Title :
A novel discrete relaxation architecture
Author :
Gu, Jun ; Wang, Wei
Author_Institution :
Dept. of Electr. & Comput. Eng., Calgary Univ., Alta., Canada
fDate :
8/1/1992 12:00:00 AM
Abstract :
The discrete relaxation algorithm (DRA) is a computational technique that enforces arc consistency (AC) in a constraint satisfaction problem (CSP). The original sequential AC-1 algorithm suffers from O(n3m3) time complexity, and even the optimal sequential AC-4 algorithm is O (n2m2) for an n-object and m-label DRA problem. Sample problem runs show that these algorithms are all too slow to meet the need for any useful, real-time CSP applications. A parallel DRA5 algorithm that reaches a lower bound of O(nm) (where the number of processors is polynomial in the problem size) is given. A fine-grained, massively parallel hardware computer architecture has been designed for the DRA5 algorithm. For practical problems, many orders of magnitude of efficiency improvement can be reached on such a hardware architecture
Keywords :
computational complexity; parallel algorithms; parallel architectures; polynomials; discrete relaxation algorithm; parallel DRA5 algorithm; polynomial; sequential AC-1 algorithm; sequential AC-4 algorithm; time complexity; Algorithm design and analysis; Application software; Artificial intelligence; Computer architecture; Computer vision; Concurrent computing; Hardware; Labeling; Pattern recognition; Scholarships;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on