Machine learning applications in plant identification, wireless channel estimation, and gain estimation for multi-user software-defined radio

<p>"This work applies machine learning (ML) techniques to selected computer vision and digital communication problems. Machine learning algorithms can be trained to perform a specific task without explicit programming. This research applies ML to the problems of: plant identification from images of leaves, channel state information (CSI) estimation for wireless multiple-input-multiple-output (MIMO) systems, and gain estimation for a multi-user software-defined radio (SDR) application. </p><p>In the first task, two methods for plant species identification from leaf images are developed. One of the methods uses hand-crafted features extracted from leaf images to train a support vector machine classifier. The other method combines five publicly available leaf datasets: Flavia, Folio, LeafSnap, Swedish, and Middle European Woods 2014, to create a new data set named F2LSM. To create a benchmark, multiple end-to-end convolutional neural network classifiers are trained to classify images in the F2LSM dataset. </p><p>The second application of ML is a novel CSI estimation technique for MIMO communication systems. The approach uses atmospheric conditions, the position of the transmitter and receiver, and the relative motion of the transmitter and receiver as features for an artificial neural network (ANN).</p><p>The third study uses two ML methods to estimate gain for a multi-user SDR system in an aircraft, where a single SDR must generate a composite waveform for multiple communication links. An accurate estimate of the composite waveform’s peak is required to set the digital-to-analog converter’s gain to a value that will avoid clipping, while minimizing quantization noise. One of the methods uses an ANN to estimate the waveform’s peak and statistical moments. The other method uses an ANN to estimate the statistical distribution parameters that closely represent the voltage distribution of the waveform"--Abstract, page iv.</p>

Name thesis:degree_name

Ph. D. in Electrical Engineering

Grantor

Missouri University of Science and Technology

Author dc:creator

• Gajjar, Viraj K.

• Computer Vision
• Deep Learning
• Digital Communications
• Machine Learning
• Signal Processing
• Statistical Signal Analysis
• Electrical and Computer Engineering

Repository record dc:identifier

https://scholarsmine.mst.edu/doctoral_dissertations/3169

OAI identifier oai:identifier

oai:scholarsmine.mst.edu:doctoral_dissertations-4174