Systolic Pattern Recognition Based on Neural Algorithm

Dinu O. Creteanu; Valeriu Beiu; Jan A. Peperstraete; Rudy Lauwereins

doi:10.1007/978-3-7091-7533-0_22

Systolic Pattern Recognition Based on Neural Algorithm

Dinu O. Creteanu, Valeriu Beiu, Jan A. Peperstraete, Rudy Lauwereins

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The paper presents a solution for pattern classification, which uses distribute processing both for computing the matching score and selecting the class with the maximum score. The proposed architecture belongs to systolic arrays, being a generalization of the classical priority queue. A detailed description of the elementary processors (EPs) reveals that the algorithm implemented by each EP (which is based on computing the Hamming distance) is common also for neural networks. The overall result is a O(M) execution time for M classes (i.e. linear), and O(1) execution time with respect to n (the size of the patterns). For testing the ideas, a simulator has been developed. It has been built starting from a set of C functions for simulating parallel processes. A short description of these functions supports our claim about the improvement of efficiency when developing a simulator starting from these functions. Several results are shortly discussed. Conclusions and further directions of research end the paper.

Original language	English
Title of host publication	International Conference on Artificial Neural Networks and Genetic Algorithms
DOIs	https://doi.org/10.1007/978-3-7091-7533-0_22
Publication status	Published - Apr 13 1993
Event	ICANNGA'93 - Innsbruck, Austria Duration: Apr 13 1993 → …

Conference

Conference	ICANNGA'93
Period	4/13/93 → …

Access to Document

10.1007/978-3-7091-7533-0_22

http://link.springer.com/chapter/10.1007%2F978-3-7091-7533-0_22

Cite this

@inproceedings{f23d0b3bd0f445f5a0c7fc942e5d8b73,

title = "Systolic Pattern Recognition Based on Neural Algorithm",

abstract = "The paper presents a solution for pattern classification, which uses distribute processing both for computing the matching score and selecting the class with the maximum score. The proposed architecture belongs to systolic arrays, being a generalization of the classical priority queue. A detailed description of the elementary processors (EPs) reveals that the algorithm implemented by each EP (which is based on computing the Hamming distance) is common also for neural networks. The overall result is a O(M) execution time for M classes (i.e. linear), and O(1) execution time with respect to n (the size of the patterns). For testing the ideas, a simulator has been developed. It has been built starting from a set of C functions for simulating parallel processes. A short description of these functions supports our claim about the improvement of efficiency when developing a simulator starting from these functions. Several results are shortly discussed. Conclusions and further directions of research end the paper.",

author = "Creteanu, {Dinu O.} and Valeriu Beiu and Peperstraete, {Jan A.} and Rudy Lauwereins",

year = "1993",

month = apr,

day = "13",

doi = "10.1007/978-3-7091-7533-0_22",

language = "English",

isbn = "978-3-211-82459-7",

booktitle = "International Conference on Artificial Neural Networks and Genetic Algorithms",

note = "ICANNGA'93 ; Conference date: 13-04-1993",

}

TY - GEN

T1 - Systolic Pattern Recognition Based on Neural Algorithm

AU - Creteanu, Dinu O.

AU - Beiu, Valeriu

AU - Peperstraete, Jan A.

AU - Lauwereins, Rudy

PY - 1993/4/13

Y1 - 1993/4/13

N2 - The paper presents a solution for pattern classification, which uses distribute processing both for computing the matching score and selecting the class with the maximum score. The proposed architecture belongs to systolic arrays, being a generalization of the classical priority queue. A detailed description of the elementary processors (EPs) reveals that the algorithm implemented by each EP (which is based on computing the Hamming distance) is common also for neural networks. The overall result is a O(M) execution time for M classes (i.e. linear), and O(1) execution time with respect to n (the size of the patterns). For testing the ideas, a simulator has been developed. It has been built starting from a set of C functions for simulating parallel processes. A short description of these functions supports our claim about the improvement of efficiency when developing a simulator starting from these functions. Several results are shortly discussed. Conclusions and further directions of research end the paper.

AB - The paper presents a solution for pattern classification, which uses distribute processing both for computing the matching score and selecting the class with the maximum score. The proposed architecture belongs to systolic arrays, being a generalization of the classical priority queue. A detailed description of the elementary processors (EPs) reveals that the algorithm implemented by each EP (which is based on computing the Hamming distance) is common also for neural networks. The overall result is a O(M) execution time for M classes (i.e. linear), and O(1) execution time with respect to n (the size of the patterns). For testing the ideas, a simulator has been developed. It has been built starting from a set of C functions for simulating parallel processes. A short description of these functions supports our claim about the improvement of efficiency when developing a simulator starting from these functions. Several results are shortly discussed. Conclusions and further directions of research end the paper.

U2 - 10.1007/978-3-7091-7533-0_22

DO - 10.1007/978-3-7091-7533-0_22

M3 - Conference contribution

SN - 978-3-211-82459-7

BT - International Conference on Artificial Neural Networks and Genetic Algorithms

T2 - ICANNGA'93

Y2 - 13 April 1993

ER -

Systolic Pattern Recognition Based on Neural Algorithm

Abstract

Conference

Access to Document

Fingerprint

Cite this