Optimizing soft handoff latency and access network selection for next generation mobile communication networks

In this contribution, a cognitive quality of service (QoS) mechanism is proposed for improved handoff decisions in mobile communication networks. The proposed design is a hybrid channel perception tool that trains observed key performance indicators (or channel characteristics) to discover the system's channel behavior. A Hidden Markov Model (HMM) framework is adopted to ensure that the available channels are not completely consumed during handoffs by allocating guard channels, and using a set of regulated threshold values, the transmission and emission matrices for learning the network, were constructed. Estimates obtained from the training are then passed to a Fuzzy-based model to select the appropriate access network or base station. Simulation results prove that the proposed system is able to sustain handoffs at an average of 80%, compared to the empirical system, which average stood at 50%. Since the mobile terminal of next generation wireless networks has multiple interfaces for connection to any best connected access technology for handoff execution, two multiple attribute decision making (MADM) methods: fuzzy analytic hierarchy process and synthetic extent analysis were used to assign weights as triangular fuzzy numbers to certain criteria that influence handoff decision, and rank the alternative access networks based on weight priority scores in order to select the most optimal access network for a particular application. A future direction of this research include a prioritization of the various decision variables to improve on the overall network performance, and the provision of efficient control mechanism to deal with the integration of newer technologies deployed for optimized environments.