Dr. Matthew Pickett and I have been collaborating on a project at HP Labs to explore the possibility of using "locally-active memristors" as the basis for extremely low-power transistorless computation. We first analyzed the thermally-induced first order phase transition from a Mott insulator to a highly conducting state. The current-voltage characteristic of a cross-point device that has a thin film of such a material sandwiched between two metal electrodes displays a current-controlled or 'S'-type negative differential. We derived analytical equations for the behavior these devices, and found that the resulting dynamical model was mathematically equivalent to the "memristive system" formulation of Leon Chua; we thus call these devices "Mott Memristors. We built Pearson-Anson oscillators based on a parallel circuit of one Mott memristor and one capacitor, and demonstrated subnanosecond and subpicoJoule switching time and energy. We then built a neuristor using two Mott memristors and two capacitors, which emulates the Hodgkin-Huxley model of the axon action potential of a neuron. Finally, through SPICE, we demonstrate that spiking neuristors are capable of Boolean logic and Turing complete computation by designing and simulating the one dimensional cellular nonlinear network based on 'Rule 137'.
Wednesday, October 24, 2012
Free and open to the public