Title :
Transform domain filtering based on pipelining structure
Author :
Lee, Jae Beom ; Lee, Byeong Gi
Author_Institution :
Dept. of Electron. Eng., Seoul Nat. Univ., South Korea
fDate :
8/1/1992 12:00:00 AM
Abstract :
The concept of transform domain filtering (TDF) is defined, and a pipelining structure is proposed as a means to implement it. TDF refers to filtering applied directly to transform domain data. The proposed structure consists of product blocks and an array of M adders, where M is the transform block length. The TDF output data can be obtained by summing the outputs of each product block. The product block in the proposed structure performs the product operation between an M×M matrix and an M-vector, which can be effectively implemented by using a distributed arithmetic structure of a systolic array structure. The proposed structure is simple, with localized interconnections, and is therefore suitable for VLSI implementations
Keywords :
filtering and prediction theory; pipeline processing; systolic arrays; transforms; VLSI; adders; distributed arithmetic structure; matrix-vector product; pipelining structure; product blocks; systolic array structure; transform block length; transform domain data; transform domain filtering; Convolution; Convolutional codes; Data compression; Discrete cosine transforms; Filtering; Filters; Image coding; Pipeline processing; Signal processing; Very large scale integration;
Journal_Title :
Signal Processing, IEEE Transactions on