Sumio Hosaka

Learn More
Although synaptic behaviours of memristors have been widely demonstrated, implementation of an even simple artificial neural network is still a great challenge. In this work, we demonstrate the associative memory on the basis of a memristive Hopfield network. Different patterns can be stored into the memristive Hopfield network by tuning the resistance of(More)
We have investigated electrical properties of a chalcogenide-based device with naturally oxidized Al electrodes. Intermediate-resistance (IR) states exhibited by current–voltage (I–V) characteristics, dynamic resistance change as a function of pulse height and decay behavior from a low-resistance state of such a device make multi-state storage feasible.(More)
A huge internal stress is usually induced by a volume change in chalcogenide as the storage media in phase-change memory (PCM) devices and deteriorates their performance. In this work, doping N into GeTe with a fast operation speed (ns order) is investigated with the aim of reducing the volume change upon crystallization. The volume change upon(More)
Lateral single-channel (SC) and multi-channel (MC) phase-change memory (PCM) cell structures are numerically analyzed for multi-state storage based on their temperature distributions and their programming characteristics. As studied by finite element modeling (FEM), shorter pulses lead to more gradual transition from the set to the reset state of these(More)
We have studied a possibility to read small pits in electron beam (EB) resist layer using a near-field optical probe for ultrahigh density optical storage. For the reading, we used small near-field optical probe in prototype atomic force cantilevered SNOM, which has an ability to obtain both atomic force microscope (AFM) and scanning near-field microscope(More)
  • 1