CN113677027A - Method and system for predicting future transmission of node based on machine learning - Google Patents

Method and system for predicting future transmission of node based on machine learning Download PDF

Info

Publication number
CN113677027A
CN113677027A CN202110958194.6A CN202110958194A CN113677027A CN 113677027 A CN113677027 A CN 113677027A CN 202110958194 A CN202110958194 A CN 202110958194A CN 113677027 A CN113677027 A CN 113677027A
Authority
CN
China
Prior art keywords
machine learning
frequency
spectrum
prediction
node
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.)
Withdrawn
Application number
CN202110958194.6A
Other languages
Chinese (zh)
Inventor
叶卓雅
胡广生
舒鹏
陈柏合
赖梓烨
陈思凡
秦云霄
白双佳
杨睿诗
陈诚斌
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.)
Fuzhou Jiju Intelligent Technology Co ltd
Original Assignee
Fuzhou Jiju Intelligent Technology Co ltd
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 Fuzhou Jiju Intelligent Technology Co ltd filed Critical Fuzhou Jiju Intelligent Technology Co ltd
Priority to CN202110958194.6A priority Critical patent/CN113677027A/en
Publication of CN113677027A publication Critical patent/CN113677027A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0453—Resources in frequency domain, e.g. a carrier in FDMA
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/147—Network analysis or design for predicting network behaviour
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Databases & Information Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention provides a method and a system for predicting node future transmission based on machine learning, which are characterized in that: extracting frequency spectrum information of the radio signal; inputting the acquired frequency spectrum information into a convolutional neural network for feature extraction and classification; and judging and distinguishing the nodes according to the extracted features and predicting the future transmission mode of the nodes. The method takes the frequency use condition of an incumbent as a feature, and combines a machine learning method to extract the feature so as to realize the prediction and planning of the frequency use.

Description

Method and system for predicting future transmission of node based on machine learning
Technical Field
The invention relates to the technical field of wireless communication, in particular to a method and a system for predicting future transmission of nodes based on machine learning.
Background
In wireless communication, frequency is an important resource. However, in a general space-time scenario, the frequency occupied by each user node appears to be "random" and "random", which makes it difficult to allocate frequency resources or protect the frequency occupation of the current user (incumbent) most reasonably in the conventional technical sense.
Disclosure of Invention
In order to overcome the defects and shortcomings in the prior art, the invention provides a novel method and a novel system for predicting the future transmission of nodes based on machine learning. The frequency use condition of an incumbent is taken as a feature, and feature extraction is carried out by combining a machine learning method so as to realize prediction and planning of frequency use.
The technical scheme is as follows:
a method for predicting node future transmission based on machine learning, comprising the steps of:
step S1: extracting frequency spectrum information of the radio signal;
step S2: inputting the acquired frequency spectrum information into a convolutional neural network for feature extraction and classification;
step S3: and judging and distinguishing the nodes according to the extracted features and predicting the future transmission mode of the nodes.
Further, the current or future available frequency band is broadcasted through the cooperation information according to the judgment and prediction results.
Further, the technique adopted for extracting the spectrum information of the radio signal is as follows: a continuous stream of collected spectral data, output only in time and frequency if a given technique exists in a two-dimensional spectral voxel; and then transforming and sampling by adopting fast Fourier transform.
Further, when the resolution of the acquired frequency information is insufficient, judgment is carried out through the frequency operation details shared in a cooperation mode, and the acquisition is continuously detected until an incumbent is detected.
That is, when the information collected by the incumbent does not match and the uncertainty is high (it is indicated that the resolution is insufficient), the collected information is defaulted to be from other interference, and is not defined as the incumbent, the default incumbent is not detected yet, and the detection and collection are continued, so that the anti-interference effect is achieved (that is, the detail compensation is performed through the cooperation information).
Further, the acquired spectrum samples are concatenated with the MAC frame and form a one-to-one mapping.
Further, before the convolutional neural network training, the process of repeating the spectrum clustering is also included.
A system for predicting node future transmissions based on machine learning, comprising: a technology identification module and a repeated spectrum use mode prediction module; the technology identification module comprises a radio frequency monitor and an AI model feeder; the repeated spectrum use mode prediction module comprises a repeated spectrum clustering module and a convolutional neural network module;
the radiofrequency monitor is used for the continuous spectral data stream collected, only in time and frequency if a given technique is present in a two-dimensional spectral voxel;
the AI model feeder connects the spectrum samples acquired by the radio frequency monitor with the MAC frame to provide a one-to-one mapping between the outputs of the transmit and receive scheduling and technology identification modules;
the technology identification module is used as the input of the repeated spectrum use mode prediction module, and the repeated spectrum use mode prediction module judges and distinguishes the nodes according to the extracted features and predicts the future transmission mode of the nodes.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements steps S2-S3 of the method for machine learning based prediction of future transmissions of a node as described above.
A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements steps S2-S3 of the method for predicting future transmissions of nodes based on machine learning as described above.
The invention and the preferred scheme thereof combine the cooperative information, predict the behavior of other CIRN by analyzing and evaluating the shared extra information, such as frequency spectrum usage and maximally using local knowledge, cooperate how to influence the performance of the intelligent radio network (CIRN), can recognize, learn and actively predict the periodic transmission mode of the existing node in near real time, the precision exceeds 95%, and can predict the future existing transmission of all nodes.
Drawings
The invention is described in further detail below with reference to the following figures and detailed description:
FIG. 1 is a system framework and workflow diagram of an embodiment of the invention.
Detailed Description
In order to make the features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail as follows:
as shown in fig. 1, the system solution for predicting node future transmission based on machine learning provided by this embodiment includes: a technology identification module and a repeated spectrum usage pattern prediction module.
Wherein the technical identification (TR): the method is mainly used for respectively identifying spectrum signals and idle signals of different radio technologies;
(1) this module uses a continuous spectrum data stream collected by a radio frequency monitor (RF-MON) with a noisy background radio signal as input, and outputs only in time and frequency if a given technique is present in a two-dimensional spectrum voxel.
(2) Instead of using raw samples, 32 average 512 point Fast Fourier Transform (FFT) samples are used, with a sample frequency of 23.04Msps (46.08 Msps if the available bandwidth is 20 MHz), which can be compensated by using incumbents with co-sharing frequency operation details if the frequency resolution is insufficient to be discernable.
(3) One module after RF-MON, is the AI model feeder. The RF-MON samples are concatenated with the MAC frames that are concatenated to the present system. It provides a one-to-one mapping between the transmit and receive schedules and the output of the TR module.
For Repetitive Spectrum Usage Pattern Prediction (RSUPP):
the module adopts machine learning-based repetitive spectrum clustering and a convolutional neural network for training so as to learn and predict the working mode of the nodes in real time.
The scheme combines the cooperative information, the behaviors of other CIRNs are predicted by analyzing and evaluating the shared additional information such as GPS position or frequency spectrum usage and maximally using local knowledge, how the cooperation affects the performance of the intelligent radio network (CIRN), the periodic transmission mode of the existing node can be identified, learned and actively predicted in near real time, the precision exceeds 95%, and the future existing transmission of all nodes can be predicted.
It can be realized that: spectrum reserved when a protected incumbent is active and a sub-band is available is shared.
Dynamically taking advantage of the period of time that the incumbent is away from the radio.
Collaborative information is used to minimize the possibility of injury to protective incumbents.
Further, based on feature extraction, it can implement the following working mechanism to achieve prediction and planning of frequency usage:
1. in an ideal case, the TR module of a radio detects an incumbent if the incumbent is in the vicinity of a given radio.
2. By filtering this information from the other radio signatures, voxels with and without incumbents detected are judged.
3. Upon receipt of this output, as well as the incumbent report, the RSUPP will create a boolean string for each existing channel since the incumbent report will have a time stamp to create a new view of the two-dimensional spectral voxels. It should be noted that when existing personnel report center frequency and bandwidth, we reduce the computational requirements by only processing channels that may overlap with existing personnel (existing channels).
4. This new view will be transferred back to the input, since the protected incumbent has a fixed and repetitive transmission pattern, which should learn and predict the incumbent's future use of the spectrum. It should be noted that this two-step procedure needs to support real-time decisions because it is not practical to apply online learning algorithms, such as Reinforcement Learning (RL), because the use of raw spectral data introduces a large state-motion space and long training time.
For using collaboration information to minimize the possibility of injury to protective incumbents:
in order to avoid that other users interfere with TR and cannot identify the incumbent, the cooperation information plays a fundamental role. The spectral power measurements shared by incumbents reduce the uncertainty of existing transmissions when the TR module is blind, and these measures provide information that RSUPP can take as a basic truth that the incumbent exists in the event that the TR is blind due to interference of other CIRNs with the incumbent. It may work in reverse, if an incumbent reports interference, but the TR height of a given radio believes that the incumbent is not detected, it may share the full spectrum with other CIRNs, as it does not cause such interference.
The algorithm model provided by this embodiment can be stored in a computer readable storage medium in a coded form, and implemented in a computer program manner, and inputs basic parameter information required for calculation through computer hardware, and outputs a calculation result.
The algorithm model provided by this embodiment can be stored in a computer readable storage medium in a coded form, and implemented in a computer program manner, and inputs basic parameter information required for calculation through computer hardware, and outputs a calculation result.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
The present invention is not limited to the above preferred embodiments, and other various forms of methods and systems for predicting node future transmissions based on machine learning may be devised by those skilled in the art, and all equivalent changes and modifications made herein without departing from the scope of the invention are intended to be covered by the appended claims.

Claims (9)

1. A method for predicting node future transmission based on machine learning, comprising the steps of:
step S1: extracting frequency spectrum information of the radio signal;
step S2: inputting the acquired frequency spectrum information into a convolutional neural network for feature extraction and classification;
step S3: and judging and distinguishing the nodes according to the extracted features and predicting the future transmission mode of the nodes.
2. The method of machine learning based prediction of node future transmissions according to claim 1, characterized by: broadcasting the current or future available frequency band through the cooperation information according to the judgment and prediction results.
3. The method of machine learning based prediction of node future transmissions according to claim 1, characterized by: the technology for extracting the frequency spectrum information of the radio signal comprises the following steps: a continuous stream of collected spectral data, output only in time and frequency if a given technique exists in a two-dimensional spectral voxel; and then transforming and sampling by adopting fast Fourier transform.
4. The method of machine learning based prediction of node future transmissions according to claim 4, characterized by: and when the resolution ratio of the acquired frequency information is insufficient, judging through the frequency operation details shared in a cooperation mode, and continuing to detect and acquire until an incumbent is detected.
5. The method of machine learning based prediction of node future transmissions according to claim 4, characterized by: and connecting the acquired spectrum samples with the MAC frame and forming a one-to-one mapping.
6. The method of machine learning based prediction of node future transmissions according to claim 1, characterized by: before the convolutional neural network training, the process of repeating the spectrum clustering is also included.
7. A system for predicting node future transmissions based on machine learning, comprising: a technology identification module and a repeated spectrum use mode prediction module; the technology identification module comprises a radio frequency monitor and an AI model feeder; the repeated spectrum use mode prediction module comprises a repeated spectrum clustering module and a convolutional neural network module;
the radiofrequency monitor is used for the continuous spectral data stream collected, only in time and frequency if a given technique is present in a two-dimensional spectral voxel;
the AI model feeder connects the spectrum samples acquired by the radio frequency monitor with the MAC frame to provide a one-to-one mapping between the outputs of the transmit and receive scheduling and technology identification modules;
the technology identification module is used as the input of the repeated spectrum use mode prediction module, and the repeated spectrum use mode prediction module judges and distinguishes the nodes according to the extracted features and predicts the future transmission mode of the nodes.
8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements steps S2-S3 of the method of machine learning based predictive node future transmissions of claim 1.
9. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements steps S2-S3 of the method of machine learning based predictive node future transmissions of claim 1.
CN202110958194.6A 2021-08-20 2021-08-20 Method and system for predicting future transmission of node based on machine learning Withdrawn CN113677027A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110958194.6A CN113677027A (en) 2021-08-20 2021-08-20 Method and system for predicting future transmission of node based on machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110958194.6A CN113677027A (en) 2021-08-20 2021-08-20 Method and system for predicting future transmission of node based on machine learning

Publications (1)

Publication Number Publication Date
CN113677027A true CN113677027A (en) 2021-11-19

Family

ID=78544147

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110958194.6A Withdrawn CN113677027A (en) 2021-08-20 2021-08-20 Method and system for predicting future transmission of node based on machine learning

Country Status (1)

Country Link
CN (1) CN113677027A (en)

Similar Documents

Publication Publication Date Title
Datla et al. A spectrum surveying framework for dynamic spectrum access networks
CN101399564B (en) Dual-time collaboration detection method for sensing idle frequency spectrum in radio frequency
CN101895895B (en) Method and device for sensing graded frequency spectra in cognitive radio network
CN109743356B (en) Industrial internet data acquisition method and device, readable storage medium and terminal
US10241973B2 (en) System and method for reducing size of raw data by performing time-frequency data analysis
US20210241441A1 (en) Methods, Apparatus and Computer-Readable Mediums Relating to Detection of Cell Conditions in a Wireless Cellular Network
US11895440B2 (en) Methods and apparatus to perform remote monitoring
CN113159145A (en) Characteristic engineering arrangement method and device
CN116488750A (en) Electromagnetic interference influence analysis method and system based on neural network
Girmay et al. Machine learning enabled Wi-Fi saturation sensing for fair coexistence in unlicensed spectrum
Honghao et al. Spectrum anomalies autonomous detection in cognitive radio using hidden markov models
Yi et al. Interference source identification for ieee 802.15. 4 wireless sensor networks using deep learning
CN110856206A (en) Interference identification method and device and terminal equipment
CN111150392A (en) Directed dynamic brain function network multi-class emotion recognition construction method and device
CN114760172A (en) Method and device for identifying radio frequency baseband comprehensive characteristic signal
Dong et al. Exploration of singular spectrum analysis for online anomaly detection in crns
CN113677027A (en) Method and system for predicting future transmission of node based on machine learning
CN116668986A (en) Human body sensing method, device, equipment and medium based on general sense integrated route
CN110996289A (en) Intelligent gateway, method and system for machine state monitoring diagnostic instrument
Zhang et al. Intelligent fault diagnosis system based on vibration signal edge computing
KR102386382B1 (en) A method and apparatus for forecasting saturation of cell capacity in a wireless communication system
Burago et al. Bandwidth-aware data filtering in edge-assisted wireless sensor systems
CN107864489B (en) Signal prediction method, system and signal detection method
CN113810500B (en) Crowd gathering risk identification method and related equipment
Webber et al. Human Motion Identity using Machine Learning on Spectral Analysis of RSS Signals

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20211119

WW01 Invention patent application withdrawn after publication