Embedded systems are constrained by the available memory, and code-compression techniques address this issue by reducing the code size of application programs. The main challenge for the development of an effective code-compression technique is to reduce code size without affecting the overall system performance. Dictionary-based code-compression schemes are the most commonly used code-compression methods, because they can provide both good compression ratio and fast decompression. We propose an XOR-based reference scheme that can enhance the compression ratio on all the existing dictionary-based algorithms by changing the distribution of the symbols. Our approach works on all kinds of computer architecture with fixed length instructions, such as RISC or VLIW. Experiments show that our approach can further improve the compression ratio with nearly no hardware, performance, and power overheads.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Jui-Chun CHEN, Chang-Hong LIN, "Improved Dictionary-Based Code-Compression Schemes with XOR Reference for RISC/VLIW Architecture" in IEICE TRANSACTIONS on Fundamentals,
vol. E93-A, no. 12, pp. 2517-2523, December 2010, doi: 10.1587/transfun.E93.A.2517.
Abstract: Embedded systems are constrained by the available memory, and code-compression techniques address this issue by reducing the code size of application programs. The main challenge for the development of an effective code-compression technique is to reduce code size without affecting the overall system performance. Dictionary-based code-compression schemes are the most commonly used code-compression methods, because they can provide both good compression ratio and fast decompression. We propose an XOR-based reference scheme that can enhance the compression ratio on all the existing dictionary-based algorithms by changing the distribution of the symbols. Our approach works on all kinds of computer architecture with fixed length instructions, such as RISC or VLIW. Experiments show that our approach can further improve the compression ratio with nearly no hardware, performance, and power overheads.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E93.A.2517/_p
Copy
@ARTICLE{e93-a_12_2517,
author={Jui-Chun CHEN, Chang-Hong LIN, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Improved Dictionary-Based Code-Compression Schemes with XOR Reference for RISC/VLIW Architecture},
year={2010},
volume={E93-A},
number={12},
pages={2517-2523},
abstract={Embedded systems are constrained by the available memory, and code-compression techniques address this issue by reducing the code size of application programs. The main challenge for the development of an effective code-compression technique is to reduce code size without affecting the overall system performance. Dictionary-based code-compression schemes are the most commonly used code-compression methods, because they can provide both good compression ratio and fast decompression. We propose an XOR-based reference scheme that can enhance the compression ratio on all the existing dictionary-based algorithms by changing the distribution of the symbols. Our approach works on all kinds of computer architecture with fixed length instructions, such as RISC or VLIW. Experiments show that our approach can further improve the compression ratio with nearly no hardware, performance, and power overheads.},
keywords={},
doi={10.1587/transfun.E93.A.2517},
ISSN={1745-1337},
month={December},}
Copy
TY - JOUR
TI - Improved Dictionary-Based Code-Compression Schemes with XOR Reference for RISC/VLIW Architecture
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2517
EP - 2523
AU - Jui-Chun CHEN
AU - Chang-Hong LIN
PY - 2010
DO - 10.1587/transfun.E93.A.2517
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E93-A
IS - 12
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - December 2010
AB - Embedded systems are constrained by the available memory, and code-compression techniques address this issue by reducing the code size of application programs. The main challenge for the development of an effective code-compression technique is to reduce code size without affecting the overall system performance. Dictionary-based code-compression schemes are the most commonly used code-compression methods, because they can provide both good compression ratio and fast decompression. We propose an XOR-based reference scheme that can enhance the compression ratio on all the existing dictionary-based algorithms by changing the distribution of the symbols. Our approach works on all kinds of computer architecture with fixed length instructions, such as RISC or VLIW. Experiments show that our approach can further improve the compression ratio with nearly no hardware, performance, and power overheads.
ER -