CN114491403A - Edge calculation data processing method, device and equipment and readable storage medium - Google Patents

Edge calculation data processing method, device and equipment and readable storage medium Download PDF

Info

Publication number
CN114491403A
CN114491403A CN202210099362.5A CN202210099362A CN114491403A CN 114491403 A CN114491403 A CN 114491403A CN 202210099362 A CN202210099362 A CN 202210099362A CN 114491403 A CN114491403 A CN 114491403A
Authority
CN
China
Prior art keywords
sequence
preset
sparse
matrix
signal sequence
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210099362.5A
Other languages
Chinese (zh)
Inventor
宋洋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Agricultural Bank of China
Original Assignee
Agricultural Bank of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Agricultural Bank of China filed Critical Agricultural Bank of China
Priority to CN202210099362.5A priority Critical patent/CN114491403A/en
Publication of CN114491403A publication Critical patent/CN114491403A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
    • H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
    • H03M7/3059—Digital compression and data reduction techniques where the original information is represented by a subset or similar information, e.g. lossy compression

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Algebra (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Abstract

The invention provides a method, a device and equipment for processing edge calculation data and a readable storage medium, wherein the method comprises the following steps: acquiring an acquisition signal sequence to be compressed; compressing the collected signal sequence according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding measurement projection sequence; the preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data; the preset observation matrix is irrelevant to the preset sparse basis matrix; and sending the measurement projection sequence to the electronic equipment of the sink node. According to the method, the acquisition signal sequence to be compressed is compressed through the preset observation matrix and the preset sparse basis matrix, the characteristics of the data do not need to be analyzed, the requirement of data processing on hardware computing performance is lowered, the data processing effect in an edge computing scene is improved, and therefore the data transmission efficiency is improved.

Description

Edge calculation data processing method, device and equipment and readable storage medium
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, and a readable storage medium for processing edge calculation data.
Background
The edge computing technology is in high-speed development, and is limited by the data communication technology, and how to occupy less communication resources when data is transmitted from the edge back to the cloud is one of the core problems that must be solved. Generally, when data transmission is performed, data collected by the sensing node is compressed and transmitted to the sink node, and the sink node decompresses the data and then restores the original data.
The traditional data compression method mainly compresses data which are collected by a sensing node, the characteristics of the data are firstly analyzed in the compression process, and then redundant data in signals are removed through a certain algorithm, so that the purpose of data compression is achieved. However, since the edge computing device of the sensing node has low computation performance, it is difficult to perform the above-mentioned complicated data compression operation. Therefore, conventional network data compression techniques are not suitable for edge computing scenarios.
Therefore, the data processing effect in the current edge computing scene is poor, so that the data transmission occupies more resources and the transmission efficiency is low.
Disclosure of Invention
The invention provides a method, a device and equipment for processing edge calculation data and a readable storage medium, which are used for solving the problems that data transmission occupies more resources and the transmission efficiency is lower due to the poor data processing effect in the current edge calculation scene.
The invention provides an edge calculation data processing method, which is applied to electronic equipment sensing nodes and comprises the following steps:
acquiring an acquisition signal sequence to be compressed;
compressing the acquired signal sequence according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding measurement projection sequence; the preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data; the preset observation matrix is uncorrelated with the preset sparse basis matrix;
and sending the measurement projection sequence to the electronic equipment of the sink node.
Further, the method as described above, the compressing the acquired signal sequence according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding measured projection sequence, includes:
performing sparse transformation on the acquired signal sequence according to the preset sparse basis matrix to generate a corresponding sparse coefficient sequence; the sequence length of the sparse coefficient sequence is smaller than the sequence length of the acquisition signal sequence;
and carrying out random linear projection processing on the sparse coefficient sequence according to the preset observation matrix so as to determine a corresponding measurement projection sequence.
Further, according to the method, the preset observation matrix is a gaussian random measurement matrix, and the preset sparse basis matrix is a fast cosine transform basis.
Further, the method as described above, before the acquiring the acquisition signal sequence to be compressed, further includes:
carrying out sparse transformation processing on an original signal sequence acquired by electronic equipment of a sensing node within a preset time period to determine a corresponding sparse coefficient sequence;
judging whether the original signal sequence is compressible or not according to the corresponding sparse coefficient sequence and a preset sparse judgment algorithm;
and if the original signal sequence is determined to be compressible, determining the original signal sequence as an acquisition signal sequence to be compressed.
A second aspect of the present invention provides a method for processing edge calculation data, which is applied to an electronic device of a sink node, and the method includes:
receiving a measurement projection sequence sent by electronic equipment of a sensing node;
reducing the measurement projection sequence according to a preset reduction algorithm, a preset observation matrix and a preset sparse basis matrix to generate a corresponding acquisition signal sequence; and the preset observation matrix is irrelevant to the preset sparse basis matrix.
Further, in the method as described above, the predetermined reduction algorithm is a basis pursuit algorithm;
the reducing processing is performed on the measurement projection sequence according to a preset reducing algorithm, a preset observation matrix and a preset sparse basis matrix to generate a corresponding acquisition signal sequence, and the reducing processing comprises the following steps:
performing inverse random linear projection processing on the measurement projection sequence by adopting a basis tracking algorithm and the preset observation matrix to determine a corresponding sparse coefficient sequence;
and performing inverse sparse transform processing on the sparse coefficient sequence by adopting a basis pursuit algorithm and the preset sparse basis matrix to generate a corresponding acquisition signal sequence.
The third aspect of the present invention provides an edge calculation data processing apparatus, an electronic device located at a sensing node, the apparatus comprising:
the acquisition module is used for acquiring an acquisition signal sequence to be compressed;
the compression module is used for compressing the acquired signal sequence according to a preset observation matrix and a preset sparse basis matrix so as to generate a corresponding measurement projection sequence; the preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data;
and the sending module is used for sending the measurement projection sequence to the electronic equipment of the sink node.
Further, in the apparatus as described above, the compression module is specifically configured to:
performing sparse transformation on the acquired signal sequence according to the preset sparse basis matrix to generate a corresponding sparse coefficient sequence; the sequence length of the sparse coefficient sequence is smaller than the sequence length of the acquisition signal sequence; and carrying out random linear projection processing on the sparse coefficient sequence according to the preset observation matrix so as to determine a corresponding measurement projection sequence.
Further, according to the apparatus as described above, the preset observation matrix is a gaussian random measurement matrix, and the preset sparse basis matrix is a fast cosine transform basis.
Further, the apparatus as described above, further comprising:
the judging module is used for carrying out sparse transformation processing on an original signal sequence acquired by the electronic equipment of the sensing node within a preset time period so as to determine a corresponding sparse coefficient sequence; judging whether the original signal sequence is compressible or not according to the corresponding sparse coefficient sequence and a preset sparse judgment algorithm; and if the original signal sequence is determined to be compressible, determining the original signal sequence as an acquisition signal sequence to be compressed.
A fourth aspect of the present invention provides an edge calculation data processing apparatus, an electronic device located at a sink node, the apparatus including:
the receiving module is used for receiving a measurement projection sequence sent by the electronic equipment of the sensing node;
the reduction module is used for reducing the measurement projection sequence according to a preset reduction algorithm, a preset observation matrix and a preset sparse basis matrix so as to generate a corresponding acquisition signal sequence; and the preset observation matrix is irrelevant to the preset sparse basis matrix.
Further, in the apparatus as described above, the preset reduction algorithm is a basis pursuit algorithm;
the reduction module is specifically configured to:
performing inverse random linear projection processing on the measurement projection sequence by adopting a basis tracking algorithm and the preset observation matrix to determine a corresponding sparse coefficient sequence; and performing inverse sparse transform processing on the sparse coefficient sequence by adopting a basis pursuit algorithm and the preset sparse basis matrix to generate a corresponding acquisition signal sequence.
A fifth aspect of the present invention provides an electronic apparatus, comprising: at least one processor, memory, and transceiver;
the processor, the memory and the transceiver circuitry are interconnected;
the memory stores computer-executable instructions; the transceiver is used for transceiving data;
the at least one processor executes computer-executable instructions stored by the memory, causing the at least one processor to perform the edge computation data processing method of any one of the first or second aspects.
A sixth aspect of the present invention provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the computer-executable instructions are used for implementing the edge calculation data processing method according to any one of the first aspect or the second aspect.
A seventh aspect of the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the edge calculation data processing method of any one of the first or second aspects.
The invention provides a method, a device and equipment for processing edge calculation data and a readable storage medium, wherein the method comprises the following steps: acquiring an acquisition signal sequence to be compressed; compressing the acquired signal sequence according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding measurement projection sequence; the preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data; the preset observation matrix is uncorrelated with the preset sparse basis matrix; and sending the measurement projection sequence to the electronic equipment of the sink node. According to the edge calculation data processing method, the acquisition signal sequence to be compressed is compressed through the preset observation matrix and the preset sparse basis matrix, the characteristics of the data do not need to be analyzed, the requirement of the data processing on the hardware calculation performance is reduced, the data processing effect in an edge calculation scene is improved, and therefore the data transmission efficiency is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
FIG. 1 is a diagram of a scenario in which an edge calculation data processing method according to an embodiment of the present invention may be implemented;
FIG. 2 is a flowchart illustrating a method for processing edge calculation data according to a first embodiment of the present invention;
FIG. 3 is a schematic diagram of data transformation of an edge calculation data processing method according to a first embodiment of the present invention;
FIG. 4 is a flowchart illustrating a method for processing edge calculation data according to a second embodiment of the present invention;
FIG. 5 is a schematic structural diagram of an edge calculation data processing apparatus according to a third embodiment of the present invention;
FIG. 6 is a schematic structural diagram of an edge calculation data processing apparatus according to a fourth embodiment of the present invention;
fig. 7 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention.
With the above figures, certain embodiments of the invention have been illustrated and described in more detail below. The drawings and the description are not intended to limit the scope of the inventive concept in any way, but rather to illustrate it by those skilled in the art with reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.
The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present invention will be described below with reference to the accompanying drawings.
For a clear understanding of the technical solutions of the present application, a detailed description of the prior art solutions is first provided. The edge computing technology is in a high-speed development, and is limited by resources of a data communication technology, and how to transmit data from an edge to a cloud with low consumption is one of core problems that must be solved.
In general, methods for reducing transmission consumption of nodes in an edge computing network are mainly classified into three categories, which are: duty cycle techniques, data driven techniques, and mobile node techniques. They improve on three aspects of networking communication, perception and collection of the network respectively. The sensing mode is that data collected by the sensing nodes are compressed and then transmitted.
In general, the sensing coverage areas of nodes in a network overlap, so that correlation exists between data collected by different nodes, and therefore, the collected data can be compressed. The traditional data compression method mainly compresses data which is collected by nodes, and in the compression process, all data are subjected to feature extraction, and then redundant data in signals are removed through a certain algorithm, so that the purpose of data compression is achieved. However, in the data collection process of the wireless sensor network, the sensing nodes in the network have simple computing power, so that the complex data compression operation is difficult to perform. Therefore, conventional network data compression techniques are not suitable for use in rimless computing scenarios.
Therefore, the data processing effect in the current edge computing scene is poor, so that the data transmission occupies more resources and the transmission efficiency is low.
Therefore, the inventor finds that, in order to solve the problem in the prior art, a compressive sensing theory can be combined to solve the problem, wherein the problem is that data transmission occupies more resources and has lower transmission efficiency due to poor data processing effect in an edge computing scene. The perception theory considers that because data collected by nodes in a network have strong spatio-temporal correlation, most information in the data can be obtained by only projecting a small amount of the data, and heavy operation work in the data encoding and decoding process is transferred from the perception nodes to the aggregation nodes. In addition, the minimum requirement of Nyquist sampling theorem on sampling frequency is broken through, and the node can accurately acquire the information of the original signal without high-frequency sampling. The reduction in the number of samples significantly reduces the communication resources consumed by the data transmission of the sampling device.
Specifically, an acquisition signal sequence to be compressed is obtained, and the acquisition signal sequence is compressed according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding measurement projection sequence. The preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data. And the preset observation matrix is irrelevant to the preset sparse basis matrix. And then, sending the measurement projection sequence to the electronic equipment of the aggregation node to realize data transmission. According to the edge calculation data processing method, the compression processing is carried out on the to-be-compressed acquisition signal sequence through the preset observation matrix and the preset sparse basis matrix, the self characteristics of data do not need to be analyzed, the requirement of data processing on hardware calculation performance is lowered, the data processing effect in an edge calculation scene is improved, and therefore the data transmission efficiency is improved.
The inventor proposes a technical scheme of the application based on the creative discovery.
An application scenario of the edge calculation data processing method provided by the embodiment of the present invention is described below. As shown in fig. 1, 1 is a first electronic device, which is an electronic device of a sink node, and 2 is a second electronic device, which is an electronic device of a sense node. The second electronic device 2 stores the acquisition signal sequence to be compressed, and the second electronic device 2 may also be an electronic device for acquiring signal data, and may compress the acquisition signal sequence simultaneously when acquiring signal data. When the edge calculation data transmission is performed, the second electronic device 2 compresses the collected signal sequence according to the preset observation matrix and the preset sparse basis matrix to generate a corresponding measurement projection sequence, wherein the preset observation matrix and the preset sparse basis matrix are uncorrelated. The second electronic device 2 then sends the measured projection sequence to the first electronic device 1. After receiving the measurement projection sequence, the first electronic device 1 performs reduction processing on the measurement projection sequence according to a preset reduction algorithm, a preset observation matrix and a preset sparse basis matrix to generate a corresponding acquisition signal sequence, thereby completing transmission of edge calculation data. After generating the acquisition signal sequence, the first electronic device 1 may perform data analysis on the acquisition signal sequence, or send the acquisition signal sequence to other electronic devices for corresponding processing.
The embodiments of the present invention will be described with reference to the accompanying drawings.
Fig. 2 is a schematic flow chart of a method for processing edge calculation data according to a first embodiment of the present invention, and as shown in fig. 2, in this embodiment, an execution subject of the embodiment of the present invention is an edge calculation data processing apparatus, and the edge calculation data processing apparatus may be integrated in an electronic device of a sensor node. The method for processing edge calculation data provided by this embodiment includes the following steps:
and step S101, acquiring a signal acquisition sequence to be compressed.
In this embodiment, the acquisition signal sequence to be compressed may be natural environment data such as temperature and humidity, or network data such as communication signal data.
In this embodiment, the sensing node can compress the collected signal sequence while collecting the signal, and simultaneously, the collected signal frequency can be lower than that of the prior art, and the nyquist sampling theorem does not need to be met.
And S102, compressing the collected signal sequence according to the preset observation matrix and the preset sparse basis matrix to generate a corresponding measurement projection sequence. The preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data. And the preset observation matrix is irrelevant to the preset sparse basis matrix.
In this embodiment, the preset observation matrix and the preset sparse basis matrix need to satisfy an irrelevant condition, or the preset observation matrix may satisfy a limited equidistance. The irrelevancy may be determined by a deterministic manner of finite equidistance.
The preset observation matrix can adopt a Gaussian random measurement matrix and is mainly used for carrying out random linear projection processing on the sequence data and uniformly changing high-dimensional sequence data into low-dimensional sequence data.
The sequence length of the measured projection sequence is smaller than the sequence length of the acquired signal sequence.
And step S103, sending the measurement projection sequence to the electronic equipment of the sink node.
In this embodiment, after the measurement projection sequence is determined, the measurement projection sequence may be sent to the electronic device of the sink node to complete data transmission, and meanwhile, the electronic device of the sink node performs reduction processing on the measurement projection sequence to obtain an acquisition signal sequence.
In order to better understand the method for processing edge calculation data of the present invention, a specific process of compressing the collected signal sequence according to the preset observation matrix and the preset sparse basis matrix to generate the corresponding measurement projection sequence will be further described below:
and carrying out sparse transformation on the acquired signal sequence according to a preset sparse basis matrix to generate a corresponding sparse coefficient sequence. The sequence length of the sparse coefficient sequence is smaller than the sequence length of the acquisition signal sequence.
And carrying out random linear projection processing on the sparse coefficient sequence according to a preset observation matrix so as to determine a corresponding measurement projection sequence.
In this embodiment, the preset sparse basis matrix may adopt a fast fourier transform basis or a fast cosine transform basis, and the fast cosine transform basis is preferred. The preset observation matrix adopts a Gaussian random measurement matrix.
The specific process can be described by the formula of compressed sensing and fig. 3, and the formula is as follows:
y＝Φx＝ΦΨs
wherein y represents a measurement projection sequence, phi represents a preset observation matrix, psi represents a preset sparse basis matrix, x represents an acquisition signal sequence, and s represents a sparse coefficient sequence. The preset observation matrix and the preset sparse basis matrix can also be combined to be called a perception matrix. The number of non-zero elements in the sparse coefficient sequence is generally smaller than the sequence length of the sparse coefficient sequence.
The preset sparse base matrix denoted by Ψ may be transposed as required, that is, the signal sequence is changed from the acquisition signal sequence to the sparse coefficient sequence, and the sparse base matrix is changed from the sparse coefficient sequence to the acquisition signal sequence, which both need to be preset, but may be transposed as required.
In the compressive sensing theory, a sparse coefficient sequence is obtained firstly, then random linear projection is carried out on the sparse coefficient sequence, and a sufficient amount of characteristic value information of the sparse coefficient sequence is obtained, so that signal compression is realized, and the process is mainly realized by an observation matrix.
In fig. 3, the dimension of the measurement projection sequence is M dimension, the measurement projection sequence includes a plurality of feature values of the sparse coefficient sequence, the dimension of S is N dimension, and M is smaller than N. The sparse coefficient sequence is a sequence set of sparse coefficients of the acquired signal sequence after sparse transformation. The dimension of the preset sparse base matrix is N x N, and the dimension of the preset observation matrix is M x N. The essence of the compressed sensing is that a small amount of projection operation is performed on a sparse signal, a small amount of characteristic information capable of characterizing data is randomly acquired, and then the original data is recovered from the small amount of characteristic information by utilizing the correlation among the data.
Meanwhile, the acquisition signal sequence of the embodiment is compressible, so that whether the acquisition signal sequence is compressible can be judged in advance before compression, specifically:
and carrying out sparse transformation processing on an original signal sequence acquired by the electronic equipment of the sensing node within a preset time period to determine a corresponding sparse coefficient sequence.
And judging whether the original signal sequence is compressible according to the corresponding sparse coefficient sequence and a preset sparse judgment algorithm.
And if the original signal sequence is determined to be compressible, determining the original signal sequence as the acquisition signal sequence to be compressed.
In this embodiment, the preset sparse determination algorithm is:
|si|≤C·i-q
wherein s isiC and q are constants for ith data in the sparse coefficient sequence, and the larger the q value is, the faster the coefficient attenuation rate after transformation is. When the sparse coefficient sequence satisfies the preset sparse judgment algorithm, it can be determined that the original signal sequence is compressible.
The embodiment of the invention provides an edge calculation data processing method, which comprises the following steps: and acquiring an acquisition signal sequence to be compressed. And compressing the acquired signal sequence according to the preset observation matrix and the preset sparse basis matrix to generate a corresponding measurement projection sequence. The preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data. And the preset observation matrix is irrelevant to the preset sparse basis matrix. And sending the measurement projection sequence to the electronic equipment of the sink node.
According to the edge calculation data processing method, the compression processing is carried out on the to-be-compressed acquisition signal sequence through the preset observation matrix and the preset sparse basis matrix, the self characteristics of data do not need to be analyzed, the requirement of data processing on hardware calculation performance is lowered, the data processing effect in an edge calculation scene is improved, and therefore the data transmission efficiency is improved.
Fig. 4 is a schematic flowchart of an edge calculation data processing method according to a second embodiment of the present invention, and as shown in fig. 4, an execution subject of the embodiment of the present invention is an edge calculation data processing apparatus, and the edge calculation data processing apparatus may be integrated in an electronic device of a sink node. The edge calculation data processing method provided by the present embodiment includes the following steps.
Step S201, receiving a measurement projection sequence sent by the electronic device of the sensing node.
In this embodiment, the electronic device of the sink node may be in network connection with the electronic devices of the plurality of sensing nodes, and after receiving the measurement projection sequence sent by the electronic device of the sensing node, the measurement projection sequence may be restored.
And S202, reducing the measurement projection sequence according to a preset reduction algorithm, a preset observation matrix and a preset sparse basis matrix to generate a corresponding acquisition signal sequence. And the preset observation matrix is irrelevant to the preset sparse basis matrix.
In this embodiment, when the preset observation matrix is not related to the preset sparse basis matrix, the original acquisition signal sequence can be restored by using the preset restoration algorithm, the preset observation matrix and the preset sparse basis matrix according to the limited equidistant property characteristic reconstruction of the sparse coefficient sequence.
The specific process of reduction is as follows: the predetermined reduction algorithm of the present embodiment adopts a basis tracking algorithm.
And performing inverse random linear projection processing on the measurement projection sequence by adopting a basis tracking algorithm and a preset observation matrix to determine a corresponding sparse coefficient sequence.
And performing inverse sparse transformation processing on the sparse coefficient sequence by adopting a basis tracking algorithm and a preset sparse basis matrix to generate a corresponding acquisition signal sequence.
The reduction process is a process of solving the system of underdetermined equations y ═ Φ Ψ s. This is a zero-norm minimization problem, a problem without a fast solution, and therefore is often converted to a solution of a one-norm minimization of s, or an algorithm with some approximate estimation. Thus, the present embodiment calculates the optimal solution by the basis pursuit algorithm. Meanwhile, the Bayesian estimation algorithm can be used as a reduction algorithm.
The basis pursuit algorithm is a typical algorithm. The frequency non-zero value of the original acquired signal still retains a larger value in the frequency domain after the random linear projection, wherein the larger values can be detected by setting a first threshold value and combining a base tracking algorithm with a preset observation matrix. Assuming that only two non-zero values of the signal are present, the interference caused by these two non-zero values can be calculated. And after the interference is removed, setting a second threshold value to obtain a sparse coefficient sequence.
And then, carrying out inverse sparse transformation processing on the sparse coefficient sequence to obtain the original acquisition signal sequence.
Fig. 5 is a schematic structural diagram of an edge calculation data processing apparatus according to a third embodiment of the present invention, and as shown in fig. 5, in this embodiment, the edge calculation data processing apparatus 300 is located in an electronic device of a sensing node, and the edge calculation data processing apparatus 300 includes:
an obtaining module 301, configured to obtain an acquisition signal sequence to be compressed.
A compressing module 302, configured to compress the acquired signal sequence according to the preset observation matrix and the preset sparse basis matrix to generate a corresponding measurement projection sequence. The preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data.
And a sending module 303, configured to send the measurement projection sequence to the electronic device of the sink node.
Optionally, in this embodiment, the compression module 302 is specifically configured to:
and carrying out sparse transformation on the acquired signal sequence according to a preset sparse basis matrix to generate a corresponding sparse coefficient sequence. The sequence length of the sparse coefficient sequence is smaller than the sequence length of the acquisition signal sequence. And carrying out random linear projection processing on the sparse coefficient sequence according to a preset observation matrix so as to determine a corresponding measurement projection sequence.
Optionally, in this embodiment, the preset observation matrix is a gaussian random measurement matrix, and the preset sparse basis matrix is a fast cosine transform basis.
Optionally, in this embodiment, the edge calculation data processing apparatus 300 further includes:
and the judging module is used for performing sparse transformation processing on an original signal sequence acquired by the electronic equipment of the sensing node within a preset time period so as to determine a corresponding sparse coefficient sequence. And judging whether the original signal sequence is compressible according to the corresponding sparse coefficient sequence and a preset sparse judgment algorithm. And if the original signal sequence is determined to be compressible, determining the original signal sequence as the acquisition signal sequence to be compressed.
The edge calculation data processing apparatus provided in this embodiment may execute the technical solutions of the method embodiments shown in fig. 2 and fig. 3, and the implementation principle and the technical effect of the apparatus are similar to those of the method embodiments shown in fig. 2 and fig. 3, and are not described in detail here.
Meanwhile, fig. 6 is a schematic structural diagram of an edge calculation data processing apparatus according to a fourth embodiment of the present invention, and as shown in fig. 6, in order to better distinguish the edge calculation data processing apparatus according to the present embodiment from the edge calculation data processing apparatus according to the previous embodiment, the edge calculation data processing apparatus according to the previous embodiment is an edge calculation data processing apparatus 300, and the edge calculation data processing apparatus according to the present invention is an edge calculation data processing apparatus 400. Both refer to the edge computing data processing device, but the modules included in the device are different, and the applied electronic devices are also different. The edge calculation data processing apparatus 400 of this embodiment is located in an electronic device of a sink node, and the edge calculation data processing apparatus 400 includes:
a receiving module 401, configured to receive a measurement projection sequence sent by an electronic device of a sensing node.
A restoring module 402, configured to restore the measurement projection sequence according to a preset restoring algorithm, a preset observation matrix, and a preset sparse basis matrix, so as to generate a corresponding acquisition signal sequence. And the preset observation matrix is irrelevant to the preset sparse basis matrix.
Optionally, in this embodiment, the preset reduction algorithm is a basis pursuit algorithm.
The restoring module 402 is specifically configured to:
and performing inverse random linear projection processing on the measurement projection sequence by adopting a basis tracking algorithm and a preset observation matrix to determine a corresponding sparse coefficient sequence. And performing inverse sparse transformation processing on the sparse coefficient sequence by adopting a basis tracking algorithm and a preset sparse basis matrix to generate a corresponding acquisition signal sequence.
The edge calculation data processing apparatus provided in this embodiment may execute the technical solution of the method embodiment shown in fig. 4, and the implementation principle and technical effect thereof are similar to those of the method embodiment shown in fig. 4, and are not described in detail herein.
The invention also provides an electronic device, a computer readable storage medium and a computer program product according to the embodiments of the invention.
As shown in fig. 7, fig. 7 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention. Electronic devices are intended for various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 7, the electronic apparatus includes: a processor 501, a memory 502, and a transceiver 503. The various components are interconnected using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions for execution within the electronic device.
The memory 502 is a non-transitory computer readable storage medium provided by the present invention. The memory stores instructions executable by the at least one processor, so that the at least one processor executes the edge calculation data processing method provided by the invention. The non-transitory computer-readable storage medium of the present invention stores computer instructions for causing a computer to execute the edge calculation data processing method provided by the present invention.
The memory 502, which is a non-transitory computer-readable storage medium, may be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the edge calculation data processing method in the embodiment of the present invention (for example, the obtaining module 301, the compressing module 302, and the transmitting module 303 shown in fig. 5 or the receiving module 401 and the restoring module 402 shown in fig. 7). The processor 501 executes various functional applications of the server and data processing, i.e., implements the edge calculation data processing method in the above method embodiment, by running non-transitory software programs, instructions, and modules stored in the memory 502. The transceiver 503 is used for transceiving data.
Meanwhile, the present embodiment also provides a computer product, and when instructions in the computer product are executed by a processor of the electronic device, the electronic device is enabled to execute the edge calculation data processing method of the first to second embodiments.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the embodiments of the invention following, in general, the principles of the embodiments of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the embodiments of the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the embodiments of the invention being indicated by the following claims.
It is to be understood that the embodiments of the present invention are not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of embodiments of the invention is limited only by the appended claims.

Claims (10)

1. An edge computing data processing method applied to an electronic device sensing nodes, the method comprising:
acquiring an acquisition signal sequence to be compressed;
compressing the acquired signal sequence according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding measurement projection sequence; the preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data; the preset observation matrix is uncorrelated with the preset sparse basis matrix;
and sending the measurement projection sequence to the electronic equipment of the sink node.
2. The method of claim 1, wherein compressing the sequence of acquired signals according to a preset observation matrix and a preset sparse basis matrix to generate a corresponding sequence of measured projections comprises:
performing sparse transformation on the acquired signal sequence according to the preset sparse basis matrix to generate a corresponding sparse coefficient sequence; the sequence length of the sparse coefficient sequence is smaller than the sequence length of the acquisition signal sequence;
and carrying out random linear projection processing on the sparse coefficient sequence according to the preset observation matrix so as to determine a corresponding measurement projection sequence.
3. The method of claim 2, wherein the predetermined observation matrix is a gaussian random measurement matrix and the predetermined sparse basis matrix is a fast cosine transform basis.
4. The method of claim 1, wherein the obtaining the sequence of acquisition signals to be compressed further comprises:
carrying out sparse transformation processing on an original signal sequence acquired by electronic equipment of a sensing node within a preset time period to determine a corresponding sparse coefficient sequence;
judging whether the original signal sequence is compressible or not according to the corresponding sparse coefficient sequence and a preset sparse judgment algorithm;
and if the original signal sequence is determined to be compressible, determining the original signal sequence as an acquisition signal sequence to be compressed.
5. An edge computing data processing method applied to an electronic device of a sink node, the method comprising:
receiving a measurement projection sequence sent by electronic equipment of a sensing node;
reducing the measurement projection sequence according to a preset reduction algorithm, a preset observation matrix and a preset sparse basis matrix to generate a corresponding acquisition signal sequence; and the preset observation matrix is irrelevant to the preset sparse basis matrix.
6. The method of claim 5, wherein the predetermined reduction algorithm is a basis pursuit algorithm;
the reducing processing is performed on the measurement projection sequence according to a preset reducing algorithm, a preset observation matrix and a preset sparse basis matrix to generate a corresponding acquisition signal sequence, and the reducing processing comprises the following steps:
performing inverse random linear projection processing on the measurement projection sequence by adopting a basis pursuit algorithm and the preset observation matrix to determine a corresponding sparse coefficient sequence;
and performing inverse sparse transform processing on the sparse coefficient sequence by adopting a basis pursuit algorithm and the preset sparse basis matrix to generate a corresponding acquisition signal sequence.
7. An edge computing data processing apparatus, characterized by an electronic device located at a sensing node, the apparatus comprising:
the acquisition module is used for acquiring an acquisition signal sequence to be compressed;
the compression module is used for compressing the acquired signal sequence according to a preset observation matrix and a preset sparse basis matrix so as to generate a corresponding measurement projection sequence; the preset observation matrix is used for carrying out random linear projection processing on the sequence data, and the sparse basis matrix is used for carrying out sparse transformation processing on the sequence data;
and the sending module is used for sending the measurement projection sequence to the electronic equipment of the sink node.
8. An edge computing data processing apparatus, characterized in that an electronic device located at a sink node, the apparatus comprises:
the receiving module is used for receiving a measurement projection sequence sent by the electronic equipment of the sensing node;
the reduction module is used for reducing the measurement projection sequence according to a preset reduction algorithm, a preset observation matrix and a preset sparse basis matrix so as to generate a corresponding acquisition signal sequence; and the preset observation matrix is irrelevant to the preset sparse basis matrix.
9. An electronic device, comprising: at least one processor, memory, and transceiver;
the processor, the memory and the transceiver circuitry are interconnected;
the memory stores computer-executable instructions; the transceiver is used for transceiving data;
the at least one processor executing the computer-executable instructions stored by the memory causes the at least one processor to perform the edge computation data processing method of any one of claims 1 to 4 or of claim 5 or 6.
10. A computer-readable storage medium having computer-executable instructions stored therein, which when executed by a processor, are configured to implement the edge calculation data processing method according to any one of claims 1 to 4 or according to claim 5 or 6.
CN202210099362.5A 2022-01-27 2022-01-27 Edge calculation data processing method, device and equipment and readable storage medium Pending CN114491403A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210099362.5A CN114491403A (en) 2022-01-27 2022-01-27 Edge calculation data processing method, device and equipment and readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210099362.5A CN114491403A (en) 2022-01-27 2022-01-27 Edge calculation data processing method, device and equipment and readable storage medium

Publications (1)

Publication Number Publication Date
CN114491403A true CN114491403A (en) 2022-05-13

Family

ID=81477205

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210099362.5A Pending CN114491403A (en) 2022-01-27 2022-01-27 Edge calculation data processing method, device and equipment and readable storage medium

Country Status (1)

Country Link
CN (1) CN114491403A (en)

Similar Documents

Publication Publication Date Title
CN110535869B (en) Data transmission method based on compression algorithm, terminal equipment and storage medium
CN111628946B (en) Channel estimation method and receiving equipment
CN107113101A (en) Non- for compressed sensing owes fixed estimation
CN107135518B (en) Method and device for maintaining heartbeat connection and terminal equipment
CN111404853B (en) Carrier frequency offset estimation method, device and computer storage medium
CN111465051B (en) SINR estimation method, system, electronic device and storage medium
WO2015094537A1 (en) Methods and apparatuses for detecting anomalies in the compressed sensing domain
CN105761215A (en) Image compression method, image reconstruction method, device and system
JP2023515846A (en) A Phase-Based Approach for Ultrasound
CN110611626B (en) Channel estimation method, device and equipment
Melek et al. ECG compression using wavelet-based compressed sensing with prior support information
CN114390519A (en) Wireless channel key generation method, device, equipment and storage medium
CN114491403A (en) Edge calculation data processing method, device and equipment and readable storage medium
Zhang et al. Side information based orthogonal matching pursuit in distributed compressed sensing
CN104104390A (en) Signal compression method, signal reconstruction method, and correlation apparatus and system
Peterson et al. Datacomp: Locally independent adaptive compression for real-world systems
WO2020080143A1 (en) Computer, sensing system, and data communication method
CN113472703B (en) OFDM channel estimation method
Zonoobi et al. Low rank and sparse matrix reconstruction with partial support knowledge for surveillance video processing
Medeiros et al. TinyML custom AI algorithms for low-power IoT data compression: A bridge monitoring case study
Zou et al. Compressive detection with sparse random projections
Lakshmi et al. Compressive estimation of UWA channels for OFDM transmission using iterative sparse reconstruction algorithms
Zhang et al. An effective source number detection method for single-channel signals based on signal reconstruction and deep learning at low SNR
CN104302017A (en) Pretreatment method for wavelet data compression in sensor network
JP6190780B2 (en) Web display waiting time estimation apparatus, method and program

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination