Nokia Siemens Networks Oy

Nokia Corporation

Nokia Solutions and Networks Oy

Kimmo Hätönen

Markus Juhani Miettinen

Albert Höglund

Antti Sorvari

Sasu Tarkoma

Jyrki Berg

Sampo Kristian Torikka

Kari Kulmala

Pekka Kumpulainen

Pekko Vehviläinen

Devices, methods and computer program products for managing quality of monitoring models for collection of performance and/or sensing data are disclosed. A communication network node device sets up a quality of monitoring function for managing a quality of monitoring, QoM, model that defines QoM classes for processing and compressing a performance and/or sensing data stream via selection of data to be included in the performance and/or sensing data stream. The communication network node device causes a first message notifying about the set up QMF to be transmitted towards at least one first target rendezvous point, RP, in the communication network, each RP being configured to aggregate at least one of performance and/or sensing data streams or QoM models.

Managing quality of monitoring models for collection of performance and/or sensing data

The present invention relates to a method and apparatus for compressing a log record information provided e.g. to a monitoring system. Frequent patterns in the log information are detected and, then, redundant frequent patterns whose value or record combination is a subset of a value or record combination of another one of said detected frequent patterns are removed. Accordingly, a general method is provided which can be applied to all event logs arising from communication networks or other monitoring systems. The detection of frequent patterns is only based on value or record combinations and is thus independent of the specific application. On the other hand, the size of the stored log files can be decreased remarkably while remaining readable for human beings.

Method and apparatus for compressing log record information

A method for teaching an anomaly detecting mechanism in a system comprising observable elements (), at least one of which has a periodic time-dependent behaviour, the anomaly detecting mechanism comprising a computerized learning mechanism (). The method comprises assembling indicators () indicating the behaviour of the elements () and arranging the assembled indicators such that each observable element's indicators are assigned to the same input data component. The learning mechanism () is taught so that the input data of the learning mechanism comprises the input data components which are based on the assembled indicators (). Points which approximate the input data are placed in the input space. A presentation of time (-) is incorporated into at least one input data component wherein the presentation of time is periodic, continuous and un-ambiguous within the period of the at least one element with periodic time-dependent behaviour.

Anomaly detection system and a method of teaching it

A method and system for monitoring the behavior of at least one observable object, e.g. a network element, of a network, wherein at least one parameter of the observable object is repeatedly detected. An actually detected parameter is input to a learning process and to an analyzing process, wherein the learning process forms a reference, based on at least two detected parameter values, for describing the behavior of the observable object. The analyzing process compares the input parameter and the reference for detecting an anomal behavior. The parameter preferably is a vector which includes several values describing properties or functioning of the observable object, and is formed based on events and/or reports from the object.

System, device and method for automatic anomaly detection

A data processing system processes data arrays that collectively describe cyclic behavior of at least one variable in several entities in a physical process. Each cycle comprises several time slots. An input routine (-) receives multiple data arrays, each data array containing multiple data items, each of which describes a variable of an entity in one time slot. A magnitude-determination routine (-) determines a specific magnitude parameter, such as average, volume or peak, for each of the several entities. A scaling routine (-) scales the data arrays between entities such that the specific magnitude parameters are suppressed and only their shape is maintained. A training routine (-) trains a clustering system with a first plurality of the scaled data arrays, to determine a set of cluster centers. After training, a clustering routine (-) applies a second plurality of the scaled data arrays to the trained clustering system.

Growing community of inventors

Helsinki, Finland

Kimmo Hätönen