Dependency graph to improve notifications’ semantic on anomaly detection

Besides identifying anomalies, detection systems must offer additional information about the occurrence, aiming to help the network administrator in order to build an accurate diagnostic. This paper presents a lightweight approach to detect anomalies, improving the semantic power of notifications sent to network administrator. The key point of the proposed anomaly detection system is a correlation system based on a directed graph which represents the possible paths of anomaly propagation through the SNMP objects in a network element. The results obtained from initial tests were encouraging and showed that our system is able to detect anomalies on the monitored network element, avoiding the high false alarms rate.