M-Sequence Recognition Method Based on Sparse Autoencoder

A technology of sparse autoencoder and recognition method, which is applied in the field of estimation and recognition of binary pseudo-random sequences, can solve the problem that the detection accuracy of low signal-to-noise ratio peak points needs to be further improved, so as to reduce training complexity and save training time , good recognition effect

Active Publication Date: 2022-05-27
HANGZHOU DIANZI UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the published research results, scholars use the matrix oblique cancellation method to determine the original polynomial of the m-sequence according to the location of the TCF peak, and use the goodness-of-fit test to improve the accuracy of peak point detection. The accuracy of point detection needs to be further improved

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
  • M-Sequence Recognition Method Based on Sparse Autoencoder
  • M-Sequence Recognition Method Based on Sparse Autoencoder
  • M-Sequence Recognition Method Based on Sparse Autoencoder

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The embodiments of the present invention are further described in detail below.

[0024] as Figure 1 and 2 As shown, the m-sequence recognition method based on the sparse autoencoder, including the following steps:

[0025] Step 1: Generate n complete m sequences with different cycles T and different signal-to-noise ratios, as follows:

[0026] In order to verify the effectiveness of the method, the INPUT set samples of order 5 to 8 are selected to train the SAE network classification model, and the corresponding original polynomials are 6, 6, 18, and 16, respectively, for a total of 46. For four m sequences of different periods, a random value is taken between -4 and 4dB as the signal-to-noise ratio value, and N noisy complete m sequences are generated.

[0027] Step 2: Estimate the third-order correlation function of each sequence in the range of [T / 2,T / 2] to obtain a third-order correlation matrix containing the characteristic information of the m-sequence, as follows:

...

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 m-sequence recognition method based on a sparse self-encoder. Firstly, the peak characteristic of the third-order correlation function of the m-sequence is introduced, and it is verified that the TCF estimated by using the complete periodic m-sequence or m-sequence fragment maintains good peak characteristic. Secondly, according to the characteristics of m-sequence TCF, an input sample construction method based on the third-order correlation feature vector is proposed. Finally, use the sparse autoencoder to build a feature learning network, use softmax regression to classify and identify the learned features, establish a sparse autoencoder network classification model, and input the pre-constructed samples into the model to train a model with optimal recognition performance model. The invention can effectively identify the m-sequence, and has good identification performance under the condition of low signal-to-noise ratio.

Description

Technical field [0001] The present invention belongs to the field of estimation and recognition of binary pseudo-random sequences in communications, particularly relates to a method of identifying m sequences using sparse self-encoded networks. Background [0002] Binary pseudo-random sequences are widely used in spread spectrum communication because of their good pseudo-random characteristics. The m sequence is the most representative pseudo-random sequence, and its estimation and recognition is the basis for information decryption in spread spectrum systems, so it has important theoretical significance and value to study the recognition algorithm of m sequence. [0003] In the existing literature on m-sequence recognition at home and abroad, Massey algorithm and Euclidean algorithm can achieve the purpose of identifying sequences to generate polynomials, but they are greatly affected by bit errors. The TripleCorrelation Function (TCF) method based on high-order statistical anal...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00
CPCG06F2218/00G06F2218/12
Inventor 强芳芳赵知劲杨安锋陈颖沈雷姜显扬
Owner HANGZHOU DIANZI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap