Enhanced handover mechanism using mobility prediction in wireless networks

RESEARCH ARTICLE Enhanced handover mechanism using mobility prediction in wireless networks Khong-Lim Yap1☯, Yung-Wey Chong ID1☯*, Weixia Liu2☯ 1 National Advanced IPv6 Centre, Universiti Sains Malaysia, USM, Penang, Malaysia, 2 Internet Innovation Research Center, Minjiang University, Fuzhou, China ☯ These authors contributed equally to this work. * a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Yap K-L, Chong Y-W, Liu W (2020) Enhanced handover mechanism using mobility prediction in wireless networks. PLoS ONE 15(1): e0227982. https://doi.org/10.1371/journal. pone.0227982 Editor: He Debiao, Wuhan University, CHINA Received: September 29, 2019 Accepted: January 4, 2020 Published: January 24, 2020 Copyright: © 2020 Yap et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Abstract The rapid increase in the usage of the mobile internet has led to a great expansion of cellular data networks in order to provide better quality of service. However, the cost to expand the cellular network is high. One of the solutions to provide affordable wireless connectivity is the deployment of a WiFi access point to offload users’ data usage. Nevertheless, the frequent and inefficient handover process between the WiFi AP and cellular network, especially when the mobile device is on the go, may degrade the network performance. Mobile devices do not have the intelligence to select the optimal network to enhance the quality of service (QoS). This paper presents an enhanced handover mechanism using mobility prediction (eHMP) to assist mobile devices in the handover process so that users can experience seamless connectivity. eHMP is tested in two wireless architectures, homogeneous and heterogeneous networks. The network performance significantly improved when eHMP is used in a homogeneous network, where the network throughput increases by 106% and the rate of retransmission decreases by 85%. When eHMP is used in a heterogeneous network, the network throughput increases by 55% and the retransmission rate decreases by 75%. The findings presented in this paper reveal that mobility prediction coupled with the multipath protocol can improve the QoS for mobile devices. These results will contribute to a better understanding of how the network service provider can offload traffic to the WiFi network without experiencing performance degradation. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Funding: This research has been supported by Digi Telecommunications Sdn Bhd under grant 304/ PNAV/650764/D113 to Y-WC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: This research has been supported by Digi Telecommunications Sdn Bhd. The support from Digi Telecommunications Sdn Introduction The growth in mobile and wireless networks has disrupted the way that networks were designed. A recent report showed that mobile data usage grew 63% in 2016 and that it will surpass 49 exabytes by 2021 [1]. Smart phones have eclipsed personal computers, and the global population of over 7.4 billion wireless users continues to consume an increasing amount of spectrum resources. Cloud-based services and video traffic as well as essential online services such as e-banking, e-learning and e-health continue to proliferate, causing the wireless traffic volume to grow exponentially. PLOS ONE | https://doi.org/10.1371/journal.pone.0227982 January 24, 2020 1 / 31 Enhanced handover mechanism using mobility prediction in wireless networks Bhd does not alter our adherence to PLOS ONE policies on sharing data and materials. Users’ expectations also change, and the demand for “anytime, anywhere” connections has become a necessity to satisfy their needs. The good news is that mobile devices are armed with multiple radio capabilities. However, current mobile devices do not have the intelligence to communicate via multiple wireless networks at the same time for improved capacity, coverage and seamless handover. There are no intelligent and autonomous mobile devices that are able to find and connect to the best radio networks that meet users’ needs. Users continue to experience intermittent connectivity and inconsistent throughput, and latency can be extremely unpredictable even when multiple radio networks are available. Since mobile devices cannot predict and optimize the handover process, the quality of service (QoS) for mobile users degrades further, especially when mobile users are on the move. To improve the QoS, cellular service providers are integrating picocell networks to offer an economically appealing way to improve the coverage, QoS and network capacity. In addition, WiFi is used to offload the cellular network to create a heterogeneous network, allowing users to enjoy internet connectivity everywhere and at anytime. The emergence of WiFi has reduced the need to depend on the cellular network, especially when users are constantly moving inside a building. The provisioning of QoS has become more challenging due to the frequent and unnecessary handover process, either between WiFi and WiFi or WiFi and the cellular network. The mobile handover process can be divided into three (3) stages, namely, handover initiation, network selection and handover execution [2]. Existing mobile devices will initiate the handover process when the connectivity between the mobile device and the connected access point (AP) or base station (BS) drops below a certain threshold of the received signal strength indicator (RSSI) level. The device will then select a network based on the RSSI because it is the simplest measurement even when the throughput is not optimal. This approach often causes data service session interruption and decreases users’ experience. Currently, mobile devices do not have the capability to connect to an optimal WiFi AP for a better QoS. This issue can indirectly cause unnecessary handover between different WiFi APs, eventually leading to service disruption whenever handover occurs. Unnecessary handover is also caused by the dense deployment of WiFi APs to cover a broad area. One of the commonly used approaches to mitigate the frequent handover issue is mobility prediction. [3] proposed a mobility prediction model based on the hidden Markov model (HMM) to predict the next service eNodeB in the long-term evolution (LTE) cellular network. Mobility prediction-based approaches show promising solutions mainly because human mobility behaviour is far from random and is highly influenced by their historical behaviour. Although mobility prediction can determine the future location of users, improve the handover process, and manage the resources e (...truncated)