| Summary: | Next Generation Wireless Networks (NGWNs) are envisaged to be heterogeneous. They will integrate multiple Radio Access Technologies (RATs) on the same platform where the RAT s are expected to work together in a coordinated fashion. Radio resources will therefore be jointly managed. Joint Call Admission Control (JCAC) approach has been adopted to jointly allocate or handle the radio resources amongst the different RATs in an NGWN. A number of JCAC algorithms have been proposed in the literature with the objective of enabling efficient utilisation of the radio resources and ensuring that required QoS requirements are met . However, studies in JCAC have not typically considered the revenue obtained by network operators in NGWNs . With the implementation of NGWNs, user demand for network services is expected to significantly increase. Consequently, one of the challenges envisaged in the NGWNs for network operators is a large decoupling between users' traffic and the revenue obtained . Therefore, it is important to analyse the network operators' revenue in NGWN s and devise means to ensure that sufficient revenue is obtained. This research analyses network operators' revenue in a heterogeneous network environment while maintaining QoS by incorporating pricing in JCAC. An analytical model based on a multi-dimensional Markov decision process is used to model JCAC algorithm in an NGWN. The JCAC algorithm used is load-based whereby calls arriving in the network are admitted to the least loaded RATs. The performance of the algorithm is evaluated using MATLAB. The analysed NGWN consists of two RATs and two service classes. QoS performance is measured with connection-level QoS metrics namely call blocking probability and call dropping probability. Other investigated performance metrics are average number of calls in the systems, user utility and revenue obtained. Performance of the NGWN is carried out with a dynamic pricing scheme incorporated in JCAC. A discount approach is applied to determine new price under the dynamic pricing scheme. User behaviour with price variation is impacted by change of call arrival rate and average call duration. Simulations are also carried out with flat pricing for comparison purposes. The results obtained show that better system performance and high revenue for network operators are obtained with the dynamic pricing scheme. Furthermore, a hybrid pricing model is proposed whereby flat and dynamic pricing schemes are integrated, giving users flexibility of choosing an appropriate pricing scheme. The analysis is conducted with different percentages of users opting for either dynamic or flat pricing schemes. Simulation results show that higher user utility and improved QoS performance are obtained with the hybrid pricing model. Additionally, reasonable operators' revenue is guaranteed with the hybrid pricing model.
|
|---|
