Min W. Kang

Date of Award

3-23-2018

Document Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Department of Electrical and Computer Engineering

First Advisor

Kenneth M. Hopkinson, PhD.

Abstract

Satellites are crucial for the modern world to function properly as they provide Global Navigation Satellite System (GNSS) and global communication. However, the data that is stored on these satellites can be corrupted by the radiation found in space, and its bits can be improperly flipped. In the past, Forward Error Correction (FEC) algorithms were selected based on their strength and implemented to correct these bit flips back to their original values. This thesis seeks to determine if the strength of the FEC algorithms Reed Solomon (RS) code and Reed Solomon Product Code (RSPC) directly translates to their effectiveness. These algorithms were coded and tested in Matrix Laboratory (MATLAB) and on a Field Programmable Gate Array (FPGA) under controlled parameters, including the data set sizes, number of bit flips introduced, and the distribution of the bit flips within the data set. From the experiment's results, these other factors significantly influenced the effectiveness of the algorithms as well. Knowing what factors influence the algorithm's effectiveness enable better decision making as to which FEC algorithm to use for a given set of circumstances. The RS codes should be used if the size of the data set is small enough for a single-instance RS code and the range of expected bit flips is narrow and lower than the code's correctable limit. If the data set is large or the range of expected bit flips varies widely and surpasses the RS code's correctable limit, the RSPC should be used for a higher overall success rate in exchange for a lower number of bit flips with a 100% correction rate.

AFIT Designator

AFIT-ENG-MS-18-M-036

DTIC Accession Number

AD1056155

Download

Share

COinS