Sökning: id:"swepub:oai:DiVA.org:kth-342796" >
Anomaly-Based Drone...
Anomaly-Based Drone Classification Using a Model Trained Convolutional Neural Network Autoencoder on Radar Micro-Doppler
-
- Karlsson, Alexander (författare)
- KTH,Teknisk informationsvetenskap,SAAB AB, Product Unit Electronic Surveillance Business Area Surveillance, Stockholm, Sweden
-
- Jansson, Magnus, Professor (författare)
- KTH,Teknisk informationsvetenskap
-
- Hamalainen, Mikael (författare)
- SAAB AB, Product Unit Electronic Surveillance Business Area Surveillance, Stockholm, Sweden
-
(creator_code:org_t)
- Institute of Electrical and Electronics Engineers (IEEE), 2023
- 2023
- Engelska.
-
Ingår i: 2023 IEEE International Radar Conference, RADAR 2023. - : Institute of Electrical and Electronics Engineers (IEEE).
- Relaterad länk:
-
https://urn.kb.se/re...
-
visa fler...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- We present an anomaly-based drone classification scheme. High dimensional spectrum data is encoded using a convolutional neural network autoencoder. This is trained on data generated from a generic mathematical drone model. Once encoded, we use quadratic discriminant analysis for non-drone classes and define anomalies in terms of the log likelihood and prior knowledge from the drone model. When integrating ten samples, we can discriminate drones from non-drone samples such as birds, with an average accuracy of 98% at 20 dB signal to noise ratio. This corresponds to an effective observation time of 90 ms.
Ämnesord
- NATURVETENSKAP -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Probability Theory and Statistics (hsv//eng)
Nyckelord
- bird
- classification
- deep learning
- drone
- high dimensional anomaly detection
- QDA
- radar
- RCS modeling
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)