KIT | KIT-Bibliothek | Impressum | Datenschutz

Deep reinforcement learning for attacking wireless sensor networks

Parras, Juan; Hüttenrauch, Maximilian 1; Zazo, Santiago; Neumann, Gerhard 1
1 Karlsruher Institut für Technologie (KIT)

Abstract:

Recent advances in Deep Reinforcement Learning allow solving increasingly complex problems. In this work, we show how current defense mechanisms in Wireless Sensor Networks are vulnerable to attacks that use these advances. We use a Deep Reinforcement Learning attacker architecture that allows having one or more attacking agents that can learn to attack using only partial observations. Then, we subject our architecture to a test-bench consisting of two defense mechanisms against a distributed spectrum sensing attack and a backoff attack. Our simulations show that our attacker learns to exploit these systems without having a priori information about the defense mechanism used nor its concrete parameters. Since our attacker requires minimal hyper-parameter tuning, scales with the number of attackers, and learns only by interacting with the defense mechanism, it poses a significant threat to current defense procedures.


Verlagsausgabe §
DOI: 10.5445/IR/1000134273
Veröffentlicht am 26.06.2021
Originalveröffentlichung
DOI: 10.3390/s21124060
Scopus
Zitationen: 1
Web of Science
Zitationen: 1
Dimensions
Zitationen: 2
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Anthropomatik und Robotik (IAR)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2021
Sprache Englisch
Identifikator ISSN: 1424-8220
KITopen-ID: 1000134273
Erschienen in Sensors
Verlag MDPI
Band 21
Heft 12
Seiten 4060
Schlagwörter POMDP; Deep Reinforcement Learning; TRPO; SSDF attack; backoff attack
Nachgewiesen in Dimensions
Scopus
Web of Science
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page