WO2023169332A1

WO2023169332A1 - Method for regenerating signal by model

Info

Publication number: WO2023169332A1
Application number: PCT/CN2023/079585
Authority: WO
Inventors: 盖玉梅
Original assignee: 盖玉梅
Priority date: 2022-03-08
Filing date: 2023-03-03
Publication date: 2023-09-14
Also published as: CN114721463A

Abstract

The present invention relates to the technical field of signal processing. Disclosed is a method for regenerating a signal by a model. The present invention comprises the following steps: A, establishing a model library composed of models; B, when a native signal file is stored, sampling a native signal per unit time in the time line direction, and recording fragment feature information at each sampling position, and a model number or algorithm code equivalent to a current sampling signal fragment in the model library to form a digital record file; and C, reading the record file in the step B, reading out or calculating corresponding signal fragments from the model library according to the corresponding fragment feature information, and continuing the signal fragments according to the time line to form a regeneration signal again. According to the present invention, the lightweight recording of an original analog signal or a digital signal by the digital record file can be realized, and high-rate storage and transmission of big data can be realized only by storing and transmitting the digital record file. Moreover, the characteristics of the analog signal of being easy to read, easy to edit, easy to calculate, low in noise, and high in anti-interference performance can also be realized.

Description

A method for model regeneration of signals

Technical field

The invention belongs to the technical field of signal processing, specifically a method for model regeneration of signals.

Background technique

Most signals in nature exist as analog signals, such as sound, temperature, light, color, pressure, displacement, etc. Digital signals are values obtained by artificially digital sampling of analog signals. Because digital signals have strong anti-interference and are easy to process and calculate, digital technology is currently widely used in various industries. However, digital signals are typical point-like discrete signals. Only a very small part of the original linear analog signal is sampled, which is a great discard of the original signal source and reduces the calculation accuracy. Therefore, increasing the sampling rate per unit time is to improve digital calculations. The direct method of accuracy is limited by the pursuit of high sampling rate and computing power. The disadvantages of digital computing such as large data volume, large redundancy, and high power consumption are becoming more and more obvious, and the bandwidth occupied by big data transmission is getting higher and higher.

Compared with digital signal processing, analog signals have significant characteristics such as easy collection, high information content, low energy consumption, low calculation workload, and low data redundancy. They are increasingly valued by computing, communications, medical, military, audio and video and other industries. However, due to the The signal is a continuous signal, which has defects such as difficulty in editing, easy interference, and high noise floor, which severely limits the current development of analog technology.

Contents of the invention

The purpose of the present invention is to address the above problems and provide a method for model regenerating signals, which includes the following steps:

A. Establish a model library composed of models;

B. When the native signal file is stored, the native signal is sampled per unit time in the direction of the time axis, and the segment characteristic information at each sampling point is recorded, as well as the model number or algorithm code equivalent to the current sampled signal segment in the model library, forming a record in digital form. document;

C. Read the record file in step B and read it from the model library according to the corresponding fragment feature information. Or calculate the corresponding signal fragments, and the signal fragments are connected back and forth according to the time axis to form a regenerated signal;

The model is a segment signal or algorithm code of the signal unit time length.

Preferably, the model is a signal segment of unit time length reconstructed based on sampling segments of analog signals and digital signals per unit time or a signal segment of unit time length formed based on a specific algorithm of analog signals and digital signals.

Preferably, the record files and model library can be stored or transmitted separately.

Preferably, the segment characteristic information includes sampling frequency, time, level value, model number or model algorithm code.

Compared with the prior art, the beneficial effects of the present invention are as follows:

The invention can realize the lightweight recording of original analog or digital signals by digital recording files. It only needs to store and transmit the recording files in digital form to achieve high-rate storage and transmission of big data; it can also realize the easy-to-read analog signals. Easy to edit, easy to calculate, low noise and high anti-interference.

Description of drawings

Figure 1 is a voltage signal waveform diagram of a native analog signal in an embodiment;

Figure 2 is the digital sampling chart per unit time of Figure 1;

Figure 3 is an analog signal of the basic level value in the embodiment;

Figure 4-6 is a waveform diagram of the regenerated analog signal after modulation in the embodiment.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

In the present invention, first, a segment model library or a segment model algorithm of unit time length of analog or digital signals is established. The model library can be established through extensive sampling of analog or digital signals that are currently visible to humans (or calculated by forming a specific algorithm), that is, per unit time in the direction of the time axis. The analog or digital signal is digitally sampled, and the signal segments within each sampling segment are reconstructed to form an analog or digital segment model of the reference potential, and the model number, time and other information of the signal segment are marked in digital form.

When acquiring and storing new analog or digital files, you only need to record basic information such as time, level value, frequency and other sampling points per unit time, as well as the model number or algorithm code in the model library that is highly similar to the current signal segment to form a digital Record files with necessary characteristics required to reproduce the original signal in the form.

When reading the digital record file, the corresponding model number is read or calculated in the model library, and based on the corresponding level value and other data, it is modulated into the same or highly similar to the original analog signal or digital signal segment. According to the time axis data, before and after Connected into a continuous analog signal or digital signal.

Example

Figure 1 is a voltage signal waveform diagram of an analog signal, and then perform digital sampling per unit time on this analog signal (Figure 2). The time and level value corresponding to the sampling point and the model number corresponding to the waveform between the two sampling points Or algorithm code is recorded to form a record file in digital form.

When the analog signal in Figure 1 needs to be read out, the digital record file is read, and the corresponding analog signal segment model (or formed through algorithm calculation) is sequentially read out in the model library according to the model number to form an analog signal with a basic level value. (As shown in Figure 3, the basic level value can be either a low level reference or a high level reference). According to the recorded file level value, the basic level segment analog signal is modulated to the original signal through the analog line. Level value, the modulated analog signal segments are connected from end to end according to the direction of the time axis into a continuous complete analog signal (as shown in Figures 4, 5, and 6).

The storage and regeneration process of digital signals is basically completed through digital lines, which is basically the same principle and method as the above-mentioned storage and regeneration of analog signals.

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations are mutually exclusive. any such actual relationship or sequence exists between them. Furthermore, the term "includes", "Comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or otherwise. Includes elements inherent to such process, method, article or equipment.

Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art will understand that various changes, modifications, and substitutions can be made to these embodiments without departing from the principles and spirit of the invention. and modifications, the scope of the invention is defined by the appended claims and their equivalents.

Claims

A method for model regeneration of signals, characterized by including the following steps:

A. Establish a model library composed of models;

B. When the native signal file is stored, the native signal is sampled per unit time in the direction of the time axis, and the segment characteristic information at each sampling point is recorded, as well as the model number or algorithm code equivalent to the current sampled signal segment in the model library, forming a record in digital form. document;

C. Read the record file in step B, read out or calculate the corresponding signal fragments from the model library according to the corresponding fragment feature information, and the signal fragments are continuously regenerated according to the time axis to form a regenerated signal;

The model is a segment signal or algorithm code of the signal unit time length.
A method for regenerating signals from a model according to claim 1, characterized in that the model is a signal segment of unit time length reconstructed based on sampling segments of an analog signal and a digital signal within a unit time or is based on an analog signal and a digital signal. A signal segment of unit time length formed by a signal-specific algorithm.
A method for model regeneration of signals according to claim 1, characterized in that the record file and the model library can be stored or transmitted separately.
A method for model regenerating signals according to claim 1, characterized in that the segment characteristic information includes sampling frequency, time, level value, model number or model algorithm code.