A Cognitive Approach to Network Monitoring in Heterogeneous Environments

Abstract— Introducing intelligence by means of cognition for managing, protecting, processing, and delivering of information in mobile communication systems is the way towards ubiquitous, converged and secure communications. In this context, this paper introduces the concept of quality of information (QoI). QoI means QoS while all the requirements for dependability, security, privacy and trust are satisfied at the highest possible level.

This work proposes and describes an approach to network monitoring in a heterogeneous communication environment based on use of cognitive techniques and learning predictive algorithms (e.g., fuzzy logic). These methodologies are used to create an autonomy in the decision making process that is based on the calculation of key performance indicators (KPIs), which in their turn would trigger the needed radio resource management algorithms. The expected output is an improved network performance in terms of maximized throughput and faster accessibility to services, minimized transport delay, improved network coverage and simplified security management. This is achieved by introducing an intelligent functionality that entails the use of cognitive learning algorithms for efficient resource allocation, provisioning of network resources or for detection of security violations into the traditional network monitoring approach. The paper describes the cognitive monitoring architecture, the required physical and logical entities, and their functionalities. Further, the paper presents an implementation approach for the described architecture.