A Hybrid Approach Combining SAW and TOPSIS for Network Interface Selection in Heterogeneous Wireless Networks

Network interface selection (NIS) is one of the main challenges of future dense and heterogeneous wireless networks, including 5G and beyond. In such cases, achieving always best connected services is challenging and might result in frequent/unnecessary handovers as well as the rank reversal phenomenon (RRP). Multi-criteria decision making (MCDM) techniques are widely utilized to describe and tackle the NIS issue in heterogeneous network environments. In this paper, we propose a novel hybrid MCDM approach that combines simple additive weighting (SAW) and techniques for order preference by similarity for an ideal solution (TOPSIS) to benefit from their complementary advantages. In the proposed approach, considering four various service types, including conversational, streaming, interactive, and background services, the fuzzy analytic hierarchy process is used to weigh the criteria (i.e., delay, jitter, bandwidth, packet loss, and monetary cost). Then, SAW and TOPSIS are combined in a novel way, aiming to offer a NIS approach that outperforms previous works in terms of handover and rank reversal efficiency. Furthermore, user preferences on networking alternatives for different services are taken into account while ranking the candidate network interfaces, making the proposed NIS approach a quality of experience-aware one. Extensive simulations demonstrate that the proposed hybrid NIS approach enables the user to select the best network interface while significantly reducing the handover frequency, unnecessary handovers, and the RRP ratio.