Telefonaktiebolaget Lm Ericsson (Publ)

Paddy Farrell

Liam Fallon

Premnath Kandhasamy Narayanan

Mingxue Wang

James O'Meara

Robin Grindrod

Yujie He

Françoise Sailhan

Ronan Barrett

A method of managing predicting anomalies in operation of a communications network. The method includes receiving network performance data, including network performance data received as time series of values representing monitored characteristics. The method also includes detecting a first anomaly in operation of the communications network and, from historical network performance data, determining if an instance of the first anomaly occurred in the past. If this is a first occurrence of the first anomaly, then based on network performance data received before detecting the first anomaly the method includes building a first model for predicting an instance of the first anomaly and deploying the first model to operate. An apparatus implementing the method is also disclosed.

Method and apparatus for managing prediction of network anomalies

A method of real-time processing of network events, wherein an event comprises a plurality of attributes. The method comprises maintaining counters, including counters indicating numbers of occurrences and co-occurrences of instances of specific attributes of received network events, maintaining a record of instances of specific attributes of recent events, receiving a first event, increasing respective counters based on the instances of attributes of the first event. The method further comprises calculating relation measure scores between a first instance of the first attribute of the first event and instances of the first attribute of events recently received. The relation measure score depends on a number of co-occurrences in which the first instance of the first attribute and a second instance of the first attribute were received within the measurement window, the total number of occurrences of the first instance and the total number of occurrences of the second instance. The method also comprises identifying a group of events as related based on their relation measure scores and creating incident information based on the identified group of events.

Method and device for real-time network event processing

A method of processing event subscriptions in a telecommunications or data network comprising a plurality of nodes organized in clusters, in multiple layers. Each cluster has a cluster leader. The method comprises maintaining by each node a routing table comprising a list of nodes in its cluster and a repository of event subscriptions; forwarding an ascending subscription by a node to all nodes within its cluster and to its cluster's leader; forwarding a descending subscription to all nodes of all layers that the node is a leader and adding a received event subscription to the event subscriptions repository and associating said subscription with a node, which forwarded it.

Method of processing event notifications and event subscriptions

Growing community of inventors

Athlone, Ireland

Paddy Farrell