality (RSQ), is considered as the QoE of an IM service user. In our scheduling algorithm, RS is estimated when several subservices are launched simultaneously, and resource allocation is done accordingly to avoid the QoE degradation due to resource insufficiency. Simulation results validate the rationality and effectiveness of the proposed algorithm, show that it outperforms classical fair scheduling algorithms.