Data Analytics for Traffic Engineering Sample Clauses

Data Analytics for Traffic Engineering. According to the NGMN white paper [NGM15], the end users will have access to a diverse set of services (ranging from high definition video and audio, web browsing, games, to keep alive messages etc.). For the provided services (e.g., augmented reality, cloud services, car to car communication, etc.) connectivity everywhere (including indoor and outdoor environments, environments with ultra-dense or limited infrastructure, or where the environment is extremely crowded, etc.) to users with various mobility shall be provided [NGM15]. Thus, the network needs to meet at the same time the data rate demanding services (such as high definition video and audio) and the low demanding ones (e.g., keep alive messages) which may however involve millions of devices accessing simultaneously the network; all the user equipment’s may have diverse mobilities. In such demanding environments the use of context information is imperative for enabling the efficient use of the network resources [MSC13]. Also according to [MSC13] the term context refers to “all the information that may be exchanged among the network entities, even heterogeneous ones so as to solve challenging networking problems such as management and control of the network resources”. This generic definition of the context information implies that all information types may be used for the optimization of the network management and control such as radio measurements, mobility information/measurements (e.g., user speed, direction, etc.), etc. Several Context Aware mechanisms are available in the literature. Such schemes take advantage of e.g.,  Location information [XPM+14] [CWH+13],  Mobility information [KKP+13],  Network statistics (access or backhaul) [QXY+10] [MBS+10],  Current Service information [XPM+14] [MCC+11]. Combination of the abovementioned parameters can improve the accuracy of the mechanisms. In general such combinations are based on objective functions and/or rule based systems, which however try to combine inputs not clearly linked to each other. The solution presented in [TP10] is an example of such a scheme. However:  In these context-aware systems the decisions require real time information related to a significant number of parameters  Even enhanced with state optimization, the rule based systems cannot address the “Curse of Dimensionality” ([SRG+12]) efficiently and in real time [MKS+10]  Transferring all these inputs to serve every radio resource management and control mechanism for a...