Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

AI-based Classification of Radar Signals

Hollmann, Ludwig LU and Fors Joki, Max (2021) In Master's thesis in Mathematical Scieces FMSM01 20202
Mathematical Statistics
Abstract (Swedish)
Military vehicles typically send out radar signals in order to detect their surroundings. Using an Electromagnetic Support Measures receiver, these can be detected and it is of interest to be able to classify them in order to identify the vehicle and type of radar. Although there already exists multiple methods to do this, it is of interest to automatize and speed up the process of classification as much as possible. Artificial Neural Networks is a form of machine learning that has proven to be successful in classifying sequential data from a large variation of sources. The purpose of this thesis is therefore to investigate how well Artificial Neural Networks can classify different types of radar signals, when the data is given on a... (More)
Military vehicles typically send out radar signals in order to detect their surroundings. Using an Electromagnetic Support Measures receiver, these can be detected and it is of interest to be able to classify them in order to identify the vehicle and type of radar. Although there already exists multiple methods to do this, it is of interest to automatize and speed up the process of classification as much as possible. Artificial Neural Networks is a form of machine learning that has proven to be successful in classifying sequential data from a large variation of sources. The purpose of this thesis is therefore to investigate how well Artificial Neural Networks can classify different types of radar signals, when the data is given on a sequential form of arrival times from radar pulses.
Both feed-forward and recurrent neural networks of different types are considered and in addition to this, methods to apply them on the specific radar data are developed. The results show that artificial neural networks are capable of classifying radar signals of this form with a precision of up to 98 percent. In addition to this it could also be done with a couple of seconds worth of data using relatively simple models. (Less)
Please use this url to cite or link to this publication:
author
Hollmann, Ludwig LU and Fors Joki, Max
supervisor
organization
course
FMSM01 20202
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Neural networks, TensorFlow, Keras, Radar signals
publication/series
Master's thesis in Mathematical Scieces
report number
LUTFMS-3429-2021
ISSN
1404-6342
other publication id
2021:E56
language
English
id
9052789
date added to LUP
2021-07-06 11:59:38
date last changed
2021-08-20 16:57:54
@misc{9052789,
  abstract     = {{Military vehicles typically send out radar signals in order to detect their surroundings. Using an Electromagnetic Support Measures receiver, these can be detected and it is of interest to be able to classify them in order to identify the vehicle and type of radar. Although there already exists multiple methods to do this, it is of interest to automatize and speed up the process of classification as much as possible. Artificial Neural Networks is a form of machine learning that has proven to be successful in classifying sequential data from a large variation of sources. The purpose of this thesis is therefore to investigate how well Artificial Neural Networks can classify different types of radar signals, when the data is given on a sequential form of arrival times from radar pulses.
Both feed-forward and recurrent neural networks of different types are considered and in addition to this, methods to apply them on the specific radar data are developed. The results show that artificial neural networks are capable of classifying radar signals of this form with a precision of up to 98 percent. In addition to this it could also be done with a couple of seconds worth of data using relatively simple models.}},
  author       = {{Hollmann, Ludwig and Fors Joki, Max}},
  issn         = {{1404-6342}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Master's thesis in Mathematical Scieces}},
  title        = {{AI-based Classification of Radar Signals}},
  year         = {{2021}},
}