Low power text compression using Huffman coding with power management controller

Data compression with low power dissipation is a useful technique in digital systems because it reduces data size with the smallest power consumed to overcome the design limitations. Huffman lossless compression is an important technique in information theory as well as in today‘s IT field. Reducing...

Full description

Saved in:
Bibliographic Details
Main Author: Mohammed, Maan Hameed
Format: Thesis
Language:English
Published: 2016
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/70450/1/FK%202016%2076%20-%20IR.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my-upm-ir.70450
record_format uketd_dc
spelling my-upm-ir.704502019-08-21T03:47:41Z Low power text compression using Huffman coding with power management controller 2016-04 Mohammed, Maan Hameed Data compression with low power dissipation is a useful technique in digital systems because it reduces data size with the smallest power consumed to overcome the design limitations. Huffman lossless compression is an important technique in information theory as well as in today‘s IT field. Reducing data size is the most important goal for Application-Specific Integrated Circuit (ASIC) design and power consumption is one of important issues which restrict the design performance. Therefore, reducing data size and embedding a Power Management Controller (PMC) in a synchronous system is a suitable technique for this problem. Using PMC to control the system to operate in fast and slow modes leads to low power dissipation. This work focused on the design of high performance Huffman compression with low power technique for all English text. Huffman tree is generated by sorting data according to their frequencies and traversing the tree to extract codeword bits for each character. Verilog HDL language used for writing Huffman codes and ModelSim tool was used for simulating the functionality of Huffman. In addition, Field-Programmable Gate Array (FPGA) was used to verify the functionality of Huffman. 8 green LEDs on the FPGA board were used as ASCII input for Huffman encoder, while the first 9 red LEDs on the board were utilized as output of the encoder. The process is reversed in the decoder implementation. A new sub module named PMC was created using clock gating and frequency scaling to improve power consumption in flexible design. Furthermore, Synopsys power compiler with 130nm technology library was used for power analysis. In this study, data size was reduced to 47.95% after compression process when compared with original data size. Moreover, with using PMC, power consumption was reduced up to 52.52% when compared with Huffman without using PMC. Text processing (Computer science) Coding theory Data compression (Computer science) 2016-04 Thesis http://psasir.upm.edu.my/id/eprint/70450/ http://psasir.upm.edu.my/id/eprint/70450/1/FK%202016%2076%20-%20IR.pdf text en public masters Universiti Putra Malaysia Text processing (Computer science) Coding theory Data compression (Computer science)
institution Universiti Putra Malaysia
collection PSAS Institutional Repository
language English
topic Text processing (Computer science)
Coding theory
Data compression (Computer science)
spellingShingle Text processing (Computer science)
Coding theory
Data compression (Computer science)
Mohammed, Maan Hameed
Low power text compression using Huffman coding with power management controller
description Data compression with low power dissipation is a useful technique in digital systems because it reduces data size with the smallest power consumed to overcome the design limitations. Huffman lossless compression is an important technique in information theory as well as in today‘s IT field. Reducing data size is the most important goal for Application-Specific Integrated Circuit (ASIC) design and power consumption is one of important issues which restrict the design performance. Therefore, reducing data size and embedding a Power Management Controller (PMC) in a synchronous system is a suitable technique for this problem. Using PMC to control the system to operate in fast and slow modes leads to low power dissipation. This work focused on the design of high performance Huffman compression with low power technique for all English text. Huffman tree is generated by sorting data according to their frequencies and traversing the tree to extract codeword bits for each character. Verilog HDL language used for writing Huffman codes and ModelSim tool was used for simulating the functionality of Huffman. In addition, Field-Programmable Gate Array (FPGA) was used to verify the functionality of Huffman. 8 green LEDs on the FPGA board were used as ASCII input for Huffman encoder, while the first 9 red LEDs on the board were utilized as output of the encoder. The process is reversed in the decoder implementation. A new sub module named PMC was created using clock gating and frequency scaling to improve power consumption in flexible design. Furthermore, Synopsys power compiler with 130nm technology library was used for power analysis. In this study, data size was reduced to 47.95% after compression process when compared with original data size. Moreover, with using PMC, power consumption was reduced up to 52.52% when compared with Huffman without using PMC.
format Thesis
qualification_level Master's degree
author Mohammed, Maan Hameed
author_facet Mohammed, Maan Hameed
author_sort Mohammed, Maan Hameed
title Low power text compression using Huffman coding with power management controller
title_short Low power text compression using Huffman coding with power management controller
title_full Low power text compression using Huffman coding with power management controller
title_fullStr Low power text compression using Huffman coding with power management controller
title_full_unstemmed Low power text compression using Huffman coding with power management controller
title_sort low power text compression using huffman coding with power management controller
granting_institution Universiti Putra Malaysia
publishDate 2016
url http://psasir.upm.edu.my/id/eprint/70450/1/FK%202016%2076%20-%20IR.pdf
_version_ 1747812840816771072