New step towards intelligent analogue processors at nanometric scale

Iranian researchers from Sharif University of Technology used memristor nanostrcutures as programmable analogue memories to save parameters of phase calculation methods. The presented method can be used in hardware implementation and in the optimization of artificial intelligence systems. This research is a step towards obtaining an intelligence analogue processor that due to its nanometric dimensions enjoys very high calculation power. The main objective of the research was to resolve one of the most important problems in sub-branches of phase science, which is hardware implementation of phase systems. A big problem against this goal is the phase nature of the data and the other reason is the fact that there are no electric devices compatible with the needs of these systems. Among the innovative specifications of the plan, mention can be made of the simplicity of the calculation system, high rate of data processing, the ability to create complicated systems through the connection of base systems, the existence of compatible hardware with the presented learning method, nanometric scale of the system, non-volatility of the used memristors, the possibility of carrying out the calculation by not specifically adjusting the resistance of elements in memristors, resistance of the system against noise and hardware malfunction, and low power consumption. The use of intersecting memristor nanostructures as analogue memories to save phase connectors and the presentation of a novel method to carry out phase conclusions in these structures are among the results of the research. The presented method can be used in hardware implementation and in the optimization of artificial intelligence systems. Taking into account the high rate of data processing, low power consumption, and very tiny dimensions of memristor systems, the presented system can have numerous applications in various industries and sciences such as robotics, intelligent cell phones, sound and image signal processing systems, and in the recreation of human brain data processing.