Digital perceptron

Abstract

In view of the neural network information parallel processing, a digital perceptron device analogous to the build-in neural network hardware systems for parallel processing digital signals directly by the processor's memory content and memory perception in one feed-forward step is disclosed. The digital perceptron device of the invention applies the configurable content and perceptive non-volatile memory arrays as the memory processor hardware. The input digital signals are then broadcasted into the non-volatile content memory array for a match to output the digital signals from the perceptive non-volatile memory array as the content-perceptive digital perceptron device.

Description

BACKGROUND OF THE INVENTION[0001]Field of the invention[0002]The invention is related to a digital signal processor, which can interpret the receptive digital signals into the perceptive output digital signals. In particular, the processor parallel processes digital information according to its pre-configured digital content and perceptive non-volatile memories without executing any sequential Boolean logic operations. That is, instead of executing the combinational logic computations given by the programmed instructions in the conventional digital processors, the invented digital signal processor intelligently processes digital information fully based on their configured content and perceptive non-volatile memory hardware.Description of the Related Art[0003]In the modern Von Neumann computing architecture as shown in FIG. 1, the Central Process Unit (CPU) executes logic operations according to the instructions and data from the main memory 11. The CPU 10 includes a main memory 11, ...

AI Technical Summary

Benefits of technology

[0009]In another aspect of this invention, the multiple-time configurability of non-volatile memories in the digital perceptrons provides the capability of real-time updating the digital content and output symbols. The digital content and output symbols can be renewed anytime according to the coding efficiency and the learning algorithms for the real world scenarios. From the perspective, the digital perceptron can evolve into a processor for better processing efficiency and more desirable functions set by the learning algorithm as the training for the processor.

Problems solved by technology

The more frequent memory accessing to complete the computation processing steps, the more energy and processing time are consumed for the digital processors.

The energy and time consumed for searching a digital symbol by addresses in a large memory database storage become very inefficient as the general practice of running programmed software algorithm with many times of memory accessing between CPU and non-volatile memory database storage, and the data comparisons in the present computing system.

Images