A new model to optimize the architecture of a fault-tolerant modular neurocomputer

Abstract In this paper, we present some results on error detection and correction in a modular neurocomputer that are based on redundant residue number systems. The error correction method developed below involves the modified Chinese Remainder Theorem with fractions and uses a Hopfield neural network to correct the errors. The suggested approach eliminates the need for extending the bases of a residue number system, a costly operation required in case of syndrome decoding with error syndromes calculation on the control bases of the system. Also the approach does not utilize the projection method, another costly operation intended to localize errors (i.e., to detect the moduli associated with faulty digits). The well-known procedures mentioned above seem inefficient in terms of practical implementation, as they employ a mixed radix number system: transition to such a system is iterative and may affect the performance of a whole neurocomputer. Owing to the exclusion of these costly operations, the suggested approach significantly simplifies error correction procedures for integer numbers.

[1]  Hortensia Mecha,et al.  Hardware implementation of a fault-tolerant Hopfield Neural Network on FPGAs , 2016, Neurocomputing.

[2]  A. Michel,et al.  Analysis and synthesis of a class of neural networks: linear systems operating on a closed hypercube , 1989 .

[3]  A. S. Madhukumar,et al.  A robust symmetrical number system based parallel communication system with inherent error detection and correction , 2009, IEEE Transactions on Wireless Communications.

[4]  Piero Maestrini,et al.  Error Correcting Properties of Redundant Residue Number Systems , 1973, IEEE Transactions on Computers.

[5]  Kamala Krithivasan,et al.  Modelling and analysis of spiking neural P systems with anti-spikes using Pnet lab , 2011, Nano Commun. Networks.

[6]  H. M. Abbas,et al.  An FPGA Implementation of a Hopfield Optimized Block Truncation Coding , 2006, 2006 6th International Workshop on System on Chip for Real Time Applications.

[7]  Chip-Hong Chang,et al.  A new algorithm for single residue digit error correction in Redundant Residue Number System , 2014, 2014 IEEE International Symposium on Circuits and Systems (ISCAS).

[8]  A. Omondi,et al.  Residue Number Systems: Theory and Implementation , 2007 .

[9]  A. S. Molahosseini,et al.  Embedded Systems Design with Special Arithmetic and Number Systems , 2017 .

[10]  Raoul Velazco,et al.  An Optimal Implementation on FPGA of a Hopfield Neural Network , 2011, Adv. Artif. Neural Syst..

[11]  Feng Lin,et al.  A Hopfield Neural Classifier and Its FPGA Implementation for Identification of Symmetrically Structured DNA Motifs , 2007, J. VLSI Signal Process..

[12]  Mohammad Umar Siddiqi,et al.  Multiple error detection and correction based on redundant residue number systems , 2008, IEEE Transactions on Communications.

[13]  Stephen S. Yau,et al.  Error Correction in Redundant Residue Number Systems , 1973, IEEE Trans. Computers.

[14]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[15]  Nikolay I. Chervyakov,et al.  Residue-to-binary conversion for general moduli sets based on approximate Chinese remainder theorem , 2017, Int. J. Comput. Math..

[16]  Thu V. Vu.  Efficient Implementations of the Chinese Remainder Theorem for Sign Detection and Residue Decoding , 1985, IEEE Trans. Computers.

[17]  Chip-Hong Chang,et al.  A non-iterative multiple residue digit error detection and correction algorithm in RRNS , 2016, IEEE Transactions on Computers.

[18]  P. V. Ananda Mohan.  Residue Number Systems: Theory and Applications , 2016 .

[19]  Emilio Del-Moral-Hernandez,et al.  Architecture Analysis of an FPGA-Based Hopfield Neural Network , 2014, Adv. Artif. Neural Syst..

[20]  Nikolay I. Chervyakov,et al.  The architecture of a fault-tolerant modular neurocomputer based on modular number projections , 2018, Neurocomputing.

[21]  Simon Haykin,et al.  Neural Networks and Learning Machines , 2010 .

[22]  Nikolay I. Chervyakov,et al.  An efficient method of error correction in fault-tolerant modular neurocomputers , 2016, Neurocomputing.

[23]  Vladimir Pavlovic,et al.  Bayesian networks as ensemble of classifiers , 2002, Object recognition supported by user interaction for service robots.