Area efficient VLSI architectures for Huffman coding

Abstract

Absmt--In this paper, we present simple and area efficient VLSI architectures for HufFman coding, an industrial standard proposed by MPEG, JPEG, and others. We use a memory of size O(nlogn) bits to store a HuEman code tree, where n is the number of symbols. This storage scheme supports real-time encoding and decoding. In addition, few simple arithmetic operations are performed on the chip for encoding and decoding. Based on our scheme, we show a design for &bit symbols. The proposed design requires 256 X 9 and 64 X U-bit memory modules to process 8-bit symbols. The chip occupies a silicon a m of 3.5 X 3.5 mm' using 1.2 micron CMOSN standard library cells. Compared with the known parallel implementation in [I] which requires upto 65536 PE's, the proposed architecture leads to a single PE design. It requires significantly less area than the known single PE design in 121. Different Huffknan codes can be stored by changing the contents of the memory, without changing the design. I. INTRODUC~ONLow bit rate coding is essential for image applications such as TV transmission, video conferencing, remote sensing via satellite, computer communication, facsimile transmission, etc. Data compression techniques reduce the storage and communication channel bandwidth needed to process such signals. Most conventional compression techniques require lossless coding method at the end of the encoder to obtain additional compression. In addition, some applications require lossless coding schemes. The lossless data encoding schemes can also reduce the cost of backup and recovery in computer systems and increased security and efficient search operations on compressed index structure of files. It is also useful in the design of databases [3], document delivery systems, communication networks, and high performance supercomputers [ll. Several lossless coding schemes have been proposed for these purposes. One of these lossless coding schemes is Huffman coding (also known as optimal variable length coding) which provides optimal coding for a fixed length

A generalized prediction method for modified memory-based high throughput VLC decoder design

“…Moreover, two images, which are encoded by MPEG2 VLC codewords, are used to perform the comparison in actual applications. Note that both [8] and [9] are memory-based architectures. Assume, under the same working frequency, the throughput of our proposal is about 3.2 times of [9] and 8 times of [8].…”

Efficient VLSI Huffman encoder implementation and its application in high rate serial data encoding

“…Therefore, the memory bits required by the various implementations are employed here as a comparison measurement of silicon area. Compared with previous works, which require 256 Ã 9 þ 64 Ã 18 ¼ 3456 bits of memory [4], and 512 Ã 12 ¼ 6144 bits of memory [5] to process 8 bit symbols, the proposed design requires 256 Ã 8 bits of input symbol memory and 73 Ã 10 bits of temporarily available memory, for a total of 2778 bits of memory to process 8 bit symbols. This verifies a substantial decrease in silicon area for the proposed architecture.…”

Optimisation of variable-length code for data compression of memoryless Laplacian source

“…It requires a very complex realization structure and also is time consuming. Hardware implementations of popular compression algorithms such as the Huffman coding [11], Lempel-Ziv coding [12], binary arithmetic coding [13], and the Rice algorithm [14] have been reported in the literature. A 12-bit A/D with a simplified Huffman encoder is presented in [15].…”