CN111984242A

CN111984242A - Method and system for decomposing synthesized signal

Info

Publication number: CN111984242A
Application number: CN202010843267.2A
Authority: CN
Inventors: 刘硕; 年夫顺; 刘毅; 陈鹏飞; 邱田华; 张海庆; 邹德军; 张维亮; 刘海岗
Original assignee: China Electronics Technology Instruments Co Ltd CETI
Current assignee: China Electronics Technology Instruments Co Ltd CETI
Priority date: 2020-08-20
Filing date: 2020-08-20
Publication date: 2020-11-24

Abstract

The present disclosure discloses a method and a system for decomposing a synthesized signal, including: acquiring a synthetic signal to be decomposed and a basic signal library; inputting the synthetic signal to be decomposed into a trained neural network model for signal decomposition to obtain a decomposed meta-signal; the neural network model is obtained by training a basic signal library. The neural network model is trained through the basic signal library, and the synthetic signal to be decomposed is input into the trained neural network model for signal decomposition, so that the decomposition of the synthetic signal is realized, and the cost is reduced.

Description

Method and system for decomposing synthesized signal

Technical Field

The present disclosure relates to a method and system for decomposition of a synthesized signal.

Background

The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.

Signal-oriented test languages include both ATLAS and ATML. ATLAS was born in the 60's of the 20 th century, and ATLAS was developed by ARINC corporation of America in 1962 to solve the problem of standardization of ATLAS in the field of aviation, and in 1976 the IEEE (institute of Electrical and Electronics Engineers) of the United states began to take over the standardization of ATLAS, and the latest IEEE Std 716 + 1995 was published by SCC20(Standard coordination Committee 20) of ATLAS in 1995. The ATLAS has some points of device independence, strong expandability and the like, but as the test requirement increases, the ATLAS gradually exposes the defects of increasingly complicated language, fuzzy signal definition and the like, so that the SCC20 starts to improve the ATLAS standard, thereby generating a new generation of test language, namely the ATLAS 2000, and establishes the IEEE1641 standard on the basis of the new generation of test language, namely the signal and test definition. The IEEE1641 standard solves the standardization problem of test descriptions and also the instrument interchange problem, but does not solve the test program migration problem. For this purpose, the IEEE has made an ATML standard family from 2002, and an IEEE 1671 series standard is generated, and test information is described in an XML language, and the standard family refers to an IEEE1641 standard, and information such as test requirements, instrument capabilities, test description and the like of a tested item (UUT) is described in a signal mode.

Wherein the BSC is a set of components defined by IEEE1641, and provides the most basic building blocks for creating signals. The BSC defines an extension mechanism, signals which are not defined in a most basic signal library can interconnect a plurality of BSC modules according to a standard framework, complex signals are generated to meet the test requirement, and a standard TSF layer is formed. The process of creating a complex signal model using a BSC is shown in fig. 7.

When the test developer possesses the signal model on the right side of fig. 7, the test developer can easily split the synthesized complex signal into the base signal before synthesis, but in many cases, the test developer only obtains the signal attribute description of the synthesized complex signal. In this case, it is difficult for the tester to perform test development work according to the attribute description of such a signal, and it is necessary to decompose the signal and obtain the basic signal for synthesizing the signal.

Currently, the following two methods are available for the synthesis signal decomposition:

1. need follow the artifical reverse decomposition of the attribute of synthetic signal according to experience, the manual work carries out the signal reverse decomposition, wastes time and energy, and the cost of labor is high, inefficiency, consuming time for a long time, and often can appear when synthetic signal complexity is higher, the situation that can't decompose almost.

2. Software developers are required to develop synthetic signal decomposition software through a traditional programming method based on logic symbols, difficulty is high, redevelopment is required under the condition that a basic signal library is changed, and reusability is poor.

Disclosure of Invention

In order to solve the above problems, the present disclosure provides a method and a system for decomposing a synthesized signal, in which a neural network model is trained through a basic signal library, and a synthesized signal to be decomposed is input into the trained neural network model for signal decomposition, so that the decomposition of the synthesized signal is realized, and the cost is reduced.

In order to achieve the purpose, the following technical scheme is adopted in the disclosure:

in one or more embodiments, a method of decomposition of a composite signal is presented, comprising:

acquiring a synthetic signal to be decomposed and a basic signal library;

inputting the synthetic signal to be decomposed into a trained neural network model for signal decomposition to obtain a decomposed meta-signal;

the neural network model is obtained by training a basic signal library.

In one or more embodiments, a system for decomposition of a composite signal is presented, comprising:

the basic signal synthesis module selects signals in the basic signal library to synthesize to obtain a synthesized new signal;

and the neural network training module is used for creating a neural network to be trained, training the neural network according to the synthesized new signal and the signal participating in the synthesis, and acquiring a trained neural network model.

And the signal decomposition module is used for inputting the synthetic signal to be decomposed into the trained neural network model to decompose the signal.

In one or more embodiments, a computer storage medium is provided for storing computer instructions which, when executed by a processor, perform the steps of the method for decomposition of a composite signal.

Compared with the prior art, the beneficial effect of this disclosure is:

1. this is disclosed trains neural network model through basic signal storehouse, carries out signal decomposition in the well-trained neural network model of synthetic signal input that will treat to decompose, has realized the decomposition of synthetic signal, and the cost is reduced has solved and has now carried out the reverse decomposition of signal through the manual work, wastes time and energy, and the cost of labor is high, inefficiency, consuming time of a specified duration, and often can appear when synthetic signal complexity is higher, the problem that can't decompose almost.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application.

FIG. 1 is an overall flow diagram of the disclosed method;

FIG. 2 is a basic signal synthesis flow diagram of the present disclosure;

FIG. 3 is a neural network model training flow diagram of the present disclosure;

fig. 4 is a decomposition flow diagram of a composite signal of the present disclosure.

FIG. 5 is a block diagram of the architecture of the system of the present disclosure;

FIG. 6 is a prior art hierarchy model relating to the IEEE1641 standard;

fig. 7 illustrates a conventional process for creating a complex signal model through a BSC.

The specific implementation mode is as follows:

the present disclosure is further described with reference to the following drawings and examples.

It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.

In the present disclosure, terms such as "upper", "lower", "left", "right", "front", "rear", "vertical", "horizontal", "side", "bottom", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only relational terms determined for convenience in describing structural relationships of the parts or elements of the present disclosure, and do not refer to any parts or elements of the present disclosure, and are not to be construed as limiting the present disclosure.

In the present disclosure, terms such as "fixedly connected", "connected", and the like are to be understood in a broad sense, and mean either a fixed connection or an integrally connected or detachable connection; may be directly connected or indirectly connected through an intermediate. The specific meanings of the above terms in the present disclosure can be determined on a case-by-case basis by persons skilled in the relevant art or technicians, and are not to be construed as limitations of the present disclosure.

Example 1

Each layer is built on top of the previous layer. It includes a signal modeling language layer, a basic signal component layer, a test signal framework layer, and a test procedure language layer, as shown in fig. 6. Wherein the BSC is a set of components defined by IEEE1641, and provides the most basic building blocks for creating signals. The BSC defines an extension mechanism, signals which are not defined in a most basic signal library can interconnect a plurality of BSC modules according to a standard framework, complex signals are generated to meet the test requirement, and a standard TSF layer is formed. The process of creating a complex signal model using a BSC is shown in fig. 7.

When the test developer possesses the signal model on the right side of fig. 7, the test developer can easily split the synthesized complex signal into the base signal before synthesis, but in many cases, the test developer only obtains the signal attribute description of the synthesized complex signal. In this case, it is difficult for the tester to perform test development work according to the attribute description of such signals, and the following two methods are currently available for the composite signal decomposition:

1. manual inverse decomposition from the properties of the synthesized signal is required based on experience, and the process is time-consuming and labor-consuming.

Therefore, the embodiment discloses a decomposition method of a synthesized signal, which is used for performing decomposition work from a complex synthesized signal to a basic signal, training a neural network model through a basic signal library, inputting the synthesized signal to be decomposed into the trained neural network model for signal decomposition, realizing the decomposition of the synthesized signal, reducing the cost, and solving the problems that the signal is reversely decomposed through manpower, time and labor are wasted, the labor cost is high, the efficiency is low, the time and the time are long, and the decomposition can be hardly performed when the complexity of the synthesized signal is higher.

It should be noted that the IEEE1641 basic signal library allows extension of signals, which allows creation of new signals besides signal synthesis, and created signals may describe the signals by themselves, and the method for decomposing synthesized signals disclosed in this embodiment cannot decompose the signals described by itself.

An Artificial Neural Network (ANN) abstracts a human brain neuron Network from the information processing perspective, establishes a certain simple model, and forms different networks according to different connection modes. A neural network is an operational model, which is formed by a large number of nodes (or neurons) connected to each other.

An artificial neural network is a nonlinear, adaptive information processing system composed of a large number of interconnected processing units. It is proposed on the basis of modern neuroscience research results, and tries to process information by simulating brain neural network processing and information memorizing modes. Artificial neural networks have four basic features:

(1) non-linear. The nonlinear relationship is a common feature in nature, and the artificial neuron is in two different states of activation or inhibition, and the behavior is mathematically expressed as a nonlinear relationship.

(2) And not limitation. A neural network is typically formed by a plurality of widely connected neurons. The overall behavior of a system depends not only on the characteristics of the individual neurons, but may be primarily determined by the interactions, interconnections, between the units. The non-limitation of the brain is simulated by a large number of connections between the units.

(3) Very qualitative. The artificial neural network has the self-adaption, self-organization and self-learning capabilities. The neural network not only can process various information, but also can process information while the nonlinear dynamical system is continuously changed, and the change process is described by adopting an iterative process.

(4) Non-convex. The direction of evolution of a system will, under certain conditions, depend on a particular state function. For example an energy function, the extreme values of which correspond to a more stable state of the system. Non-convexity means that the function has a plurality of extreme values, so that the system has a plurality of stable equilibrium states, which leads to the diversity of the system evolution.

The artificial neural network is a parallel distributed system, adopts a mechanism completely different from the traditional artificial intelligence and information processing technology, overcomes the defects of the traditional artificial intelligence based on logic symbols in the direction of processing intuition and unstructured information, and has the characteristics of self-adaption, self-organization and real-time learning.

A method of decomposition of a composite signal, comprising:

acquiring a synthetic signal to be decomposed and a basic signal library;

inputting the synthetic signal to be decomposed into a trained neural network model for signal decomposition to obtain a decomposed meta-signal, wherein the meta-signal is a basic signal forming the synthetic signal to be decomposed and is contained in a basic signal library;

the neural network model is obtained by training a basic signal library.

Further, verifying the decomposition result;

and recombining the decomposed meta-signals, comparing the recombined signals with the synthesized signals to be decomposed, if the recombined signals are the same, the verification is passed, and if the recombined signals are different, the verification is not passed.

Preferably, if the verification is passed, outputting a decomposition result;

and if the verification fails, retraining the neural network model, and decomposing the synthetic signal to be decomposed through the retrained neural network model until the decomposition result passes the verification.

Further, the training process of the neural network model is as follows:

creating a neural network to be trained;

acquiring a basic signal library;

randomly selecting basic signals in a basic signal library to synthesize, and acquiring new synthesized signals;

inputting all new signals into a neural network for pre-decomposition, and verifying a pre-decomposition result;

and when the pre-decomposition results of all the new signals are correct, the training of the neural network model is completed.

Preferably, the process of verifying the result of the pre-decomposition comprises:

and comparing the pre-decomposed signal with a basic signal selected by the new synthesis signal, wherein if the pre-decomposed signal is consistent with the basic signal, the pre-decomposed result is correct, and if the pre-decomposed result is inconsistent with the basic signal, the pre-decomposed result is wrong.

Preferably, when the pre-decomposition result is wrong, the new signal is re-pre-decomposed until the pre-decomposition result is verified to be correct.

Preferably, the selected attributes of all the basic signals are merged to perform signal synthesis.

Preferably, the selected basis signals are the same or different.

The basic signal library is an IEEE1641-BSC basic signal library, the basic signal library allows the extension of signals, the extension mode allows the creation of new signals besides signal synthesis, and the created signals can describe the signals by themselves.

The decomposition method of the synthesized signal is specifically discussed by combining the basic signal synthesis module, the neural network training module and the signal decomposition module.

The cooperative relationship among the basic signal synthesis module, the neural network training module, and the signal decomposition module, as shown in fig. 1, specifically includes:

step one, inputting an IEEE1641-BSC basic signal library into a basic signal synthesis module;

step two, a basic signal synthesis module selects basic signals in a basic signal library at will and synthesizes the basic signals to obtain new signals through a signal synthesis method, wherein the signal attributes obtained through synthesis are obtained by taking and collecting all basic signal attributes participating in synthesis;

recording the signals and attributes obtained by synthesis and the basic signals and the number of the signals participating in the synthesis;

inputting the basic signals and the number of the basic signals participating in the synthesis, the signals obtained by the synthesis and the attributes into a neural network training module;

step five, setting n to be 0, inputting the synthesized signals and the number into an unfinished signal decomposition module obtained by neural network training for pre-decomposition, if a correct result is obtained, inputting the synthesized signals and the attributes and the basic signals and the number participating in the synthesis into the neural network training, establishing a network topological relation, and obtaining a neural network model;

step six, after the continuous decomposition is carried out for m times, m is equal to n, wherein the size of m is defined by a user, the neural network model is considered to be trained to achieve the expected effect, the training result is output to obtain a signal decomposition module, and if m is greater than n, the training process is continued in the step 2;

step seven, inputting the synthetic signal and attribute to be decomposed and an IEEE1641-BSC basic signal library as input into a signal decomposition module;

step eight, obtaining element signals and number sets belonging to an IEEE1641-BSC basic signal library;

step nine, combining the element signals obtained after decomposition, and verifying the element signals with the synthetic signals and attributes before decomposition;

step ten, if the verification is passed, the decomposition is proved to be correct, the step eleven is continued, if the verification is failed, the decomposition is proved to be wrong, and the step two is continued;

step eleven, outputting a decomposition result;

and step twelve, ending.

The basic signal synthesis module process, as shown in fig. 2, includes the following specific steps:

step one, inputting an IEEE1641-BSC basic signal library;

secondly, constructing an attribute object of each signal in a basic signal library, and performing attribute description in a class mode, wherein specific attribute units are also required to be described;

integrating all the described objects into a set;

step four, randomly selecting n signal objects in the set, wherein the selected signal objects can be repeated or can not be different from each other;

step five, signal combination of the n signal objects is carried out, and the specific method is that attributes of all signals are collected;

recording the synthesized signal and recording the synthesized signal set;

recording and synthesizing n kinds of signals selected by the signals, and recording a basic signal set;

and step eight, finishing.

The training process of the neural network model in the neural network training module is shown in fig. 3:

step one, establishing neural network training according to a neural network algorithm;

step two, inputting an IEEE1641-BSC basic signal library;

inputting a synthesized signal set, wherein the number of elements in the signal set is n, and n is greater than 0;

step four, initializing i to be 0;

judging whether i is larger than n;

step six, if i > n is false, continuing the step seven, and if i > n is true, skipping to the step fourteen;

step seven, selecting the ith element in the signal set;

step eight, training through a neural network;

step nine, obtaining a basic signal set through training;

step ten, comparing the basic signal set obtained by training with a basic signal set selected by actual synthesis, and judging whether the basic signal set is consistent;

step eleven, if the two are consistent, continuing to step twelve, and if the two are not consistent, skipping to step thirteen;

step twelve, i is i +1, and the step five is skipped;

step thirteen, adjusting the neural training method, and skipping to step eight;

fourteen, outputting the trained neural network model into a signal decomposition module;

and step fifteen, ending.

The process of using the signal decomposition module is shown in fig. 4, and the specific steps are as follows:

step one, instantiating a signal decomposition module;

step two, inputting an IEEE1641-BSC basic signal library;

inputting a signal to be synthesized;

step four, obtaining a decomposed meta-signal set;

step five, carrying out signal synthesis on the obtained element signals again;

step six, judging whether the synthesized signal is the same as the input synthesized signal to be decomposed, if so, continuing the step seven, and if not, skipping to the step eight;

step seven, outputting the decomposed meta-signal set, and skipping to step ten;

step eight, prompting a decomposition error and requesting to check a signal decomposition module;

step nine, carrying out neural network training again;

and step ten, finishing.

The method for decomposing the synthetic signal provided in the embodiment is used for solving the problem that the synthetic signal in the signal testing development process is decomposed into the basic signal under the condition of an unknown signal model, the neural network model is trained through the basic signal used by the synthetic signal, the synthetic signal to be decomposed is input into the trained neural network model for decomposition, the extremely difficult work of signal decomposition is solved, the signal decomposition is converted into training by utilizing the characteristics of the neural network, the obtained neuron topological structure is used for decomposing the synthetic signal, the software development workload and the development difficulty are greatly reduced, and the working cost is reduced.

Under the condition that the BSC basic signal library is changed, the multiplexing of the whole code can be realized only by modifying the input IEEE1641-BSC basic signal library without redeveloping internal logic, and the reusability of the device is greatly improved.

Example 2

In this embodiment, a decomposition system of a synthesized signal is disclosed, and the structure is shown in fig. 5, in the drawing, a signal decomposition module marked on the right side of a dotted line and a signal decomposition module marked on the left side of the dotted line are the same module, which specifically includes:

Example 3

This embodiment discloses a computer storage medium storing computer instructions which, when executed by a processor, perform the steps of a method of decomposition of a composite signal as described in embodiment 1.

The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Although the present disclosure has been described with reference to specific embodiments, it should be understood that the scope of the present disclosure is not limited thereto, and those skilled in the art will appreciate that various modifications and changes can be made without departing from the spirit and scope of the present disclosure.

Claims

1. A method of decomposition of a composite signal, comprising:

acquiring a synthetic signal to be decomposed and a basic signal library;

the neural network model is obtained by training a basic signal library.

2. A method of decomposition of a composite signal according to claim 1, further comprising verifying the decomposition result;

and recombining the decomposed meta-signals, comparing the recombined signal with the synthesized signal needing to be decomposed, if the recombined signal is the same as the synthesized signal, the verification is passed, and if the recombined signal is different from the synthesized signal, the verification is not passed.

3. A method of decomposing a composite signal according to claim 2, wherein if the verification is passed, the decomposition result is outputted;

4. The method of claim 1, wherein the training process of the neural network model comprises:

creating a neural network to be trained;

acquiring a basic signal library;

selecting basic signals in a basic signal library to synthesize, and acquiring new synthesized signals;

5. A method of decomposition of a composite signal as claimed in claim 4, characterized in that the verification of the result of the pre-decomposition is carried out by:

6. A method of decomposing a composite signal as claimed in claim 5, characterized in that in the event of an error in the result of the pre-decomposition, the new signal is re-pre-decomposed until the result of the pre-decomposition verifies correct.

7. A method of decomposing a composite signal according to claim 4, characterized in that the signal is synthesized by taking the union set of the attributes of all the selected basis signals.

8. A method of decomposition of a composite signal as claimed in claim 1, characterized in that the selected basis signals are identical or different.

9. A system for decomposition of a composite signal, comprising:

10. A computer storage medium storing computer instructions which, when executed by a processor, perform the steps of a method of decomposition of a composite signal according to any one of claims 1 to 8.