Supercharge Your Innovation With Domain-Expert AI Agents!

Undersampling frequency hopping communication signal deep learning recovery method

A technology of frequency hopping communication and deep learning, applied in the field of communication, can solve the problems of high computational complexity, low computational efficiency, low reconstruction accuracy, etc., and achieve the effects of improving computational efficiency, reducing computational load, and improving accuracy.

Active Publication Date: 2020-09-18
XIDIAN UNIV
View PDF9 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007]Currently, existing signal restoration methods have high computational complexity, low computational efficiency, and low reconstruction accuracy, which limit their practical application

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Undersampling frequency hopping communication signal deep learning recovery method
  • Undersampling frequency hopping communication signal deep learning recovery method
  • Undersampling frequency hopping communication signal deep learning recovery method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] With the development of wireless communication technology, various emerging communication methods continue to emerge, and wireless spectrum resources are increasingly tight. Frequency hopping communication has become a new solution to frequency band congestion due to its high frequency band utilization. Frequency hopping communication has strong anti-interference ability. Low interception, easy networking, superior security performance and other advantages. The principle of the deep learning recovery method of the under-sampled frequency-hopping communication signal of the present invention is that, according to the deep learning, the internal law and representation level of the sample data can be learned, and the mapping relationship between the input signal and the output signal can be learned, so that the error of the reconstructed signal is smaller, The purpose of improving the performance of frequency hopping communication is achieved.

[0033] In the prior art, a ...

Embodiment 2

[0045] The undersampling frequency hopping signal deep learning recovery method is the same as that of embodiment 1, and the construction convolutional neural network and variational autoencoder network structure described in step 1 of the present invention, wherein the convolutional neural network, the network parameters are set to: convolution kernel The size is 10, the number is 16, the step size is 1, and the filling method is "Same"; the convolutional neural network is essentially an input-to-output mapping, which can learn a large number of mapping relationships between inputs and outputs. Without any precise mathematical expression between input and output, as long as the convolutional network is trained with a known pattern, the network has the ability to map between input and output pairs. The convolutional neural network uses its local weights The shared special structure has unique advantages in speech recognition and image processing. Its layout is closer to the act...

Embodiment 3

[0047] The undersampling frequency hopping signal deep learning recovery method is the same as embodiment 1-2, and the data preprocessing process described in step 3 is as follows: by the expression Y=AX of compressed sensing, wherein Y={Y 1 [n],Y 2 [n]...,Y m [n]} T is the preprocessed vector, A is the observation matrix, X={X 1 [n],X 2 [n]...,X m [n]} is the preprocessed vector, let Y=a+jb, A=B+jC, X=c+jd, available from Y=AX, a+jb=(B+jC)*(c+ jd), dismantling, making the real part and the real part of the complex number equal, and the imaginary part and the imaginary part equal, the following relation can be obtained,

[0048]

[0049] Let the new observation matrix The above formula can be transformed into

[0050]

[0051] which is

[0052]

[0053] Multiply both sides of the above formula by A T ,Available

[0054]

[0055] in, is the input signal of a convolutional neural network or a variational autoencoder, [c-d] T is the output signal of a co...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an undersampling frequency hopping communication signal deep learning recovery method, and solves the problems of high calculation complexity, low calculation efficiency and low reconstruction precision of a signal recovery method in the prior art. The method comprises the following steps: constructing a convolutional neural network and a variational auto-encoder network, introducing a new network structure and parameter setting, and constructing an optimal neural network structure; linearly measuring a frequency hopping signal; preprocessing the data, and obtaining a data format input into the neural network in a data preprocessing mode of separating real parts and imaginary parts of the complex numbers; training a convolutional neural network and a variational auto-encoder network; frequency hopping signal recovery. Two networks of weight sharing and sparse connection are constructed, a data format input into a neural network is obtained in a data preprocessing mode, and an original frequency hopping signal is reconstructed through a trained optimal neural network structure and parameter setting. The method greatly improves the accuracy of signal recovery,reduces the reconstruction error and calculation complexity, and is used for a frequency hopping communication system.

Description

technical field [0001] The invention belongs to the technical field of communication, and in particular relates to frequency hopping signal recovery, in particular to a method for deep learning recovery of undersampled frequency hopping communication signals, which can be used in military and civilian communications. Background technique [0002] Frequency hopping communication is a common spread spectrum communication method. Its working principle is that the carrier frequency of the sender and the receiver changes according to a certain rule when transmitting signals. The carrier frequency is hopped by the change of the pseudo-random sequence. , has the advantages of strong anti-interference ability, low interception and easy networking. With the development of wireless communication technology, various emerging communication methods continue to emerge, and wireless spectrum resources are becoming increasingly scarce. Frequency hopping communication has become a new soluti...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06F17/14
CPCG06F17/142G06N3/045G06F2218/00G06F18/214Y02D30/70
Inventor 齐佩汉王凡周涛谢爱平梁琳琳周小雨李赞王丹洋关磊都毅毛维安
Owner XIDIAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com