Investigation block error rate and bit error rate for linear block codes

• Main Author: Han, Eugene Zhiwei
• Other Authors: Li Kwok Hung
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/67869
• Institution: Nanyang Technological University

Uncoded transmissions are suitable for a low-noise environment, but these ideal conditions are not usually available. While a multitude of options exists for reducing errors during transmission, it is challenging to estimate the volume of additional resources required for an improvement in its error-reduction capability. As such, this report compares uncoded communication systems with some coding schemes, examining their resource efficiency and overall effectiveness. Extensive research has been conducted before and the literature of bit error rate (BER) and block error rate (BLER) relating to binary linear block code is widely available. This project strives to reinforce those findings by employing myriad simulative methods. Hamming-coded transmissions aid in reducing error caused by noise, but their performance varies across different orders of coding systems. Focussing on Hamming codes, a common platform where various types of Hamming codes can be normalized is built, allowing them to be compared based on BER and BLER. BER and BLER are examined in greater details as well, through both theoretical models and simulations. A relationship has been found between the two. It is observed that BLER is higher than BER in all cases. For Hamming codes, the relationship is BER = 3 � BLER, where � is the length of the code. This linearity applies to non-perfect linear block code too. This report also highlights the importance of supporting theory with applications. An investigation into the discrepancy in errors correction between degrees of error reveals that Hamming code performs unpredictably when an error of greater than 1 is introduced. As such, careful analysis shows that many factors like � �/ �0 and predicted errors must be considered for each type of communication system.