Title :
Shared memristance restoring circuit for memristor-based Cellular Neural Networks
Author :
Youngsu Kim ; SangHak Shin ; Kyeong-Sik Min
Author_Institution :
Sch. of Electr. Eng., Kookmin Univ., Seoul, South Korea
Abstract :
In this paper, a memristance restoring circuit in the memristor-based Cellular Neural Networks (CNNs) is proposed. The proposed restoring circuit of memristance can be shared among many rows of synaptic circuits in CNN array to minimize the area overhead. Assuming 10×10 CNN cell array, if we share the restoring circuit among 10 rows of synaptic circuits in CNN array, the area overhead of the restoring circuit can be reduced to 1/10. The circuit simulation results show that the proposed restoring circuit can reduce the percentage variation in memristance from 21.1% to 0.12% during the restoring cycles.
Keywords :
cellular neural nets; memristors; CNN array; memristance restoring circuit; memristor based cellular neural networks; synaptic circuits; Arrays; Cellular neural networks; Circuit simulation; Educational institutions; Integrated circuit modeling; Memristors; Semiconductor device modeling; Cellular neural networks; memristors; shared memristance restoring circuits; synaptic weighting circuits;
Conference_Titel :
Cellular Nanoscale Networks and their Applications (CNNA), 2014 14th International Workshop on
Conference_Location :
Notre Dame, IN
DOI :
10.1109/CNNA.2014.6888625