Quality of Monitoring for Cellular Networks

Abstract: 5G networks and beyond introduces a larger number of Network Elements (NEs) and functions than previous cellular generations. This will result in increasing the NEs Management Plane (M-Plane) data. Thus, the conventional centralized Network Management Systems (NMS) will face fundamental scaling challenges in processing the data. In this paper, we present a novel concept of Quality of Monitoring (QoM), which uses a set of classes for compression of M-plane data at the mobile edge. QoM classes specify the quality by which MPlane data can be collected from the NEs. QoM as a solution aggregates the raw M-Plane data into a smaller set of operatordefined performance metrics. It further compresses the data using a modified version of Piece-Wise Constant Approximation (PWCA) algorithm, which is a lossy-compression technique. PWCA significantly reduces the data size by discarding the portions of data with redundant and less important information. QoM concept is based on publish/subscribe paradigm and uses Apache Kafka for streaming of the compressed data. To study the feasibility of the concept, we use M-Plane data of four KPIs of different NEs from a live LTE network. As result, the QoM as a solution minimizes the amount of M-Plane data by removing redundant and similar data at the network edge. This results in reducingand saving the network resources in data transmission.