Title :
A novel target recognition scheme for WSNs
Author :
Al-Naeem, Mohammed ; Khan, Asad I.
Author_Institution :
Clayton Sch. of IT, Monash Univ., Clayton, VIC, Australia
Abstract :
Many existing pattern recognition schemes in wireless sensor networks suffer from pattern displacement, pattern scaling, and pattern rotation issues. We propose a novel one-shot learning associative memory method for target recognition in wireless sensor networks. This method, known as Macroscopic Object Heuristics Algorithm (MOHA), is able to address all of the above issues. Our proposed scheme is also capable of reducing the power and memory consumptions of wireless sensor networks. The experimental results show that the proposed scheme can effectively and efficiently handle pattern displaced, pattern scaling, and pattern rotation issues.
Keywords :
content-addressable storage; learning (artificial intelligence); pattern recognition; telecommunication computing; wireless sensor networks; WSN; graph neuron; macroscopic object heuristics algorithm; object recognition; one-shot learning associative memory; pattern displacement; pattern recognition; pattern rotation issues; pattern scaling; sensor field; target recognition; wireless sensor networks; Arrays; Equations; Mathematical model; Pattern recognition; Silicon; Target recognition; Wireless sensor networks;
Conference_Titel :
Neural Networks (IJCNN), The 2012 International Joint Conference on
Conference_Location :
Brisbane, QLD
Print_ISBN :
978-1-4673-1488-6
Electronic_ISBN :
2161-4393
DOI :
10.1109/IJCNN.2012.6252700