Visualization of Self-Organizing Networks Operated by the ANR Algorithm

• Main Authors: Ho Van Quan, Astrom, T., Jern, M., Moe, J., Gunnarsson, F., Kallin, H.
• Format: Conference Proceeding
• Language: English
• Subjects: Computer network management; Computer networks; Costs; Data analysis; Data visualization; Geospatial; Geovisual analytics; Information analysis; Large scale data; Large-scale systems; Multi-dimensional; Multi-layer; Network visualization; Radio network; Self-organizing network; Self-organizing networks; Time-varying; Visual analytics
• DOI: 10.1109/RIVF.2009.5174664

Cellular radio networks are continually growing in both node count and complexity. It therefore becomes more and more difficult to manage the networks and necessary to use time and cost effective automatic computer algorithms to organize the network's neighbor cell relations. Ericsson has developed such an algorithm, called automatic neighbor relations (ANR), which solves a part of this problem by automatically creating and updating neighbor cell relation (NCR) lists, based on measured network data. Network operators need to have an understanding of the algorithm and of its reliability and performance, which is not easily perceived. This paper presents a visualization tool that visualizes the performance of ANR and gives the user a possibility to control it via policies. The tool allows the operators to follow the evolution of the network and to find problems occurring over time. In addition, it also supports finding potential problems that can occur in the future. The tool was evaluated by a group of relevant domain users and the results from the evaluation were highly positive.