Nowadays, networks are at the center of the next industrial revolution. In fact, 5G in a short time will connect people, industries and things, so understanding how the network is performing its critical mission in this new paradigm is a key aspect. Network analytics increases the knowledge of the network and its users, leading the network managers to make smarter, data-driven decisions about the operations that they will execute in the network. In this article, a new methodology is introduced to analyze real data contained in a Call Details Record of a mobile network. With this novel methodology, the extraction of extreme points using the orthogonal projection decrease the complexity of the classification algorithm to obtain key information about network usage. Experimental results show how the proposed methodology selects and classifies network behavior patterns using a simple classification algorithm and how these patterns could be used to find, for instance, anomalies in the network, track human mobility, undertake network planning, detect events in the network, etc.INDEX TERMS 5G, data analysis, mobile networks, network analytic.

No abstract

In recent years, the growth the in the number of heterogeneous interconnected systems, as well as the emergence of new requirements in applications and services are progressively changing the original simplicity and transparency of the Internet architecture. When this architecture was designed, the main goal was to interconnect stationary host. Therefore, the appearance of mobile communications has made necessary to adapt traditional protocols in order to accommodate mobile users. This implies a new interaction between the mobile network and the fixed access network. This paper describes the main IP mobility protocols both centralized and distributed paradigms, and emergent approaches based on software defined networking. Moreover, a novel classification is presented, which relates the integration of the mobility protocol with the access network. Analytical models evaluate the registration updates cost and the packet loss rate of the classified protocols.

Mobile data traffic in the Internet has experienced an exponential growth due to the widespread presence of multimedia capable mobile devices and the deployment of multiple wireless networks. With this continuous development of mobile communications, the achievement of an efficient IP mobility management protocol has revealed as one of the major challenges in next-generation wireless networks. Mobility management solutions are responsible for maintaining the ongoing communications while the user roams among distinct networks. Mobile IPv6 and Proxy Mobile IPv6 are the most representative solutions standardized by the IETF. Recently, the IPv6 mobility support has been newly integrated into the kernel sources and Linux mobility ready kernels are available from versions 3.8.1. In this paper, we conduct an analytic and experimental evaluation of Mobile IPv6 and Proxy Mobile IPv6. We develop an analytic model of the signaling and handover latency. Moreover, we present an experimental study of these protocols based on their open source implementations. We provide numerical results based on experiments made in real scenarios under different network conditions.

Cyber-physical systems allow creating new applications and services which will bring people, data, processes, and things together. The network is the backbone that interconnects this new paradigm, especially 5G networks that will expand the coverage, reduce the latency, and enhance the data rate. In this sense, network analytics will increase the knowledge about the network and its interconnected devices, being a key feature especially with the increment in the number of physical things (sensors, actuators, smartphones, tablets, and so on). With this increment, the usage of online networking services and applications will grow, and network operators require to detect and analyze all issues related to the network. In this article, a methodology to analyze real network information provided by a network operator and acquire knowledge of the communications is presented. Various real data sets, provided by Telecom Italia, are analyzed to compare two different zones: one located in the urban area of Milan, Italy, and its surroundings, and the second in the province of Trento, Italy. These data sets describe different areas and shapes that cover a metropolitan area in the first case and a mainly rural area in the second case, which implies that these areas will have different comportments. To compare these comportments and group them in a single cluster set, a new technique is presented in this paper to establish a relationship between them and reduce those that could be similar.

With the continuous development of mobile communications and the Internet of Things technology, the enhanced network performance can be seen as one of the major challenges in the fourth industrial revolution context, where new services and applications with strict performance requirements have emerged, such as driverless vehicles, smart cities, factories, and manufacturing, among others. These new services and applications also drive growth of the data traffic, which is increasing exponentially. Thus, in mobile network environments, industry and academia are proposing new mechanisms to overcome the traffic bottlenecks and reduce the signaling overhead that affects current networks. Centralized Mobility Management solutions are prone to several problems such as the aforementioned signaling overhead or scalability issues. To overcome these limitations, Distributed Mobility Management approaches are being considered. In this paper, an analytical cost model and experimental evaluation will be developed for evaluating the performance of the Distributed Mobility Management implementations. Furthermore, a new approach will be proposed to improve network performance.

No abstract

No abstract