Date of Award
4-2016
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computer Engineering and Sciences
First Advisor
Fareena Saqib
Second Advisor
William W. Arrasmith
Third Advisor
Lucas Stephane
Fourth Advisor
Samuel Kozaitis
Abstract
Atmospheric turbulence compensation techniques are used to apply filters and to remove noise from the information, such as images. In mission control applications, the processing is required in real time. This thesis focuses on understanding the process and developing architecture for hardware acceleration on the FPGA. Error calculation is a computational extensive component and is a continuous research area to correctly and efficiently measure it in real-time for data processing. The Error is used as a metric to compute the corrected data, such as an image pixel representing the phase, color and intensity to perform image processing. The thesis discusses the design details, resource overhead and performance of the proposed architecture for the error calculation of different sized images, such as 128x128 and 256x256, where each pixel is carries information in form of double precision floating-point numbers that are complex. The computation involves the memory operation and the floating-point operations involving the complex-numbers operations. As a prototype, this project is designed for the computation of 2x2 pixels image. The Analysis is carried out to the 128 x 128 pixels from the results of the prototype up to 1024 x 1024 pixels.
Recommended Citation
Badri, Nikhileswar, "Hardware Acceleration for Real Time Atmospheric Turbulence Compensation" (2016). Theses and Dissertations. 1530.
https://repository.fit.edu/etd/1530
