Title :
DCAD: Dynamic Cell Anomaly Detection for operational cellular networks
Author :
Ciocarlie, Gabriela ; Lindqvist, Ulf ; Nitz, Kenneth ; Novaczki, Szabolcs ; Sanneck, Henning
Author_Institution :
SRI Int., Menlo Park, CA, USA
Abstract :
The Self-Organizing Networks (SON) concept includes the functional area known as self-healing, which aims to automate the detection and diagnosis of, and recovery from, network degradations and outages. In this paper, we present Dynamic Cell Anomaly Detection (DCAD), a tool that implements an adaptive ensemble method for modeling cell behavior [5], [6]. DCAD uses Key Performance Indicators (KPIs) from real cellular networks to determine cell-performance status; enables KPI data exploration; visualizes anomalies; reduces the time required for successful detection of anomalies; and accepts user input.
Keywords :
cellular radio; cryptography; data visualisation; self-organising feature maps; telecommunication computing; telecommunication security; DCAD; KPI data exploration; SON; adaptive ensemble method; anomaly visualization; cell behavior modeling; cell-performance status determination; dynamic cell anomaly detection; key performance indicators; network degradation detection automation; network degradation diagnosis automation; network degradation recovery automation; network outage detection automation; network outage diagnosis automation; network outage recovery automation; operational cellular networks; self-healing functional area; self-organizing networks; time requirement reduction; user input acceptance; Adaptation models; Computer architecture; Data models; Degradation; Microprocessors; Self-organizing networks; Support vector machines; Key Performance Indicators (KPIs); Self-Healing; Self-Organizing Networks (SON); cell anomaly detection; performance management;
Conference_Titel :
Network Operations and Management Symposium (NOMS), 2014 IEEE
Conference_Location :
Krakow
DOI :
10.1109/NOMS.2014.6838271
