Hybrid MTJ/CNTFET-Based Binary Synapse and Neuron for Process-in-Memory Architecture

This letter develops a reliable, integrated binary synapse and neuron model for hardware implementation of binary neural networks. Thanks to the nonvolatile nature of magnetic tunnel junctions and the unique features of carbon nanotube field-effect transistors, the modeled design does not require external memory to store weights and also consumes low static power. Also, due to the circuit structure, which did not use sequential parts, the developed circuit is immune to soft error. Because, in binary neural networks, weights are limited to two values of -1 and 1, the occurrence of soft errors dramatically reduces the accuracy of the network. Simulation results indicate that the design in this work consumes at least 9% lower power, occupies 34% lower area, and offers a 49% lower power delay area product. Also, Monte Carlo simulations have been performed to study the effect of the process variation on the network. The result of the Monte Carlo simulations shows that the proposed neuron has no logical error in 10 000 simulations. Consequently, the accuracy of the network utilization by the neuron is equal to the software-implemented network and does not decrease even in the presence of process variations.

Automated summary

This study presents a reliable and integrated binary synapse and neuron model for implementing binary neural networks in hardware. By utilizing magnetic tunnel junctions and carbon nanotube field-effect transistors, the designed circuit eliminates the need for external memory to store weights and consumes low static power. Furthermore, the circuit is immune to soft errors due to its sequential parts-free structure. Simulation results demonstrate that the proposed design consumes at least 9% lower power, occupies 34% less area, and provides a 49% lower power delay area product compared to existing designs. Additionally, Monte Carlo simulations confirm the accuracy and resilience of the proposed neuron under process variations.