Abstract
dc:description.abstract<p>In this work, machine learning theory is applied to the design of a radar detector in order to train a machine learning-based detector that is robust against Doppler shifts. The radar system is designed to work with data that would be otherwise intractable to conventional optimal detector design, such as transmitted noise waveforms and the effects of one-bit quantization at the receiver. The detection performance of the one-bit receiver is shown to match the performance of the derived square-law sign correlator detector. The resulting learning-based detector also introduces Doppler tolerance to the system, which allows for the successful detection of a waveform that has been Doppler shifted due to sufficiently high target velocity. This is achieved by selecting the training data to best represent all expected Doppler shifts as well as interference effects. For further performance gains, the training data can be adjusted based on a priori estimates of the true Doppler and noise covariance values. Additionally, the advantages to using one-bit data are highlighted, which includes the reduction of computational power and memory requirements, and it is proven that this learning-based detector can be trained to detect waveforms even through the harsh non-linear effects of one-bit quantization.</p>
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy in Electrical Engineering - (Ph.D.)
- Discipline thesis:degree_discipline
- Electrical and Computer Engineering
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Wensell, Kyle Peter
- 0009-0000-8097-171X
- Contributors dc:contributor
-
- Alexander Haimovich
- Ali Abdi
- Joerg Kliewer
Subjects
dc:subject × 9Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.njit.edu/dissertations/1761