Multi-Criteria Handover in SDN-Based Multi-RAT Networks

Beyond 5G (B5G) networks are expected to be dense, heterogeneous wireless networks that coexist within a wide range of Radio Access Technologies (RATs). Conventional handover decision mechanisms do not capture the characteristic, parametric, and performance differences associated with each coexisting RAT in the multi-RAT network. Furthermore, a single base station does not provide the management processes required to organize the cross-RAT handover decision. Software-defined networking (SDN) has the potential to be effective for Multi-RAT handover algorithm implementation. In this work, we propose a Multiple Criteria Decision-Making (MCDM) based handover algorithm in SDN-based dense multi-RAT networks. The handover decision involves the calculation of entropy-based weights for each decision criterion. The candidate nodes are ranked using the Technique for Order of Preference by Similarity to the Ideal Solution (TOPSIS), which calculates the similarity of each alternative to the ideal possible and the worst possible alter-native and ranks them accordingly. The chosen case study uses Mininet WiFi to implement an SDN-controlled 5GIWIFi Multi-RAT environment and uses MCDM-based handover to optimize both RSSI and delay. Results show that the proposed MCDM-based handover improves the system throughput compared to sole RSSI-based handover, in all tested cases, by at least 18%, and there is at least 43 % reduction of handover failure.