Edinburgh Research Explorer

Kullback-Leibler Divergence (KLD) Based Anomaly Detection and Monotonic Sequence Analysis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Related Edinburgh Organisations

Original languageUndefined/Unknown
Title of host publicationProc. of the Vehicular Technology Conference (VTC)
Place of PublicationSan Francisco, CA, USA
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages1-5
Number of pages5
DOIs
Publication statusPublished - 1 Sep 2011

Abstract

Cognitive Radio systems require detailed feedback about their environment, and detecting anomalies is core to this task. The KLD metric can be used to detect a variety of anomalies in radio signals, and has been previously demonstrated to be both effective, and efficient enough to run in real-time. In tests, it was observed that some anomalous signals caused the KLD to increase monotonically for long time periods, while others did not. After analysing the KLD equation and comparing the findings with the results from the tests, we present a hypothesis for how such monotonic sequences could occur and demonstrate that this agrees very closely with results in observed signals.

ID: 1844460