Quantum nerve network testing method for multiple users

Abstract

The quanta NN multiuser detecting method relates to the imitation realizing the method by the classic computer, the method makes the quanta NN network construct the multiuser detecting apparatus, the net core adopts the feedback quanta NN cell to simplify the construction of the multiuser detecting setting, the net evolvement uses the quanta combing working character to process the quick searching the excellent, the complexity of the multiuser detecting setting can be reduced; the invention is showed as the below steps: the one feedback user detecting setting is designed, the method expressing the multiuser receiving signal by the quanta storage, the one quanta NN multiuser detecting setting, the combining working evolvement operator F1; the combining working evolvement operator F1 acts on the output quanta state of the quanta NN net multiuser detecting setting; the above one-step is repeated till the renovated output quanta state has not the change compared to state before the renovation, the one random evolvement operator F2 is designed to substitute for the combining evolvement operator F1; the above one-step is repeated till the renovated output quanta state has not the change compared to the state before the renovation.

Description

technical field [0001] Quantum neural network (QNN-Quantum Neural Networks) is a new intelligent computing paradigm that combines conventional artificial neural network (ANN) with quantum computing theory. The invention relates to a method for using a quantum neural network for multi-user detection and a simulation implementation of the method on a classical computer, and the research content belongs to the technical field of communication signal processing. Background technique [0002] In 1995, Professor Kak of the United States first proposed the concept of quantum neural computing. Later, some scholars proposed quantum neural network models such as quantum derivative neural network, quantum dot neural network, quantum associative memory model, and quantum entanglement neural network. Studies have shown that due to the use of quantum computing properties such as quantum parallelism and quantum entanglement, quantum neural network (QNN) has better performance than ANN in t...

AI Technical Summary

Problems solved by technology

[0005] Technical problem: The purpose of the present invention is to provide a method for simulating and realizing quantum neural network multi-user detection on a classical computer, so as to solve the problem that the classical neural network multi-user detector has high hardware complexity, is difficult to converge to the global optimal point, and is not suitable for large-scale The problem of multi-user communication with the number of users

Images