Associative memory circuit based on memory resistor

Abstract

The invention discloses an associative memory circuit based on a memory resistor. The associative memory circuit based on the memory resistor comprises the memory resistor, a first resistor, a second resistor and a calculation comparator. The first resistor and the memory resistor are connected to a first input end of the calculation comparator in series in sequence. A non-series-connection connecting end of the memory resistor serves as a first input end of the associative memory circuit. A series-connection connecting end of the first resistor and the memory resistor serves as a second input end of the associative memory circuit. One end of the second resistor is connected to the first input end of the calculation comparator and the other end of the second resistor is grounded. A second input end of the calculation comparator is used for being connected with reference voltage. The output end of the calculation comparator is used as the output end of the associative memory circuit. The first input end and the second input end of the associative memory circuit are used for receiving conditioned stimulus signals and receiving unconditioned stimulus signals respectively. The output end of the associative memory circuit is used for outputting response signals. By the adoption of the associative memory circuit based on the memory resistor, the forming process and the forgetting process of biological associative memory can be simulated according to the relation between the application time of the conditioned stimulus signals and the application time of the unconditioned stimulus signals.

Description

technical field [0001] The invention belongs to the field of artificial neural networks, and more specifically relates to an associative memory circuit based on a memristor. Background technique [0002] Associative memory is a cognitive behavior of biological neural network, and its neurobiological basis is the plastic change of synapses. Associative memory is also an important function in artificial neural network theory, which is widely used in the fields of intelligent control, pattern recognition and artificial intelligence. However, in the traditional artificial neural network circuit, dozens of transistors, capacitors, adders and other components are needed to build a neuron and a synapse. To realize the associative memory function requires the construction of a complex and huge circuit system. [0003] At present, some research institutions realize the associative memory function through CMOS circuits. At that time, in view of the fact that CMOS circuits realize the...

AI Technical Summary

Problems solved by technology

However, in the traditional artificial neural network circuit, dozens of transistors, capacitors, adders and other components are needed to build a neuron and a synapse. To realize the associative memory function requires the construction of a complex and huge circuit system

[0003] At present, some research institutes have realized the associative memory function through CMOS circuits. At that time, in view of the realization of the associative memory function by CMOS circuits, many electronic components were needed, the circuit scale was large, and the power consumption was high.

At the same time, this method does not represent the basic biological model of associative memory based on the temporal relationship of stimulus signals.

Images