Efficient Vertical Handover Strategies in Heterogeneous Networks using Heuristic Approaches

Abstract

A heterogeneous network incorporates cellular networks, Wi-Fi, WLAN, etc. to deliver an adaptable and globalized wireless network infrastructure. This network significantly expands the service coverage area of the conventional mobile network, enhances network utilization, and reduces packet delivery delay due to the presence of various kinds of radio access technology. However seamless communication remains an important challenge in heterogeneous networks. In such networks, a vertical handover process plays an important role in providing seamless and uninterrupted connectivity to all mobile nodes. Several vertical handover decision (VHD) algorithms are found in the literature that consider a variety of network parameters to take the handover decision. VHD algorithms have become more challenging to implement due to the complex nature of heterogeneous networks and the diversification of system functionalities. As a result, designing an effective and reliable VHO algorithm should ensure improved network resource utilization as well as a desirable QoS. Thus, developing a multi-criteria-based vertical handover decision algorithm, enabling an optimal network scanning method, establishing an adaptive hybrid channel allocation mechanism during handover, incorporating all requisite contextual information in the decision algorithm, and optimizing the handover decision criterion using a machine learning algorithm are the objectives of the thesis. An optimized multicriteria-based VHD algorithm is proposed for a heterogeneous network. This algorithm uses a two-step execution procedure, the first step evaluates whether the handover is needed or not before initiating a handover process, if needed, then a set of feasible solutions are generated for the vertical handover problem formulated as a multi-objective optimization problem with two objective functions such as quality factor of networks and handover processing cost. In the second step, an optimal solution (i.e., an appropriate target network) for handover is obtained using a hybrid approach of the Fuzzy Analytic Hierarchy Process (FAHP) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The simulation study shows that the proposed algorithm performs well as compared to some other existing algorithms in terms of performance metrics such as handover rate, delay, cost, and energy consumption. An energy-efficient network scanning algorithm is proposed using media media-independent handover (MIH) standard. MIH facilitates seamless and energy-efficient handover of mobile nodes across heterogeneous networks by avoiding network scanning. But sometimes scanning avoidance leads to inconsistent handover and causes increased handover failure rates. So, the proposed work introduces an optimal network scanning procedure by incorporates two additional functional units into MIH framework i.e., the Energy Consumption Unit (ECU) and the Handover Decision-making Unit (HDU). The objective of ECU is to limit the amount of network scans, whereas HDU is responsible for computing handover decisions by taking both network conditions and user preferences into account. An adaptive hybrid channel allocation mechanism (Ad-HCA) is proposed for vertical handover in heterogeneous networks. his study introduces a new hybrid channel allocation mechanism that facilitates the effective use of both fixed and dynamic channels in a highly complex and dynamic environment. The suggested mechanism adaptively changes the size of guard channels in accordance with the current HO blocking rate and provides a better percentage of channel utilization with excellent QoS. This study shows that the overall utilization of the band is improved when priority is given to the handover calls over new calls. Furthermore, borrowed channels are re-assigned to handover calls before returning them to the central pool whenever necessary. The proposed algorithm reduces HO as well as new call blocking probabilities by maintaining it to below 0.01 and 0.02 respectively. It maximizes the channel utilization rate up to 95% approximately. A context-aware VHD algorithm is proposed by incorporating all the requisite contextual information related to users, networks, and application details that can adapt to the varying nature of diversified networks having different specifications and requirements. The suggested algorithm undoubtedly assures that the requested handover call does not have to wait unnecessarily for resources in the system or depart unserviced due to its mobility. Furthermore, a utility function-based cell load balancing mechanism is proposed to improve the overall resource utilization of the cell. The simulation results show that the proposed algorithm significantly reduces the call-blocking probability of handover calls as well as total calls below 1.5% by maximizing the channel utilization rate. It also reduces repeated and unnecessary handovers that occur due to the users’ high and nonuniform mobility in a heterogeneous network with different access technologies. An intelligent handover method is developed that jointly uses a hybrid MCDM approach for handover decision-making and a Q-learning strategy to determine the optimal handover triggering point. Instead of using static handover triggering values, the proposed algorithm optimizes those values by adopting the fluctuations of the environment through the Q-learning approach. Here the optimal value of triggering points is learned from the environments based on various user velocities. This work outperforms other existing algorithms by reducing handover ping-pong effects, delays, and handover failure rates by an average of 55%, 33%, and 45% respectively.