Anomaly Detection and Diagnosis for Automatic Radio Network Verification

Ciocarlie, Gabriela F.; Connolly, Christopher I.; Cheng, Chih‐Chieh; Lindqvist, Ulf; Nováczki, Szabolcs; Sanneck, Henning; Naseer-Ul-Islam, Muhammad

doi:10.1007/978-3-319-16292-8_12

Cited by 19 publications

(19 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In previous work [5], we proposed a novel SON verification framework that uses anomaly detection and diagnosis techniques and operates within a certain spatial scope larger than an individual cell, e.g., a small group of cells or cell cluster being in scope for a SON optimization, an existing administrative network domain, etc. Key Performance Indicators (KPIs) are continuously collected from network cells and used as indicators (e.g., call-drop statistics, channel-qualityindicator statistics, handover statistics, throughput, etc.)…”

Section: A Son Verificationmentioning

confidence: 99%

“…With the previous approach [5], clustering is first applied for the largest scope (the largest number of cells) and then, the state of the network can be determined for subsets of the largest number of cells. However, in a real deployment, if the scope needs to be enlarged (e.g., when new cells are added to the network), we would require an incremental approach for topic modeling, which will gradually update the clusters with information from the larger scope.…”

Section: Network-level Anomaly Detection and Scope Changesmentioning

confidence: 99%

“…A second component of the anomaly detection framework [5] was an interpretation module that could characterize topic model clusters as either normal or abnormal. This simple classifier used the semantics of the KPI to generate labels; consequently, classification was achieved only for KPIs that were supposed to maintain a low level (e.g., drop call rate) or a high level (e.g., call success rate).…”

Section: A Son Verificationmentioning

confidence: 99%

“…for network performance. The first component of the SON verification framework [5] was the network-level anomaly detection component, which aimed to detect anomalies within a given scope of analysis (at the level of either a network or group of cells). The anomaly detection framework used a multi-level clustering approach in which KPI data was first clustered to form different states in which the network could be, and to express the overall state of the network as a combination of states that correspond to different parts of the network.…”

Section: A Son Verificationmentioning

confidence: 99%

See 3 more Smart Citations

Managing scope changes for cellular network-level anomaly detection

Ciocarlie

Cheng

Connolly

et al. 2014

2014 11th International Symposium on Wireless Communications Systems (ISWCS)

Self Cite

View full text Add to dashboard Cite

The Self-Organizing Networks (SON) concept is increasingly being used as an approach for managing complex, dynamic mobile radio networks. In this paper we focus on the verification component of SON, which is the ability to automatically detect problems such as performance degradation or network instability stemming from configuration management changes. In previous work, we have shown how Key Performance Indicators (KPIs) that are continuously collected from network cells can be used in an anomaly detection framework to characterize the state of the network. In this study, we introduce new methods designed to handle scope changes. Such changes can include the addition of new KPIs or cells in the network, or even re-scoping the analysis from the level of a cell or group of cells to the network level. Our results, generated using real cellular network data, suggest that the proposed network-level anomaly detection can adapt to such changes in scope and accurately identify different network states based on all types of available KPIs.

show abstract

Section: A Son Verificationmentioning

confidence: 99%

Section: Network-level Anomaly Detection and Scope Changesmentioning

confidence: 99%

Section: A Son Verificationmentioning

confidence: 99%

Section: A Son Verificationmentioning

confidence: 99%

See 2 more Smart Citations

Managing scope changes for cellular network-level anomaly detection

Ciocarlie

Cheng

Connolly

et al. 2014

2014 11th International Symposium on Wireless Communications Systems (ISWCS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Good examples of recent practical approaches to SH in real operational networks are found in [10,11,12,13,14,15]. For example, [10] addressed the problem of verifying the effect of network configuration changes by monitoring the state of the network and determining if the changes resulted in degradations.…”

Section: Developments In Sh Researchmentioning

confidence: 99%

Towards proactive context-aware self-healing for 5G networks

et al. 2017

View full text Add to dashboard Cite

In this paper, we suggest a new research direction and a future vision for Self-Healing (SH) in Self-Organizing Networks (SONs). The problem we wish to solve is that traditional SH solutions may not be sufficient for the future needs of cellular network management because of their reactive nature, i.e., they start recovering after detecting already occurred faults instead of preparing for possible future faults in a pre-emptive manner. The detection delays are especially problematic with regard to the zero latency requirements of 5G networks. To address this problem, existing SONs need to be upgraded from reactive to proactive response. One of the dimensions in SH research is to employ more holistic context information that includes, e.g., user location and mobility information, in addition to traditional context information mostly gathered from sources inside the network. Such extra information has already been found useful in SH. In this paper, we suggest how user context information can not only be incorporated in SH but also how future context could be predicted based on currently available information. We present a user mobility case study as an example to illustrate our idea.

show abstract