No abstract

We present coupled Spin Torque Nano Oscillators (STNOs) as "electronic neurons" for efficient brain-inspired computation. The coupled STNOs show two distinct outputs, depending on whether the frequencies are locked or not. The locking mechanisms are based on magnetic coupling or injection locking. The neuron firing threshold can be set by tuning the locking range of the coupled STNOs. We employ a crossbar array of programmable memory devices like memristors to implement "electronic synapses" that work seamlessly with the coupled STNO's for hardware implementation of neural networks. Results show that injection locking based neuron model can be attractive from scaling point of view and computation like character recognition can be performed with energy consumption per neuron of ~1.8X and ~3X lower than the digital and the analog CMOS counterpart, respectively.Index Terms-coupled spin torque nano-oscillators, frequency locking, injection locking, low power, neural networks, programmable resistive crossbar.

No abstract

No abstract

Abstract-In this letter, we investigate the nature of shunt leakage currents in large-area (on the order of square centimeters) thin-film a-Si:H p-i-n solar cells and show that it is characterized by following universal features: 1) voltage symmetry; 2) power-law voltage dependence; and 3) weak temperature dependence. The voltage symmetry offers a robust empirical method to isolate the diode current from measured "shunt-contaminated" forward dark IV . We find that space-charge-limited current provides the best qualitative explanation for the observed features of the shunt current. Finally, we discuss the possible physical origin of localized shunt paths in the light of experimental observations from literature.