Anomaly Detection based on Compressed Data: an Information Theoretic Characterization
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2021
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912086199631872 |
|---|---|
| author | Marchioni, Alex Enttsel, Andriy Mangia, Mauro Rovatti, Riccardo Setti, Gianluca |
| author_facet | Marchioni, Alex Enttsel, Andriy Mangia, Mauro Rovatti, Riccardo Setti, Gianluca |
| contents | We analyze the effect of lossy compression in the processing of sensor signals that must be used to detect anomalous events in the system under observation. The intuitive relationship between the quality loss at higher compression and the possibility of telling anomalous behaviours from normal ones is formalized in terms of information-theoretic quantities. Some analytic derivations are made within the Gaussian framework and possibly in the asymptotic regime for what concerns the stretch of signals considered. Analytical conclusions are matched with the performance of practical detectors in a toy case allowing the assessment of different compression/detector configurations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2110_02579 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Anomaly Detection based on Compressed Data: an Information Theoretic Characterization Marchioni, Alex Enttsel, Andriy Mangia, Mauro Rovatti, Riccardo Setti, Gianluca Information Theory Signal Processing We analyze the effect of lossy compression in the processing of sensor signals that must be used to detect anomalous events in the system under observation. The intuitive relationship between the quality loss at higher compression and the possibility of telling anomalous behaviours from normal ones is formalized in terms of information-theoretic quantities. Some analytic derivations are made within the Gaussian framework and possibly in the asymptotic regime for what concerns the stretch of signals considered. Analytical conclusions are matched with the performance of practical detectors in a toy case allowing the assessment of different compression/detector configurations. |
| title | Anomaly Detection based on Compressed Data: an Information Theoretic Characterization |
| topic | Information Theory Signal Processing |
| url | https://arxiv.org/abs/2110.02579 |