CN113301654B - Unmanned aerial vehicle communication frequency band distribution system and method based on semi-supervised learning algorithm - Google Patents
Unmanned aerial vehicle communication frequency band distribution system and method based on semi-supervised learning algorithm Download PDFInfo
- Publication number
- CN113301654B CN113301654B CN202110389059.4A CN202110389059A CN113301654B CN 113301654 B CN113301654 B CN 113301654B CN 202110389059 A CN202110389059 A CN 202110389059A CN 113301654 B CN113301654 B CN 113301654B
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- module
- frequency band
- frequency
- 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.)
- Active
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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/18502—Airborne stations
- H04B7/18506—Communications with or from aircraft, i.e. aeronautical mobile service
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
Abstract
The invention discloses an unmanned aerial vehicle communication frequency band distribution system and method based on a semi-supervised learning algorithm, wherein the system comprises a ground control center and an airborne control end. Then, when the unmanned aerial vehicle normally works, the airborne control end of the unmanned aerial vehicle is utilized to carry out circulation detection on the communication frequency band information and the working condition of the current unmanned aerial vehicle, the idle conditions of other frequency bands are predicted, when the current communication frequency band is interfered, the frequency band switching can be rapidly carried out, the normal telemetering information and data transmission of the unmanned aerial vehicle are prevented from being interfered unnecessarily, and therefore the working efficiency and the anti-interference performance of the unmanned aerial vehicle are improved.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicle communication, in particular to an unmanned aerial vehicle communication frequency band distribution system and method based on a semi-supervised learning algorithm.
Background
In recent years, the artificial intelligence technology is gradually perfected, the rapid development of the unmanned aerial vehicle industry is promoted, and along with the continuous expansion of the application scene of the unmanned aerial vehicle, the unmanned aerial vehicle is widely applied to the aspects of aerial photography, wireless image transmission, environmental monitoring, industrial surveying and mapping and the like in the civil field. At present, the state has reserved a special frequency spectrum division frequency band for a special aerospace craft, and a civil unmanned aerial vehicle can only utilize the traditional industrial, telecommunication and medical frequency bands to complete real-time measurement and control and information map transmission with the ground, wherein 433MHz-440MHz, 840.5MHz-845MHz, 1.438GHz-1.44GHz and 2.4GHz-2.44GHz are available free open frequency bands, and the frequency bands of 433MHz and below are difficult to meet the bandwidth requirement of the unmanned aerial vehicle for realizing high-definition map transmission. The fixed unmanned aerial vehicle of communication frequency channel receives other interference signal's influence very easily in the use, can cause phenomena such as unmanned aerial vehicle communication interruption, image transmission failure, remote control information are malfunctioning probably.
The Chinese patent document discloses a method and a device for allocating network channel resources aggregated by unmanned aerial vehicles, which are disclosed in the publication No. CN111182640A, wherein the communication channel of the unmanned aerial vehicle is divided into time frames at equal intervals, the method is applied to a central unmanned aerial vehicle to allocate tasks for unmanned aerial vehicles of various services, and the state of a wireless network is comprehensively considered when channel resources are allocated, so that the real-time performance and the effectiveness of communication and video services of the unmanned aerial vehicles are guaranteed.
Chinese patent literature discloses a method for allocating channel resources of an unmanned aerial vehicle supporting different QoS, which is described in detail in publication No. CN112040440A, and the method includes the steps of movement prediction, flow rate test, resource allocation for video transmission and other services, and the like, and allocates channel resources according to the service flow rate requirements of the unmanned aerial vehicle, thereby ensuring the bandwidth requirements of video services, and avoiding link interruption caused by movement of the unmanned aerial vehicle.
How to detect and identify interference signals in a communication frequency band pre-used by an unmanned aerial vehicle with high efficiency and low cost, and quickly distribute and switch the communication frequency band is a difficult problem of ensuring the information transmission quality of the unmanned aerial vehicle and improving the communication anti-interference performance of the unmanned aerial vehicle at present.
The method and the device for allocating the network channel resource aggregated by the unmanned aerial vehicle are only applied to communication between the central unmanned aerial vehicle and other unmanned aerial vehicles in the working state, the central unmanned aerial vehicle classifies service data according to types, and the central unmanned aerial vehicle allocates appropriate channels to the unmanned aerial vehicles according to different service requirements. It only preferentially guarantees video services and, if there is interference on the allocated frequency band, the video transmission quality is still affected.
In the above-mentioned "method for allocating channel resources of an unmanned aerial vehicle supporting different QoS", before allocating a communication frequency band to an unmanned aerial vehicle, a minimum transmission time is calculated according to a minimum bandwidth required by a communication demand, and an actual allocation value needs to be adjusted according to a congestion condition of a next time frame channel, and if the current communication channel resources are relatively short, the allocation value may be reduced as much as possible to make more channel resources for other services.
Disclosure of Invention
The purpose of the invention is: aiming at the conditions that the current unmanned aerial vehicle communication frequency band is easily interfered by noise and the corresponding method is insufficient, the unmanned aerial vehicle communication frequency band distribution system and method based on the semi-supervised learning algorithm are provided.
In order to achieve the purpose, the invention adopts the following technical scheme: the unmanned aerial vehicle communication frequency band distribution system based on the semi-supervised learning algorithm comprises a ground control center and an airborne control end, wherein the airborne control end is installed on an unmanned aerial vehicle; the method is characterized in that: the ground control center and the airborne control end both comprise the same wireless communication module and a frequency spectrum sensing module, and the wireless communication module of the ground control center is wirelessly connected with the wireless communication module of the airborne control end; the spectrum sensing module is used for sensing the frequency spectrums of all available frequency bands of the unmanned aerial vehicle to obtain frequency spectrum data of the frequency bands occupied by the unmanned aerial vehicle and the frequency bands pre-occupied by the unmanned aerial vehicle;
the airborne control end further comprises a GPS module, a central processing unit, a data analysis module, a signal classification and prediction module, a frequency band switching module and a power supply system, wherein the GPS module is used for collecting geographic position information of the unmanned aerial vehicle in a flight state in real time and transmitting the geographic position information to the ground control center through the wireless communication module; the central processing unit is respectively in bidirectional communication connection with the wireless communication module of the airborne control end, the spectrum sensing module of the airborne control end, the GPS module, the data analysis module, the signal classification and prediction module and the frequency band switching module, and is used for sending a working instruction to the wireless communication module of the airborne control end, the spectrum sensing module of the airborne control end and the GPS module and receiving spectrum data occupied by the unmanned aerial vehicle and occupying a pre-occupied frequency band; the data analysis module is used for reading the frequency spectrum data from the central processing unit, carrying out frequency domain analysis and sending the analyzed frequency spectrum data to the signal classification and prediction module; the signal classification and prediction module is used for predicting the idle frequency band according to the frequency spectrum data to obtain idle frequency band prediction information and transmitting the idle frequency band prediction information to the frequency band switching module; the frequency band switching module is used for switching the unmanned aerial vehicle from the current interfered communication frequency band to an idle frequency band; the power supply system is respectively connected with the wireless communication module of the airborne control end, the frequency spectrum sensing module of the airborne control end, the GPS module, the central processing unit, the data analysis module, the signal classification and prediction module and the voltage input end of the frequency band switching module;
the ground control center also comprises a flight remote control module, a signal processing module and a data processing module, wherein the flight remote control module is connected with the wireless communication module of the ground control center and is used for generating a remote control command according to a user command and transmitting the remote control command to the unmanned aerial vehicle through the wireless communication module of the ground control center in a wireless communication mode; before the ground control center communicates with the unmanned aerial vehicle, the data processing module is connected with a spectrum sensing module of the ground control center and receives spectrum data of a pre-occupied frequency band of the unmanned aerial vehicle sent by the spectrum sensing module, the data processing module integrates and displays the received spectrum data, and after the pre-occupied frequency band is determined to be undisturbed, the data processing module establishes communication connection with the unmanned aerial vehicle by utilizing the frequency band; the signal processing module is respectively in two-way connection with the data processing module and the wireless communication module of the ground control center, and the signal processing module is used for filtering and denoising the received flight state signal and the frequency spectrum signal of the unmanned aerial vehicle and transmitting the signals back to the data processing module.
As a preferred scheme of the present invention, the spectrum sensing module includes a VERT2450 antenna and an NI USRP-2901 module, and the data connection between the VERT2450 antenna and the NI USRP-2901 module is provided.
An unmanned aerial vehicle communication frequency band allocation method based on a semi-supervised learning algorithm is characterized in that the method adopts the system to allocate the unmanned aerial vehicle communication frequency band, and specifically comprises the following steps:
before the unmanned aerial vehicle takes off, a frequency spectrum sensing module of a ground control center carries out frequency spectrum scanning on a frequency band pre-occupied by the unmanned aerial vehicle, a data processing module of the ground control center judges whether the frequency band pre-occupied by the unmanned aerial vehicle is in an interference state, if interference exists, the frequency band is switched, and frequency spectrum scanning is continued until an interference-free idle frequency band is found;
step two, the ground control center adopts the idle frequency band in the step one to be in communication connection with the unmanned aerial vehicle, the ground control center sends remote control information to the unmanned aerial vehicle, and the unmanned aerial vehicle receives the remote control information to take off and execute an air flight task;
in the process of executing the air flight task, the ground control center receives flight state information and image acquisition information of the unmanned aerial vehicle through the wireless communication module, and meanwhile, the spectrum sensing module of the unmanned aerial vehicle performs spectrum scanning in real time to obtain spectrum data of an occupied frequency band and a pre-occupied frequency band of the unmanned aerial vehicle, integrates and stores the spectrum data, and then sends the spectrum data to the central processing unit;
after receiving the frequency spectrum data, the central processing unit carries out quantization processing on the obtained frequency spectrum data to obtain frequency spectrum data sets with marks and without marks, and the data analysis module trains the frequency spectrum data sets with the marks and without marks by using a semi-supervised learning algorithm to obtain the trained frequency spectrum data;
inputting the frequency spectrum data trained in the step four into a signal classification and prediction module, training and classifying the frequency spectrum data by using a naive Bayes classifier through the signal classification and prediction module, training out a naive Bayes classifier model, predicting the idle probability of each frequency band through the naive Bayes classifier model, and sending the idle probability prediction result of each frequency band to a frequency band switching module to complete idle frequency band prediction;
after the idle frequency band is predicted, when the communication frequency band used by the unmanned aerial vehicle is interfered, the unmanned aerial vehicle sends a frequency band switching request to a central processing unit in the unmanned aerial vehicle, and the central processing unit controls a frequency band switching module to switch the unmanned aerial vehicle from the current interfered frequency band to the idle frequency band according to a preset distribution principle; the ground control center communicates with the unmanned aerial vehicle according to the switched frequency band, and receives flight state information and image acquisition information transmitted to the unmanned aerial vehicle;
and step seven, repeating the step two to the step six, and ensuring that the normal working communication of the unmanned aerial vehicle is not interfered until the unmanned aerial vehicle finishes the flight task.
Further, the preset allocation principle is as follows: and according to the predicted idle probability of each frequency band, allocating the frequency band with the highest idle probability to the unmanned aerial vehicle.
Through the design scheme, the invention can bring the following beneficial effects: the invention provides an unmanned aerial vehicle communication frequency band allocation system and method based on a semi-supervised learning algorithm. Firstly, the ground control center carries out frequency spectrum scanning on a pre-used frequency band, judges whether interference exists on the frequency band or not, and jumps to the next frequency band to continue frequency spectrum scanning if the interference exists on the frequency band. Then, when the unmanned aerial vehicle works normally, the airborne control end of the unmanned aerial vehicle is used for carrying out cyclic detection on the communication frequency band information and the working condition of the current unmanned aerial vehicle, the idle conditions of other frequency bands are predicted, when the current communication frequency band is interfered, the frequency bands can be rapidly switched, the situation that the normal telemetering information and data transmission of the unmanned aerial vehicle are interfered unnecessarily is avoided, and therefore the working efficiency and the anti-interference performance of the unmanned aerial vehicle are improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limitation and are not intended to limit the invention in any way, and in which:
fig. 1 is a block diagram of a communication frequency band allocation system of an unmanned aerial vehicle based on a semi-supervised learning algorithm.
Fig. 2 is a flowchart of an unmanned aerial vehicle communication frequency band allocation method based on a semi-supervised learning algorithm.
Fig. 3 is a flow chart of classification and prediction of data of communication frequency bands of the unmanned aerial vehicle based on a semi-supervised learning algorithm.
FIG. 4 is a flow diagram of a naive Bayes classifier model prediction process.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions of the present invention are clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the present invention is not limited by the following examples, and specific embodiments can be determined according to the technical solutions and practical situations of the present invention. Well-known methods, procedures, components and circuits have not been described in detail so as not to obscure the present invention.
Referring to fig. 1,2,3 and 4, the communication frequency band allocation system of the unmanned aerial vehicle based on the semi-supervised learning algorithm includes a ground control center and an airborne control end, wherein the airborne control end is installed on the unmanned aerial vehicle; the ground control center and the airborne control end both comprise the same wireless communication module and a frequency spectrum sensing module, and the wireless communication module of the ground control center is in wireless connection with the wireless communication module of the airborne control end and is used for realizing communication connection between the unmanned aerial vehicle and the ground control center and carrying out data interaction; the spectrum sensing module is used for sensing the frequency spectrums of all available frequency bands of the unmanned aerial vehicle, and the purpose of sensing the frequency spectrums of the frequency bands is to measure the frequency spectrum use conditions of the frequency bands so as to judge whether the sensed frequency bands are occupied; the frequency spectrum sensing module comprises a VERT2450 antenna and an NI USRP-2901 module, the VERT2450 antenna is used for omni-directionally acquiring communication frequency band signals and transmitting the acquired signals to the NI USRP-2901 module, and the NI USRP-2901 module is used for demodulating and decoding the signals acquired by the VERT2450 antenna so as to acquire frequency spectrum data of a frequency band occupied by the unmanned aerial vehicle and a pre-occupied frequency band; wherein the VERT2450 antenna is a dual band 2.4 to 2.48GHz and 4.9 to 5.9GHz vertical omni-directional antenna, and the NI USRP-2901 module is a tunable RF transceiver operating with full duplex MIMO, having a wide frequency range of 70MHz to 6 GHz.
The airborne control end also comprises a GPS module, a central processing unit, a data analysis module, a signal classification and prediction module, a frequency band switching module and a power supply system, wherein the GPS module is used for acquiring the geographical position information of the unmanned aerial vehicle in the flight state in real time and transmitting the information to the ground control center through the wireless communication module, so that the ground control center can master the geographical position and the flight state of the unmanned aerial vehicle in real time; the model that central processing unit adopted the STM company to produce is STM32F 407's microcontroller chip, central processing unit respectively with the wireless communication module of airborne control end, the spectrum perception module of airborne control end, the GPS module, data analysis module, signal classification and prediction module and frequency channel switching module both way communication are connected, central processing unit is used for the wireless communication module to airborne control end, the spectrum perception module and the GPS module of airborne control end send operating instruction, and the spectral data and the real-time geographical position information of unmanned aerial vehicle that receive unmanned aerial vehicle shared and pre-occupied frequency channel, central processing unit carries out quantization to the spectral data who obtains, known by prior art, the quantization, quantize in the range dimension promptly, quantize continuous amplitude into discrete, just can encode after the discretization. The specific quantization process is as follows: removing abnormal values which may cause overlarge standard deviation by performing equally-spaced discrete encoding processing on continuous amplitude dimensions, and scaling the data (such as replacing a large unit) to enable the data to fall into a specific region (such as 0-1) so as to obtain quantized spectral data; meanwhile, the wireless communication module is in communication connection with a ground control center, and real-time data acquired by the unmanned aerial vehicle are fed back to the ground control center; the data analysis module is used for reading the frequency spectrum data from the central processing unit, carrying out frequency domain analysis and sending the analyzed frequency spectrum data to the signal classification and prediction module; the signal classification and prediction module is used for predicting the idle frequency band according to the received frequency spectrum data to obtain idle frequency band prediction information and transmitting the idle frequency band prediction information to the frequency band switching module; the frequency band switching module is used for switching the unmanned aerial vehicle from the current interfered communication frequency band to a target communication frequency band so as to adjust the working frequency band of the unmanned aerial vehicle, wherein the target communication working frequency band is the frequency band with the maximum idle probability in all the predicted frequency bands; the power supply system is respectively connected with the wireless communication module of the airborne control end, the frequency spectrum sensing module of the airborne control end, the GPS module, the central processing unit, the data analysis module, the signal classification and prediction module and the voltage input end of the frequency band switching module, and is used for providing rated working voltage for each module of the unmanned aerial vehicle. In order to highlight the design point of the invention, the unmanned aerial vehicle module structure part shown in fig. 1 only comprises the functions of switching the communication frequency band, transmitting images and videos and receiving normal remote control information by the unmanned aerial vehicle and a ground control center, and does not comprise the existing structures such as the flight structure of the unmanned aerial vehicle and other functions of the unmanned aerial vehicle.
The ground control center also comprises a flight remote control module, a signal processing module and a data processing module. The ground control center is provided with a display interface for monitoring the flight and communication states of the unmanned aerial vehicle in real time. The ground control center is connected with the unmanned aerial vehicle in real-time communication through the wireless communication module, so that the operation end can master the flight state of the unmanned aerial vehicle. The flight remote control module is connected with a wireless communication module of the ground control center, the flight remote control module is used for generating a remote control instruction according to a user instruction and transmitting the remote control instruction to the unmanned aerial vehicle through the wireless communication module of the ground control center in a wireless communication mode, and the wireless communication module with strong anti-jamming capability is adopted to ensure reliable and stable transmission of remote control information, images and video information; before the ground control center communicates with the unmanned aerial vehicle, the data processing module is connected with the spectrum sensing module of the ground control center and receives spectrum data of a frequency band pre-occupied by the unmanned aerial vehicle sent by the spectrum sensing module, and the data processing module analyzes the channel occupation condition of the received spectrum data. After determining that the unoccupied frequency band is not interfered, establishing communication connection with the unmanned aerial vehicle by using the frequency band; the signal processing module is respectively in two-way connection with the data processing module and the wireless communication module of the ground control center, and the signal processing module is used for filtering and denoising the received flight state signal and the frequency spectrum signal of the unmanned aerial vehicle and transmitting the signals back to the data processing module for operation and display.
An unmanned aerial vehicle communication frequency band allocation method based on a semi-supervised learning algorithm comprises the following steps:
firstly, before the unmanned aerial vehicle takes off, a frequency spectrum sensing module of a ground control center carries out frequency spectrum scanning on frequency bands pre-occupied by the unmanned aerial vehicle, such as 433MHz-440MHz, 840.5MHz-845MHz, 1.438GHz-1.44GHz and 2.4GHz-2.44GHz, a data processing module of the ground control center judges whether the frequency bands pre-occupied by the unmanned aerial vehicle are in an interference state, if interference exists in the frequency bands expected to be used, the frequency bands are switched to the next frequency band, and frequency spectrum scanning is continued until an interference-free idle frequency band is found; after the unmanned aerial vehicle can realize normal communication with ground control center, ground control center just can send remote control information, controls unmanned aerial vehicle and carries out normal flight work.
Secondly, after the unmanned aerial vehicle works normally, a spectrum sensing module of the unmanned aerial vehicle starts to perform real-time spectrum scanning; the method comprises the following steps that a frequency spectrum sensing module of the unmanned aerial vehicle integrates and stores frequency spectrum data of a frequency band occupied by the unmanned aerial vehicle and a pre-occupied frequency band, and then sends the frequency spectrum data to a central processing unit;
meanwhile, the ground control center receives the image and video information collected by the unmanned aerial vehicle, and the GPS module sends positioning information of the unmanned aerial vehicle, so that the ground control center can master and remotely control the information of the unmanned aerial vehicle.
After receiving the frequency spectrum data, the central processing unit carries out quantization processing on the obtained frequency spectrum data, removes abnormal values which possibly cause overlarge standard deviation by carrying out equally-spaced discrete coding processing on continuous amplitude dimensions, and scales the data into a relatively small interval according to a certain proportion so as to obtain a marked frequency spectrum data set and an unmarked frequency spectrum data set, then the data analysis module obtains the frequency spectrum data set from the central processing unit, analyzes the relation between the data distribution information and the class mark by utilizing an unmarked data sample so as to obtain the optimal generalization performance of the unmarked data, and obtains the communication frequency band data information under the normal flight state of the unmanned aerial vehicle, the specific data classification and prediction process is shown in figure 3, and then a naive Bayes classifier is adopted in the signal classification and prediction module for the known marked data class, assuming that all attributes are suitable for observed unlabeled sample data, extracting the optimal generalization performance from most unlabeled sample data by using part of labeled sample data, and gradually expanding the training set in the calculation process, wherein x is a given spectral data sample obtained by spectral scanning, y isnFor a real category label dataset, the unmanned aerial vehicle communication frequency band data is: dn＝(xn,yn) And Θ {1,2, 3.., N }, where DnA data sample set to be trained; n represents a set of data samples D to be trainednThe number of categories that may exist in (1); for a given spectral data sample x obtained by spectral scanning, its true class sample data is labeled y ∈ Θ. For a given sample, first assuming it was generated by a naive bayes classifier model, the data sample is generated substantially by the following probability density:
wherein, P (y) is class prior probability, P (x) is data sample probability used as normalization proving factor, P (y | x) is posterior probability, P (x | y) is class conditional probability, and P (x | y) is class conditional probabilitynY) is the estimated conditional probability under the nth attribute, xnFor the value of x on the nth attribute, let hn(x) The predictive markers representing the first N attributes of a given spectral data sample x obtained by spectral scanning by the model h, and Θ ═ 1,2, 3.
Wherein, P (y)j| x) is the estimated conditional probability of each attribute; p (phi)i| x) is the posterior probability generated by the ith naive Bayes component, P (y)j|ΦiX) is the probability of the ith naive bayes component by x and its class being j. P (y)j,Φi| x) is the joint probability of a class j belonging to the ith naive bayes component generated by the sample x; each class corresponds to a naive Bayes component, P (y)j|ΦiX) is only related to the naive bayes component Φ to which a given sample of spectral data obtained by a spectral scan belongs. In estimating P (y)j|ΦiX) requires knowledge of the sample's label, and P (Φ)i| x) does not involve sample labeling, and both labeled and unlabeled data can be utilized.
For marked sample set Dh＝{(x1,y1),(x2,y2),(x3,y3),...,(xh,yh) And unlabeled sample set Dl＝{(xh+1,yh+1),(xh+2,yh+2),(xh+3,yh+3),...,(xh+l,yh+l) H & lt l, l + h ═ n, if the number of samples is sufficient, DqRepresents a training set DhThe prior probability of the set consisting of the medium-q samplesIt can be roughly estimated that:
for discrete attributes of data, Dq,jExpressed in the training set DqWherein the j-th attribute takes the value yjThen the conditional probability can be estimated approximately as:
for the conditional probability P (y)j| x), the probability represents the joint probability of all the attributes of x involved in the probability, and the probability can be represented as the idle probability of the measured frequency band according to the probability of the occurrence of the sample. If the capacity of the data set received by the central processing unit is large and the prediction speed of the work task is high, the probability estimation value calculated for the given data set through the naive Bayes classifier model is stored in advance. By the method, search training can be carried out according to historical data when the communication frequency band is subsequently redistributed, so that the processing efficiency is improved. After the unmanned aerial vehicle sends out a prediction request, prediction information can be transmitted to the spectrum switching module in time, and the problems of calculation redundancy, too long time consumption and the like caused by too large data processing amount are prevented. And after the calculation is finished, according to the calculated frequency band idle probability, when the current communication frequency band is interfered, the predicted idle frequency band information is transmitted to the frequency band switching module in time.
Step four, after the idle frequency band is predicted, when the communication frequency band used by the unmanned aerial vehicle currently has interference, the unmanned aerial vehicle sends a frequency band switching request to a central processing unit in the unmanned aerial vehicle, the central processing unit controls a frequency band switching module to switch the unmanned aerial vehicle from the current interfered frequency band to the idle frequency band according to a preset distribution principle, the preset distribution principle is a predicted idle probability according to each frequency band, the frequency band with the highest idle probability is distributed for the unmanned aerial vehicle, so that the interference is avoided to influence the normal communication of the unmanned aerial vehicle, and the specific signal classification flow chart is shown in fig. 4. The ground control center receives the flight state information and the image acquisition information of the unmanned aerial vehicle through the wireless communication module, then the spectrum sensing module, the data analysis module and the signal classification and prediction module of the unmanned aerial vehicle continue to scan the frequency spectrum and predict the frequency band, and then the cyclic operation is carried out from the second step, so that the normal working communication of the unmanned aerial vehicle is not interfered.
Claims (4)
1. The unmanned aerial vehicle communication frequency band distribution system based on the semi-supervised learning algorithm comprises a ground control center and an airborne control end, wherein the airborne control end is installed on an unmanned aerial vehicle; the method is characterized in that: the ground control center and the airborne control end both comprise the same wireless communication module and a frequency spectrum sensing module, and the wireless communication module of the ground control center is wirelessly connected with the wireless communication module of the airborne control end; the spectrum sensing module is used for sensing the frequency spectrums of all available frequency bands of the unmanned aerial vehicle to obtain frequency spectrum data of the frequency bands occupied by the unmanned aerial vehicle and the frequency bands pre-occupied by the unmanned aerial vehicle;
the airborne control end further comprises a GPS module, a central processing unit, a data analysis module, a signal classification and prediction module, a frequency band switching module and a power supply system, wherein the GPS module is used for collecting geographic position information of the unmanned aerial vehicle in a flight state in real time and transmitting the geographic position information to the ground control center through the wireless communication module; the central processing unit is respectively in bidirectional communication connection with the wireless communication module of the airborne control end, the spectrum sensing module of the airborne control end, the GPS module, the data analysis module, the signal classification and prediction module and the frequency band switching module, and is used for sending a working instruction to the wireless communication module of the airborne control end, the spectrum sensing module of the airborne control end and the GPS module and receiving spectrum data occupied by the unmanned aerial vehicle and occupying a pre-occupied frequency band; the data analysis module is used for reading the frequency spectrum data from the central processing unit, carrying out frequency domain analysis and sending the analyzed frequency spectrum data to the signal classification and prediction module; the signal classification and prediction module is used for predicting the idle frequency band according to the frequency spectrum data to obtain idle frequency band prediction information and transmitting the idle frequency band prediction information to the frequency band switching module; the frequency band switching module is used for switching the unmanned aerial vehicle from the current interfered communication frequency band to an idle frequency band; the power supply system is respectively connected with the wireless communication module of the airborne control end, the frequency spectrum sensing module of the airborne control end, the GPS module, the central processing unit, the data analysis module, the signal classification and prediction module and the voltage input end of the frequency band switching module;
the ground control center also comprises a flight remote control module, a signal processing module and a data processing module, wherein the flight remote control module is connected with the wireless communication module of the ground control center and is used for generating a remote control command according to a user command and transmitting the remote control command to the unmanned aerial vehicle through the wireless communication module of the ground control center in a wireless communication mode; before the ground control center communicates with the unmanned aerial vehicle, the data processing module is connected with a spectrum sensing module of the ground control center and receives spectrum data of a pre-occupied frequency band of the unmanned aerial vehicle sent by the spectrum sensing module, the data processing module integrates and displays the received spectrum data, and after the pre-occupied frequency band is determined to be free of interference, the data processing module establishes communication connection with the unmanned aerial vehicle by using the frequency band; the signal processing module is respectively in two-way connection with the data processing module and the wireless communication module of the ground control center, and the signal processing module is used for filtering and denoising the received flight state signal and the frequency spectrum signal of the unmanned aerial vehicle and transmitting the signals back to the data processing module.
2. The semi-supervised learning algorithm based unmanned aerial vehicle communication frequency band allocation system of claim 1, wherein: the frequency spectrum sensing module comprises a VERT2450 antenna and an NI USRP-2901 module, and data connection is formed between the VERT2450 antenna and the NI USRP-2901 module.
3. An unmanned aerial vehicle communication frequency band allocation method based on a semi-supervised learning algorithm, which is characterized in that the method adopts the system of claim 1 or 2 to allocate the unmanned aerial vehicle communication frequency band, and specifically comprises the following steps:
before the unmanned aerial vehicle takes off, a frequency spectrum sensing module of a ground control center carries out frequency spectrum scanning on a frequency band pre-occupied by the unmanned aerial vehicle, a data processing module of the ground control center judges whether the frequency band pre-occupied by the unmanned aerial vehicle is in an interference state, if interference exists, the frequency band is switched, and frequency spectrum scanning is continued until an interference-free idle frequency band is found;
step two, the ground control center adopts the idle frequency band in the step one to be in communication connection with the unmanned aerial vehicle, the ground control center sends remote control information to the unmanned aerial vehicle, and the unmanned aerial vehicle receives the remote control information to take off and execute an air flight task;
in the process of executing the air flight task, the ground control center receives flight state information and image acquisition information of the unmanned aerial vehicle through the wireless communication module, and meanwhile, the spectrum sensing module of the unmanned aerial vehicle performs spectrum scanning in real time to obtain spectrum data of an occupied frequency band and a pre-occupied frequency band of the unmanned aerial vehicle, integrates and stores the spectrum data, and then sends the spectrum data to the central processing unit;
after receiving the frequency spectrum data, the central processing unit carries out quantization processing on the obtained frequency spectrum data to obtain frequency spectrum data sets with marks and without marks, and the data analysis module trains the frequency spectrum data sets with the marks and without marks by using a semi-supervised learning algorithm to obtain the trained frequency spectrum data;
inputting the frequency spectrum data trained in the step four into a signal classification and prediction module, training and classifying the frequency spectrum data by using a naive Bayes classifier through the signal classification and prediction module, training out a naive Bayes classifier model, predicting the idle probability of each frequency band through the naive Bayes classifier model, and sending the idle probability prediction result of each frequency band to a frequency band switching module to complete idle frequency band prediction;
after the idle frequency band is predicted, when the communication frequency band used by the unmanned aerial vehicle currently has interference, the unmanned aerial vehicle sends a frequency band switching request to a central processing unit in the unmanned aerial vehicle, and the central processing unit controls a frequency band switching module to switch the unmanned aerial vehicle from the current interfered frequency band to the idle frequency band according to a preset distribution principle; the ground control center communicates with the unmanned aerial vehicle according to the switched frequency band, and receives flight state information and image acquisition information transmitted to the unmanned aerial vehicle;
and step seven, repeating the step two to the step six, and ensuring that the normal working communication of the unmanned aerial vehicle is not interfered until the unmanned aerial vehicle finishes the flight task.
4. The unmanned aerial vehicle communication frequency band allocation method based on the semi-supervised learning algorithm as recited in claim 3, wherein: the preset distribution principle is as follows: and according to the predicted idle probability of each frequency band, allocating the frequency band with the highest idle probability to the unmanned aerial vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110389059.4A CN113301654B (en) | 2021-04-12 | 2021-04-12 | Unmanned aerial vehicle communication frequency band distribution system and method based on semi-supervised learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110389059.4A CN113301654B (en) | 2021-04-12 | 2021-04-12 | Unmanned aerial vehicle communication frequency band distribution system and method based on semi-supervised learning algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113301654A CN113301654A (en) | 2021-08-24 |
| CN113301654B true CN113301654B (en) | 2022-06-14 |
Family
ID=77319603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110389059.4A Active CN113301654B (en) | 2021-04-12 | 2021-04-12 | Unmanned aerial vehicle communication frequency band distribution system and method based on semi-supervised learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113301654B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11929885B2 (en) | 2022-02-10 | 2024-03-12 | Honeywell International Inc. | Situation aware edge analytics framework for communication systems |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9622133B1 (en) * | 2015-10-23 | 2017-04-11 | The Florida International University Board Of Trustees | Interference and mobility management in UAV-assisted wireless networks |
| US9930567B1 (en) * | 2011-02-03 | 2018-03-27 | Horizon Hobby, LLC | Three dimensional spread spectrum remote control system |
| CN108833049A (en) * | 2018-06-12 | 2018-11-16 | 北京邮电大学 | Deception formula anti-interference method and device based on cognition in a kind of unmanned plane network |
| CN109041195A (en) * | 2018-07-19 | 2018-12-18 | 浙江工业大学 | A kind of energy-collecting type wireless relay network througput maximization approach based on semi-supervised learning |
| WO2019149162A1 (en) * | 2018-02-02 | 2019-08-08 | 索尼公司 | Device and method for wireless communication system and computer readable storage medium |
| CN111866890A (en) * | 2020-06-08 | 2020-10-30 | 中国人民解放军空军工程大学 | Efficient spectrum sharing method and system in cognitive unmanned aerial vehicle network |
| CN111884740A (en) * | 2020-06-08 | 2020-11-03 | 江苏方天电力技术有限公司 | Unmanned aerial vehicle channel optimal allocation method and system based on frequency spectrum cognition |
| CN112543499A (en) * | 2020-11-20 | 2021-03-23 | 北京科技大学 | Resource allocation method and device for unmanned aerial vehicle-assisted wireless energy-carrying communication Internet of things |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9084276B2 (en) * | 2009-09-11 | 2015-07-14 | Aerovironment, Inc. | Dynamic transmission control for a wireless network |
| US10575312B2 (en) * | 2016-12-01 | 2020-02-25 | Electronics And Telecommunications Research Institute | Method of assigning channel for UAS control and non-payload communication (CNPC) system |
| KR102535213B1 (en) * | 2018-01-23 | 2023-05-22 | 한국전자통신연구원 | METHOD AND APPARATUS FOR CHANNEL SELECTION FOR CONTROL AND NON-PAYLOAD COMMUNICATION FOR Unmanned Aircraft Systems BASED ON CHANNEL INTERFERENCE ANALYSIS |
-
2021
- 2021-04-12 CN CN202110389059.4A patent/CN113301654B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9930567B1 (en) * | 2011-02-03 | 2018-03-27 | Horizon Hobby, LLC | Three dimensional spread spectrum remote control system |
| US9622133B1 (en) * | 2015-10-23 | 2017-04-11 | The Florida International University Board Of Trustees | Interference and mobility management in UAV-assisted wireless networks |
| WO2019149162A1 (en) * | 2018-02-02 | 2019-08-08 | 索尼公司 | Device and method for wireless communication system and computer readable storage medium |
| CN108833049A (en) * | 2018-06-12 | 2018-11-16 | 北京邮电大学 | Deception formula anti-interference method and device based on cognition in a kind of unmanned plane network |
| CN109041195A (en) * | 2018-07-19 | 2018-12-18 | 浙江工业大学 | A kind of energy-collecting type wireless relay network througput maximization approach based on semi-supervised learning |
| CN111866890A (en) * | 2020-06-08 | 2020-10-30 | 中国人民解放军空军工程大学 | Efficient spectrum sharing method and system in cognitive unmanned aerial vehicle network |
| CN111884740A (en) * | 2020-06-08 | 2020-11-03 | 江苏方天电力技术有限公司 | Unmanned aerial vehicle channel optimal allocation method and system based on frequency spectrum cognition |
| CN112543499A (en) * | 2020-11-20 | 2021-03-23 | 北京科技大学 | Resource allocation method and device for unmanned aerial vehicle-assisted wireless energy-carrying communication Internet of things |
Non-Patent Citations (3)
| Title |
|---|
| An efficient wideband spectrum sensing algorithm for unmanned aerial vehicle communication networks;Xu W etal;《IEEE Internet of Things Journal》;20181231;全文 * |
| 空天地海一体化网络中的高能效频谱共享方法;李俊峰等;《移动通信》;20201104;全文 * |
| 面向无人机通信的频谱资源利用与优化;陈勇志;《中国优秀硕士学位论文全文数据库》;20190617;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113301654A (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Baroffio et al. | A visual sensor network for parking lot occupancy detection in smart cities | |
| US20200410787A1 (en) | Systems and methods for processing vehicle sensor data | |
| Yang et al. | Machine-learning-enabled cooperative perception for connected autonomous vehicles: Challenges and opportunities | |
| CN111884740B (en) | Unmanned aerial vehicle channel optimal allocation method and system based on frequency spectrum cognition | |
| CN113301654B (en) | Unmanned aerial vehicle communication frequency band distribution system and method based on semi-supervised learning algorithm | |
| CN111884939B (en) | Data transmission method, device, mobile carrier and storage medium | |
| Dong et al. | WAVE: Edge-device cooperated real-time object detection for open-air applications | |
| CN113179541B (en) | Unmanned aerial vehicle communication method and device based on microwave image transmission and cellular network | |
| KR20220151674A (en) | Power Savings and Related Improvements in Communication Networks | |
| JPWO2019187041A1 (en) | System, its server computer, control method and computer program | |
| JP7276432B2 (en) | Communication system and terminal | |
| Collotta et al. | Wireless sensor networks to improve road monitoring | |
| CN112739984A (en) | Mobile phone multimodal position sensing | |
| Tulay et al. | Road state inference via channel state information | |
| Xu et al. | Autonomous rate control for mobile Internet of Things: A deep reinforcement learning approach | |
| Rahman et al. | Dynamic error-bounded lossy compression to reduce the bandwidth requirement for real-time vision-based pedestrian safety applications | |
| CN113115072A (en) | Video target detection tracking scheduling method and system based on end cloud cooperation | |
| Mukherjee et al. | A supervised‐learning‐based spatial performance prediction framework for heterogeneous communication networks | |
| Pyo et al. | Real-time wireless channel monitoring cooperated with roadside sensors for autonomous mobility assistance system (AMAS) | |
| US20180113892A1 (en) | Data processor and data processing method | |
| Schippers et al. | Data-Driven Digital Mobile Network Twin Enabling Mission-Critical Vehicular Applications | |
| US20220303547A1 (en) | Image information reception apparatus, image information transmission apparatus, and program and method used therein | |
| CN117713967A (en) | Intelligent frequency spectrum sensing anti-interference communication system for ad hoc network unmanned aerial vehicle | |
| CN114640853B (en) | Unmanned aerial vehicle image processing system that cruises | |
| CN115935278B (en) | Environment recognition method, electronic device, and computer-readable storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |