DNA Inspired Digital Signal Pattern Matching Algorithm

In this paper, a new algorithm for supervised pattern classification is proposed. One application area of DNA computing is the pattern matching problem which arise signal processing applications. The noise tolerance pattern matching can be achieved using hybridization property of DNA computing. The proposed algorithm overcomes some of the limitations that DNA computing as in this field. In the algorithm, a set of sample data is replicated and compared with test data permitting some of different features. These processes emulate the replication and hybridization of DNA computing. The results of classification for cardiovascular disease show better accuracy. Because of the simpler procedure, the proposed algorithm can be implemented on hardware employing parallel architecture, which mimics the as massive parallelism of DNA computing.