The Internet of Things (IoT) refers to variety of smart devices such as smartphones, tablets, and sensors that can interact and exchange of data among devices through the Internet. The diversity of IoT devices and their services have posed a larger range requirements of availability, throughput, latency, and performance in heterogeneous connectivity environments. Meanwhile, the existing networks often struggle with such of limitations in complex control protocols and difficulty in internetworking with billions of smart devices with different requirements such as latency and bandwidth allocations. These obstacles become substantial barriers to deploy services, as well as isolate between multiple co-existing tenants on the same physical network, deploy simultaneous protocols in the network, be stable to maintain the bandwidth and latency according to predefined QoS demands. These obstacles have recently been facilitated by Software Defined Network (SDN) and Network Function Virtualization (NFV) technologies that enable the programming and monitoring in data plane. In this study, firstly, the authors investigate and propose a SDN/NFV based architecture for multi-tenant networks with plenty of network slices working in a shared physical infrastructure. Secondly, P4 and ONOS Controller are used to implement a deep programming in BMv2 devices to efficiently maintain the network motoring in order to guarantee the E2E latency of communicating channels. Finally, the VXLAN technologies are exploited to for network slicing with different purposes and applications, and Inband Network Telemetry (INT) is used to monitor network latency.

In mobile networks, handover (HO) is one of the most important and complex KPIs (Key Performance Indicators), which directly affect to Quality of Service (QoS), Quality of Experience (QoE), and mobility performance. Moreover, its configuration parameters such as handover thresholds and handover neighbor lists are the key factors for implementing network optimization such as load balancing and energy saving. In a study before, the authors proposed clustering and forecasting models using ML algorithms and Time Series models to cluster, forecast, and manage the HO behavior of a huge number of cells. In this study, on the other hand, the authors firstly investigated more network KPIs to analyze their relationship with HO KPIs, and then, proposed new clustering, forecasting, and abnormal detection models that are expected to make them much more comprehensive. Finally, the performances of the proposed models were evaluated by applying them to a real dataset collected from the HO KPIs and other KPIs of more than 6000 cells of a real network during the years, 2016 and 2017. The results showed that the study was successful in identifying the relationship among network KPIs and significantly improving the performance of the HO clustering, forecasting, abnormal detection models. Moreover, the study also introduced the integration of emerging technologies such as machine learning (ML), big data, softwaredefined network (SDN), and network functions virtualization (NFV) to establish a practical and powerful computing platform for future self-organizing networks (SON).

Mobile Edge Computing (MEC) is an emerging technology and an essential component of 5G networks to bring cloud services closer to users. That means data collection, storage, processing, computing, communication, and network control are implemented at network edges. MEC is expected to be able to satisfy a variety of delay-sensitive services and applications. On the other hand, the development of vehicles to everything (V2X) communication brings many requirements to future networks to guarantee full intelligence, automatic, and faster computation, management, and optimization to fulfill network QoS (quality of service) and QoE (quality of experience). To deal with those requirements, recently, softwaredefined networking (SDN), network functions virtualization (NFV), big data, and machine learning (ML) have been proposed as emerging technologies and the necessary tools for MEC and vehicular networks. This study aims to integrate those technologies to build a comprehensive architecture and an experimental framework for future 5G MEC called Open5GMEC. Moreover, the authors analyzed challenges and proposed relevant solutions for future vehicular communications in 5G networks. Finally, based on this framework, we successfully implemented several powerful ML-based applications for V2X such as object detection, network slicing, and migration services, which are executed at Broadband Mobile Lab (BML), National Chiao Tung University (NCTU).