Abstract: An amplifier for a cable network includes circuitry configured to provide a diagnostic parameter. Exemplary parameter include one or more of: measurements of a frequency response, suckouts in the frequency domain, north upstream echo signal or power, north downstream signal or power, north upstream echo to downstream power ratio, north upstream echo to residual echo power ratio, downstream echo canceller coefficients, south downstream signal or power, upstream information, south downstream echo signal or power, south upstream signal or power, south downstream echo to upstream power ratio, south downstream echo to residual echo power ratio, upstream echo canceller coefficients, and north upstream signal or power.

Abstract: Embodiments may include techniques for hierarchical evaluation of the anomaly level of a system and its sub-components using domain knowledge, so as to provide improved accuracy and explainability compared to conventional methods. For example a method of anomaly detection in a hierarchical computer network may comprise defining a tree-like topological structure which describes how the hierarchical computer network comprises sub-components, wherein each node of the tree-like topological structure represents a sub-component of the hierarchical computer network, and wherein at least some of the sub-components are monitored to generate signals indicating an operational condition of each sub-component, collecting a plurality of time-series of maximum absolute anomaly scores for each monitored signal, and computing an anomaly score for a root node of the tree-like topological structure.

Abstract: Embodiments may include techniques for hierarchical evaluation of the anomaly level of a system and its sub-components using domain knowledge, so as to provide improved accuracy and explainability compared to conventional methods. For example a method of anomaly detection in a hierarchical computer network may comprise defining a tree-like topological structure which describes how the hierarchical computer network comprises sub-components, wherein each node of the tree-like topological structure represents a sub-component of the hierarchical computer network, and wherein at least some of the sub-components are monitored to generate signals indicating an operational condition of each sub-component, collecting a plurality of time-series of maximum absolute anomaly scores for each monitored signal, and computing an anomaly score for a root node of the tree-like topological structure.

Abstract: There is provided a system for identifying a system level anomalous, comprising code for receiving data-points indicative of a time-based dynamic state of processing nodes of a distributed system, each associated with a timestamp, a metric, and a metric value, clustering the data-points into clusters, comprising a sub-set of the data-points having a same metric and a timestamp within a same metric anomaly time interval, computing metric anomaly scores for each respective cluster denoting likelihood of an anomaly being present, analyzing metric anomaly scores of each system level anomalous time interval including at least two of the metric anomaly time intervals, computing a single anomalous event score for each respective system level anomalous time interval according to the analysis, and generating an alert indicative of the system level anomalous event for a certain system level anomalous time interval when the single anomalous event score is according to an anomaly requirement.

Abstract: There is provided a system for identifying a system level anomalous, comprising code for receiving data-points indicative of a time-based dynamic state of processing nodes of a distributed system, each associated with a timestamp, a metric, and a metric value, clustering the data-points into clusters, comprising a sub-set of the data-points having a same metric and a timestamp within a same metric anomaly time interval, computing metric anomaly scores for each respective cluster denoting likelihood of an anomaly being present, analyzing metric anomaly scores of each system level anomalous time interval including at least two of the metric anomaly time intervals, computing a single anomalous event score for each respective system level anomalous time interval according to the analysis, and generating an alert indicative of the system level anomalous event for a certain system level anomalous time interval when the single anomalous event score is according to an anomaly requirement.