Communication frequency determination method and apparatus, electronic device, and storage medium

Abstract

Embodiments of the present application disclose a communication frequency determination method and device, electronic equipment and a storage medium. The method comprises: obtaining communication data corresponding to at least one communication condition parameter; inputting the communication data into a communication frequency prediction model, and calculating a posterior probability corresponding to each candidate communication frequency value by the communication frequency prediction model according to the communication data, the posterior probability being used to represent a probability that the candidate communication frequency value meets a communication quality requirement; and determining a target communication frequency value from the candidate communication frequency values according to the posterior probability corresponding to each candidate communication frequency value. The communication frequency determination method, device, electronic equipment and storage medium can improve the flexibility and accuracy of selecting a communication frequency, and improve the practicability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, in particular to a communication frequency determination method and device, electronic equipment and storage medium. BACKGROUND

[0002] With the development of communication technology, wireless communication has become an irreplaceable mainstream communication method, whether in ordinary people's life or in military and other aspects has a wide range of applications. Wireless communication has problems such as channel stability and low reliability in signal transmission, and its communication quality depends heavily on the selected communication frequency.

[0003] In order to ensure the communication quality in the process of wireless communication, adaptive technology can be used to adaptively select the communication frequency and establish the communication link, but the current adaptive technology for selecting the communication frequency has the problems of poor practicability and low flexibility. SUMMARY

[0004] The embodiments of the present application disclose a communication frequency determination method and device, electronic equipment and storage medium, which can improve the flexibility and accuracy of selecting the communication frequency, and improve the practicability.

[0005] The embodiments of the present application disclose a communication frequency determination method, comprising:

[0006] obtaining communication data corresponding to at least one communication condition parameter;

[0007] inputting the communication data into a communication frequency prediction model, and calculating a posterior probability corresponding to each candidate communication frequency value through the communication frequency prediction model according to the communication data, wherein the posterior probability is used to represent the probability that the candidate communication frequency value meets the communication quality requirement;

[0008] determining a target communication frequency value from the candidate communication frequency values according to the posterior probability corresponding to each candidate communication frequency value.

[0009] The embodiments of the present application disclose a communication frequency determination device, comprising:

[0010] a data acquisition module configured to obtain communication data corresponding to at least one communication condition parameter;

[0011] a probability calculation module configured to input the communication data into a communication frequency prediction model, and calculate a posterior probability corresponding to each candidate communication frequency value through the communication frequency prediction model according to the communication data, wherein the posterior probability is used to represent the probability that the candidate communication frequency value meets the communication quality requirement;

[0012] A frequency determination module is configured to determine a target communication frequency value from the plurality of candidate communication frequency values according to a posterior probability corresponding to each candidate communication frequency value.

[0013] The electronic device disclosed by the embodiment of the present application comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to realize the method described above.

[0014] The computer readable storage medium disclosed by the embodiment of the present application stores a computer program, and the computer program is executed by a processor to realize the method described above.

[0015] The communication frequency determination method and device, electronic device and storage medium disclosed by the embodiment of the present application can obtain communication data corresponding to at least one communication condition parameter, input the communication data into a communication frequency prediction model, calculate a posterior probability corresponding to each candidate communication frequency value according to the communication data through the communication frequency prediction model, and use the posterior probability to represent the probability that the candidate communication frequency value meets the communication quality requirement, and determine a target communication frequency value from the plurality of candidate communication frequency values according to the posterior probability corresponding to each candidate communication frequency value. In the embodiment of the present application, the communication frequency prediction model can calculate the posterior probability of the availability of each candidate communication frequency value according to the communication data, so as to quickly identify the target frequency value from the plurality of candidate communication frequency values, solve the problem that the frequency selection range is limited by the frequency set in the adaptive technology, improve the flexibility and accuracy of selecting the communication frequency, and improve the practicability. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The flowchart of the communication frequency determination method in one embodiment;

[0018] Figure 2 The flowchart of training the communication frequency prediction model in one embodiment;

[0019] Figure 3 The flowchart of determining the joint probability density between the communication parameter set and the SNR in the communication frequency prediction model in one embodiment;

[0020] Figure 4 The flowchart of the communication frequency determination method in another embodiment;

[0021] Figure 5 Fig. 1 is a schematic diagram of a method for determining a communication frequency according to an embodiment;

[0022] Figure 6 Fig. 2 is a schematic diagram of a device for determining a communication frequency according to an embodiment;

[0023] Figure 7 Fig. 3 is a structural block diagram of an electronic device according to an embodiment. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0025] It should be noted that the terms "comprising" and "having" and any variations thereof in the embodiments of the present application and the drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally further include steps or units not listed or can optionally further include other steps or units inherent to the process, method, product or device.

[0026] Short wave communication uses electromagnetic waves of 2-30 MHz (megahertz) for signal transmission, and is the earliest wireless communication method that appears and is widely used worldwide. Due to the fact that it mainly travels through the reflection of the ionosphere on the earth's surface, it is still an important means of long-distance wireless communication, and has the advantages of no relay, strong anti-destruction, flexible maneuvering, simple erection, etc., and is widely used in various fields. However, due to the complex characteristics of the earth's ionosphere, which changes with factors such as season, time of day, and location, short wave transmission has significant frequency selectivity and fading characteristics, and there are adverse factors such as multipath interference and Doppler shift. Therefore, short wave communication has the problems of unstable channel and low reliability, and the communication effect is seriously dependent on the selected communication frequency.

[0027] In the related technical solutions, adaptive technology is mainly used to select a usable communication frequency for a short-wave communication link. The adaptive technology mainly refers to a technology capable of continuously measuring communication signal and system changes, automatically changing system structure and parameters, and enabling the system to adapt to communication condition changes and resist interference. For short-wave communication, selecting and replacing a communication frequency is the best way to improve the quality of short-wave communication. Therefore, the short-wave channel quality can be detected based on the adaptive technology, and a communication frequency can be automatically selected and a communication link can be established based on the detection result, which can be referred to as automatic link establishment (ALE).

[0028] The adaptive technology mainly includes a first-generation adaptive technology (1G-ALE), a second-generation adaptive technology (2G-ALE), and a third-generation adaptive technology (3G-ALE).

[0029] The first-generation adaptive technology is realized by an independent frequency detection device and a frequency management system, and is characterized by separation of communication and detection.

[0030] The second-generation adaptive technology is characterized by directly using RTCE technology in a communication system to realize a short-wave adaptive radio station integrating short-wave channel detection, evaluation, and communication functions. The basic principle is to perform bidirectional link quality analysis (LQA) on a set of pre-set frequencies (usually ten to twenty), sort the communication frequencies according to the channel quality, form an LQA matrix, and automatically establish a communication link based on the formed LQA matrix.

[0031] Compared with the second-generation adaptive technology, the third-generation adaptive technology has undergone a large number of improvements, including the use of a more efficient modulation and demodulation mode, the separation of a calling channel and a service channel to improve network access efficiency and network capacity, and the expansion of data transmission capacity.

[0032] In the current way of selecting a short-wave communication frequency using adaptive technology, the communication frequency is selected from a pre-set ALE frequency set. The pre-set ALE frequency set is usually limited and cannot meet the needs of communication tasks in different scenarios, and it is possible that a communication frequency with better communication quality cannot be selected. In addition, the frequency selection and link establishment time in the adaptive technology increases rapidly with the increase in the number of frequencies included in the ALE frequency set. When the number of frequencies included in the ALE frequency set is too large, the frequency selection and link establishment time is too long, and the practicability is poor. Therefore, the current way of selecting a communication frequency using adaptive technology has the problems of poor practicability, low flexibility, and low accuracy.

[0033] In addition, in addition to the adaptive technology selecting the communication frequency of the wireless short wave, there is also an access network mode frequency selection technology, the main principle of which is to improve the service quality of the communication object on the other side by increasing the guarantee resources on the communication side. The access network mode frequency selection technology also has the problems of limited frequency selection range and poor practicability due to long frequency selection link establishment time.

[0034] The embodiment of the present application provides a communication frequency determination method and device, electronic equipment and storage medium, which can solve the problem of limited frequency selection range in the adaptive technology, improve the flexibility and accuracy of selecting the communication frequency, and improve the practicability.

[0035] As shown in FIG. 1, Figure 1 In one embodiment, a communication frequency determination method is provided, which can be applied to an electronic equipment, which can include but is not limited to a mobile phone, a tablet computer, a smart wearable device (such as smart glasses, a smart watch, etc.), a notebook computer, a PC (Personal Computer), and the like, and the electronic equipment can also be a server or a server cluster, etc., and the embodiment of the present application does not limit the electronic equipment. The method can include the following steps:

[0036] In step 110, communication data corresponding to at least one communication condition parameter is obtained.

[0037] The communication condition parameter can refer to a parameter associated with the communication demand in the wireless communication process, and the communication data corresponding to the at least one communication condition parameter can refer to the expected data corresponding to each communication condition parameter when the communication is expected. The communication condition parameter can include but is not limited to at least one of the following: communication time, channel bandwidth, service type, calling geographic location, calling station power level, calling antenna type, called geographic location, called station power level, and called antenna type.

[0038] Optionally, the communication time can refer to the duration of the communication, considering that the ionosphere relied on by the short wave communication transmission has a cross-time zone feature, the communication time can be represented in Universal Time Coordinated (UTC) so that the expected communication data is more consistent with the real communication data in the communication process. The channel bandwidth can include, but is not limited to, 3 kHz (kilohertz), 6 kHz, 12 kHz, 24 kHz, etc. The service type can include, but is not limited to, datagram, vocoded speech, etc. The calling station geographical position and the called station geographical position can be represented by longitude and latitude, etc. The calling station power level and the called station power level can include, but are not limited to, 20 W (watt), 125 W, 400 W, 1000 W, 5 kW, 20 kW (kilowatt), etc. The calling station antenna type and the called station antenna type can include, but are not limited to, 1.5-meter whip antenna, 4-meter whip antenna, 10-meter whip antenna, dipole antenna, fishbone antenna, log-periodic antenna, three-wire broadband antenna, vertical cage antenna, etc.

[0039] The communication data described above can include specific values of the expected communication time corresponding to each communication condition parameter, and the communication data can be used to simulate the communication situation in the real service communication process. For example, the communication data can include, but is not limited to, communication time 3 minutes, channel bandwidth 12 kHz, service type vocoded speech, calling station geographical position (x1, y1), calling station power level 400 W, calling station antenna type dipole antenna, called station geographical position (x2, y2), called station power level 125 W, and called station antenna type log-periodic antenna.

[0040] As an implementation manner, the communication data can be represented in the form of a data set, and each specific value in the communication data can be sequentially sorted according to the arrangement order of the preset communication condition parameters. The electronic device can determine the communication condition parameter corresponding to each specific value according to the arrangement position of the specific value in the data set.

[0041] In step 120, the communication data is input into the communication frequency prediction model, and the posterior probability corresponding to each candidate communication frequency value is calculated by the communication frequency prediction model according to the communication data. The posterior probability is used to represent the probability that the candidate communication frequency value meets the communication quality requirement.

[0042] The electronic device can input the communication data into a pre-trained communication frequency prediction model. In embodiments of the present application, the communication frequency prediction model can be a probabilistic model, which has the ability to analyze the probability of each candidate communication frequency value under various communication conditions. The communication frequency prediction model can iterate through each candidate communication frequency value, and calculate the posterior probability corresponding to each candidate communication frequency value according to the communication data. The candidate communication frequency value can be a communication frequency value that can cover multiple frequency bands available for communication. The number of candidate communication frequency values is not limited in embodiments of the present application, for example, it can be several hundred or several thousand. Compared with the frequency set in the adaptive technology, the number of optional communication frequency values in the present application is larger, and the speed of selecting the communication frequency value is faster and more efficient because it does not need to scan and detect each candidate communication frequency value.

[0043] In some embodiments, the communication frequency prediction model can be trained by a large number of sample communication data. Each sample communication data can include a communication frequency value, sample data corresponding to each communication condition parameter, and a communication quality indicator corresponding to each sample communication data. Alternatively, the candidate communication frequency value can be each communication frequency value contained in the large number of sample communication data, but is not limited thereto. The communication quality indicator can be used to represent the communication quality corresponding to each sample communication data. The communication quality indicator can include at least one of a signal-to-noise ratio (SNR) value, a communication delay, a communication quality level, a communication quality score, etc. The communication quality level and the communication quality score can be obtained by analyzing and comprehensively evaluating the sample communication data.

[0044] Alternatively, the posterior probability calculated by the communication frequency prediction model according to the input communication data can include the probability of each candidate communication frequency value meeting the specified communication quality condition under the communication demand corresponding to the communication data. The communication quality condition can refer to the condition that meets the communication quality requirement. For example, taking the SNR as the communication quality indicator, the communication quality condition can be that the SNR value is greater than a preset SNR threshold (such as 18 dB, 15 dB, 20 dB, etc.). For example, taking the communication quality level as the communication quality indicator, the communication quality condition can be that the communication quality level is greater than a preset level, etc.

[0045] For each candidate communication frequency value, the probability of meeting the specified communication quality condition can be calculated. For example, there can be 1000 candidate communication frequency values A1-A100. The communication frequency prediction model can calculate the probability of each candidate communication frequency value A1-A100 meeting the condition that the SNR value is greater than the preset SNR threshold according to the communication data.

[0046] Optionally, the posterior probability calculated by the communication frequency prediction model according to the input communication data can include the probability of each candidate communication frequency value under the communication demand corresponding to the communication data at each communication quality indicator. For example, taking the SNR as the communication quality indicator, 1000 candidate communication frequency values A1-A1000 and 30 SNR values Y1-Y30 are included, and the communication frequency prediction model can calculate the probability of each candidate communication frequency value A1-A1000 corresponding to the 30 SNR values Y1-Y30 according to the communication data, wherein the probability of the candidate communication frequency value A1 can include P(A1-1)-P(A1-30), P(A1-1) represents the probability of the candidate communication frequency value A1 corresponding to the SNR value Y1, and P(A1-30) represents the probability of the candidate communication frequency value A1 corresponding to the SNR value Y30.

[0047] In some embodiments, the communication frequency prediction model can include but is not limited to a Naive Bayes classifier model, which is a series of simple probability classifiers based on the assumption that the features are strongly independent and the use of Bayes' theorem. The Naive Bayes classifier model can learn the rules between the communication frequency value, each communication condition parameter and the communication quality indicator according to a large amount of sample communication data, so that when the communication data is input into the Naive Bayes classifier model, the communication quality indicator to which each candidate communication frequency value belongs can be classified according to the communication data. Further, the communication quality indicator to which the candidate communication frequency value belongs can be the communication quality indicator with the maximum probability.

[0048] Step 130, determining the target communication frequency value from each candidate communication frequency value according to the posterior probability corresponding to each candidate communication frequency value.

[0049] The target communication frequency value can be used to establish a wireless communication link in service communication. The target communication frequency value can be a communication frequency value with better communication quality under the condition of meeting the communication demand corresponding to the communication data. The determined target communication frequency value can be one or more (referring to two or more). The frequency value determination rule can be set in advance, and the target communication frequency value can be determined from each candidate communication frequency value according to the frequency value determination rule.

[0050] In some embodiments, the frequency value determination rule can include but is not limited to any one of the following:

[0051] I. The candidate communication frequency values arranged in the first N positions are determined as the target communication frequency values in the order of the posterior probability corresponding to each candidate communication frequency value from large to small, wherein N is a positive integer.

[0052] The posterior probability corresponding to each candidate communication frequency value can be the probability of each candidate communication frequency value under the communication demand corresponding to the communication data under the specified communication quality condition. Therefore, the greater the posterior probability corresponding to the candidate communication frequency value, the more optimal the communication quality corresponding to the candidate communication frequency value. The posterior probabilities corresponding to the candidate communication frequency values can be sorted in descending order of posterior probability, and the candidate communication frequency values ranked in the top N are selected as the target communication frequency values. In this way, the communication quality of the determined target communication frequency values can be ensured to be better, and the accuracy of the communication frequency selection is improved.

[0053] II. The candidate communication frequency value with a posterior probability greater than a probability threshold value is determined as the target communication frequency value.

[0054] The posterior probability corresponding to each candidate communication frequency value can be the probability of each candidate communication frequency value under the communication demand corresponding to the communication data under the specified communication quality condition. The probability threshold value can be set according to actual needs, such as 70%, 77%, etc., but is not limited thereto. The candidate communication frequency value with a posterior probability greater than the probability threshold value can be directly selected as the target communication frequency value.

[0055] Alternatively, if the posterior probabilities of the candidate communication frequency values are all less than the probability threshold value, the candidate communication frequency values ranked in the top N can be selected as the target communication frequency values according to the method in rule I. In this way, the communication quality of the determined target communication frequency values can be ensured to be better, and the accuracy of the communication frequency selection is improved.

[0056] III. The communication quality indicators to which the candidate communication frequency values belong are determined according to the posterior probabilities corresponding to the candidate communication frequency values, and the candidate communication frequency values ranked in the top N are determined as the target communication frequency values in order from better to worse quality represented by the communication quality indicators.

[0057] The posterior probability corresponding to each candidate communication frequency value can include the probability of each candidate communication frequency value under the communication demand corresponding to the communication data under each communication quality indicator.

[0058] In some embodiments, the communication quality indicators include SNR, and the posterior probability corresponding to each candidate communication frequency value can include the probability of each candidate communication frequency value corresponding to a plurality of signal-to-noise ratio (SNR) values. The SNR values to which the candidate communication frequency values belong can be determined according to the posterior probabilities corresponding to the candidate communication frequency values (it can be understood that the candidate communication frequency values are classified, and each SNR value is a category), and the candidate communication frequency values ranked in the top N are determined as the target communication frequency values in order from large to small according to the SNR values to which they belong.

[0059] As a specific implementation, the electronic device can obtain a maximum probability corresponding to each candidate communication frequency value and a target SNR value corresponding to the maximum probability according to probabilities of respective candidate communication frequency values and a plurality of SNR (Signal to Noise Ratio) values respectively. The target SNR value corresponding to each candidate communication frequency value is an SNR value to which the candidate communication frequency value belongs. For example, the probabilities corresponding to the candidate communication frequency value A1 can include P(A1-1)~P(A1-30), P(A1-1) represents a probability of the candidate communication frequency value A1 corresponding to the SNR value Y1, P(A1-30) represents a probability of the candidate communication frequency value A1 corresponding to the SNR value Y30, and the like. The maximum probability in P(A1-1)~P(A1-30) can be obtained, and the SNR value corresponding to the maximum probability is taken as the target SNR value corresponding to the candidate communication frequency value A1.

[0060] After the target SNR value corresponding to each candidate communication frequency value is determined, the candidate communication frequency values arranged in the first N positions can be determined as target communication frequency values in a descending order of the target SNR values corresponding to the candidate communication frequency values, where N is a positive integer. In this way, the SNR values to which the candidate communication frequency values belong are classified first, and then the target communication frequency values are determined, so that the communication quality of the determined target communication frequency values is better, and the accuracy of the selection of the communication frequency is improved.

[0061] It should be noted that the target communication frequency values can also be determined by other manners, and are not limited to the above-mentioned manners.

[0062] In the embodiments of the present application, the communication frequency prediction model can calculate the posterior probability of the availability of each candidate communication frequency value according to the communication data, so as to quickly identify the target frequency value from the plurality of candidate communication frequency values, solve the problem that the frequency selection range in the adaptive technology is limited by the frequency set, improve the flexibility and accuracy of the selection of the communication frequency, and improve the practicability.

[0063] As shown in FIG. 1, Figure 2 As shown in FIG. 1,

[0064] In step 202, a plurality of sample communication data are collected according to a preset communication parameter set. The communication parameter set includes a plurality of communication characteristic parameters, and the plurality of communication characteristic parameters include a communication frequency and at least one communication condition parameter. Each sample communication data includes sample data corresponding to each communication characteristic parameter, and includes an SNR value corresponding to the sample communication data.

[0065] The communication parameter set can be set in advance according to actual needs, and the communication parameter set can include a plurality of communication characteristic parameters, which can be parameters capable of affecting the communication quality. In the embodiments of the present application, the communication characteristic parameters can include a communication frequency and at least one communication condition parameter, which can be the same as the communication condition parameters corresponding to the communication data input into the communication frequency prediction model in the application process of the communication frequency prediction model. Alternatively, the communication condition parameters corresponding to the communication data input into the communication frequency prediction model can also be a subset of the communication condition parameters included in the communication parameter set, which is not limited herein.

[0066] In some embodiments, the plurality of pieces of sample communication data can be collected by at least one of the following ways:

[0067] The first way is to generate the plurality of pieces of sample communication data conforming to the preset communication parameter set by a short-wave propagation prediction model.

[0068] The short-wave propagation prediction model can include, but is not limited to, ICEPAC (Ionospheric Communications Enhanced Profile Analysis & Circuit), VOACAP, REC533 and the like. The short-wave propagation prediction model can be obtained by training a large amount of ionospheric observation data (including electron density, propagation loss and the like of ionospheric E layer, F1 layer, F2 layer and the like), and the short-wave propagation prediction model can establish a link loss and noise model based on the ionospheric observation data.

[0069] The electronic device can randomly generate sample data corresponding to each communication characteristic parameter included in the preset communication parameter set, and input the sample data into the short-wave propagation prediction model. The short-wave propagation prediction model can analyze the sample data and calculate a predicted SNR value. The predicted SNR value and the sample data can constitute a piece of sample communication data, and the predicted SNR value is the SNR value corresponding to the sample communication data. The efficiency of generating the sample communication data by the short-wave propagation prediction model is higher, and the data amount of the sample communication data can be ensured, so as to avoid the situation that the communication frequency prediction model trained due to too little sample communication data is inaccurate and the like.

[0070] The second way is to collect the plurality of pieces of sample communication data from historical communication records according to the preset communication parameter set.

[0071] The electronic device can acquire a historical communication record, which can include historical communication data of each time a service communication is made and an SNR value corresponding to the historical communication data. According to a preset set of communication parameter, sample data corresponding to each communication characteristic parameter can be acquired from the historical communication data included in the historical communication record, and the SNR value corresponding to the historical communication data can be taken as the SNR value corresponding to the sample communication data. By collecting the sample communication data using the historical communication record, the authenticity of the sample communication data can be ensured, and the accuracy of the trained communication frequency prediction model is further improved.

[0072] In step 204, the joint probability density between the set of communication parameters and the SNR is calculated by the communication frequency prediction model to be trained according to the multiple pieces of sample communication data.

[0073] The input of the communication frequency prediction model can be communication data corresponding to the set of communication parameters, and the output can be an SNR value. Therefore, the communication frequency prediction model can learn the joint probability density between the set of communication parameters and the SNR (i.e., the joint probability density between the output and the output) according to the multiple pieces of sample communication data corresponding to the set of communication parameters, which can be used to describe the probability distribution between the set of communication parameters and the SNR.

[0074] Suppose the set of communication parameters is denoted as variable X = {X1, X2, X3, …, Xn}, where Xi (i = 1, 2, 3, …, n) represents each communication characteristic parameter, and the SNR is denoted as variable Y = {y1, y2, y3, y4, …, ym}, where yk (k = 1, 2, 3, …, m) represents each SNR value. The communication frequency prediction model can calculate the joint probability density between X and Y according to the multiple pieces of sample communication data.

[0075] As shown in FIG. 3, in one embodiment, step 204 can include steps 302-306. Figure 3

[0076] In step 302, the multiple pieces of sample communication data are analyzed by the communication frequency model to be trained to obtain a prior probability corresponding to the SNR.

[0077] The prior probability corresponding to the SNR can be used to represent the probability of occurrence of each SNR value in the multiple pieces of sample communication data. Suppose the SNR is denoted as variable Y = {y1, y2, y3, y4, …, ym}, then the prior probability corresponding to the SNR can be P(Y = yk), where k = 1, 2, 3, …, m.

[0078] In step 304, the conditional probability corresponding to the set of communication parameters is determined by the communication frequency model to be trained according to the multiple pieces of sample communication data and the prior probability.

[0079] ​The conditional probability corresponding to the communication parameter set can be used to represent the occurrence probability of each sample data corresponding to each communication characteristic parameter under each SNR value in the multiple sample communication data. Assuming that the communication parameter set is denoted as variable X={X1, X2, X3, …, Xn}, and the SNR is denoted as variable Y={y1, y2, y3, y4, …, ym}, the conditional probability corresponding to the communication parameter set can be P(X=xm|Y=yk)=P(X1=x1, X1=x2, X1=x3, …, X2=x1, X2=x2, X2=x3, …, Xn=x1, Xn=x2, Xn=x3 …|Y=yk), where k=1, 2, 3, …, m.

[0080] wherein xm represents various values of the communication characteristic parameter appearing in the sample communication data, for example, X2 represents the channel width, and the channel width in the sample communication data includes 3 kHz, 6 kHz, 12 kHz, and 24 kHz, so X2=x1=3 kHz, X2=x2=6 kHz, X2=x3=12 kHz, and X2=x4=24 kHz.

[0081] Alternatively, the conditional probability corresponding to the communication parameter set can be calculated by using formula (1):

[0082]

[0083] wherein P(X=xm∩Y=yk) represents the intersection between the sample data corresponding to each communication characteristic parameter and each SNR value.

[0084] In step 306, the joint probability density between the communication parameter set and the SNR is determined according to the prior probability and the conditional probability.

[0085] Alternatively, the joint probability density between the communication parameter set and the SNR can be calculated by using formula (2):

[0086] P1=∑ k [P(Y=yk)P(X=xm|Y=yk)] formula (2);

[0087] wherein P1 represents the joint probability density between the communication parameter set and the SNR.

[0088] After the joint probability density between the communication parameter set and the SNR is learned by the communication frequency model, the corresponding posterior probability can be calculated for each candidate communication frequency value according to the input communication data.

[0089] It should be noted that the above-mentioned sample communication data may also include communication quality indicators other than SNR, such as communication quality level and communication quality score. A mapping relationship between SNR value and communication quality level can be established in advance, and the SNR value corresponding to each sample communication data can be converted into the corresponding communication quality level according to the mapping relationship. Then, the sample data and communication quality level are input into the communication frequency model to train the communication frequency model.

[0090] In this embodiment of the application, the communication frequency model can be trained based on a large amount of sample communication data, so that the communication frequency model can accurately obtain the availability probability of each candidate communication frequency value under the communication requirement based on the input communication requirement, thereby enabling the rapid and accurate selection of communication frequency.

[0091] like Figure 4 As shown, in another embodiment, a communication frequency determination method is provided, which can be applied to the above-mentioned electronic device. The method may include the following steps:

[0092] Step 410: Collect multiple sample communication data according to a preset set of communication parameters; wherein, the set of communication parameters includes multiple communication characteristic parameters, including communication frequency and at least one communication condition parameter; each sample communication data includes sample data corresponding to each communication characteristic parameter, and includes SNR value corresponding to the sample communication data.

[0093] Step 420: The communication frequency prediction model to be trained calculates the joint probability density between the communication parameter set and SNR based on multiple sample communication data.

[0094] Step 430: Obtain communication data corresponding to at least one communication condition parameter.

[0095] Step 440: Input the communication data into the communication frequency prediction model, and calculate the posterior probability corresponding to each candidate communication frequency value based on the communication data through the communication frequency prediction model.

[0096] Step 450: Determine the target communication frequency value from the candidate communication frequency values ​​based on the posterior probability corresponding to each candidate communication frequency value.

[0097] The descriptions of steps 410 to 450 can be found in the relevant descriptions in the above embodiments, and will not be repeated here.

[0098] Step 460: Conduct business communication based on the target frequency value, record the corresponding communication logs, and update the communication frequency prediction model based on the communication logs.

[0099] When multiple target frequency values ​​exist, the electronic device can select any one of them to establish a communication link and transmit data for service communication through this link. After completing this service communication, a communication log corresponding to this service communication can be recorded. This communication log may include the communication data and actual SNR value corresponding to this service communication, and the communication data may include data corresponding to various communication characteristic parameters.

[0100] The communication frequency prediction model can be updated based on the communication logs. The communication data and actual SNR value corresponding to the current business communication are input into the communication frequency prediction model to dynamically adjust and update the probability model (such as the prior probability, conditional probability and joint probability density mentioned above) in the communication frequency prediction model.

[0101] Figure 5 This is a schematic diagram of a communication frequency determination method in one embodiment. Figure 5 As shown, sample communication data can be input into the communication frequency prediction model to be trained. The acquired communication data corresponding to at least one communication condition parameter can be input into the trained communication frequency prediction model. Based on this communication data, the trained communication frequency prediction model can calculate the posterior probability corresponding to each candidate communication frequency value, thereby selecting the target communication frequency value. Service communication can be performed according to the target communication frequency value, generating corresponding communication logs. These logs are then used to dynamically update the trained communication frequency prediction model, forming a feedback loop that continuously improves the accuracy of the communication frequency prediction model.

[0102] In this embodiment, the communication frequency prediction model can be dynamically updated continuously using communication data and actual SNR values ​​from actual business communications. Incremental learning provides data support for the communication frequency prediction model to determine the target communication frequency value next time, thereby improving the accuracy of the communication frequency prediction model.

[0103] like Figure 6 As shown, in one embodiment, a communication frequency determination device 600 is provided, which can be applied to the above-mentioned electronic device. The communication frequency determination device 600 may include a data acquisition module 610, a probability calculation module 620, and a frequency determination module 630.

[0104] The data acquisition module 610 is used to acquire communication data corresponding to at least one communication condition parameter.

[0105] In one embodiment, at least one communication condition parameter includes at least one of the following: communication time, channel bandwidth, service type, calling geographical location, calling radio power level, calling antenna type, called geographical location, called radio power level, and called antenna type.

[0106] The probability calculation module 620 is configured to input the communication data into the communication frequency prediction model, and calculate, by the communication frequency prediction model, a posterior probability corresponding to each candidate communication frequency value according to the communication data, where the posterior probability is used to represent a probability that the candidate communication frequency value meets the communication quality requirement.

[0107] The frequency determination module 630 is configured to determine the target communication frequency value from the candidate communication frequency values according to the posterior probabilities corresponding to the candidate communication frequency values.

[0108] In an embodiment, the frequency determination module 630 is further configured to determine the candidate communication frequency values arranged in the first N positions as the target communication frequency values in a descending order of the posterior probabilities corresponding to the candidate communication frequency values, where N is a positive integer.

[0109] In an embodiment, the posterior probabilities corresponding to the candidate communication frequency values include probabilities corresponding to the candidate communication frequency values and a plurality of signal-to-noise ratio (SNR) values respectively. The frequency determination module 630 is further configured to obtain a maximum probability corresponding to each candidate communication frequency value and a target SNR value corresponding to the maximum probability according to the probabilities corresponding to the candidate communication frequency values and the plurality of SNR values respectively, and determine the candidate communication frequency values arranged in the first N positions as the target communication frequency values in a descending order of the target SNR values corresponding to the candidate communication frequency values, where N is a positive integer.

[0110] In the embodiments of the present application, the communication frequency prediction model can calculate the posterior probabilities of the availability of the candidate communication frequency values according to the communication data, so as to quickly identify the target frequency value from the plurality of candidate communication frequency values, solve the problem that the frequency selection range in the adaptive technology is limited by the frequency set, improve the flexibility and accuracy of selecting the communication frequency, and improve the practicability.

[0111] In an embodiment, the communication frequency determination apparatus 600 described above further includes a training module in addition to the data acquisition module 610, the probability calculation module 620, and the frequency determination module 630.

[0112] The training module is configured to collect a plurality of pieces of sample communication data according to a preset communication parameter set, where the communication parameter set includes a plurality of communication characteristic parameters, the plurality of communication characteristic parameters include a communication frequency and at least one communication condition parameter, each piece of sample communication data includes sample data corresponding to each communication characteristic parameter, and includes an SNR value corresponding to the sample communication data, and is configured to calculate, by a to-be-trained communication frequency prediction model, a joint probability density between the communication parameter set and the SNR according to the plurality of pieces of sample communication data.

[0113] In an embodiment, the training module is further configured to generate a plurality of pieces of sample communication data conforming to the preset communication parameter set by using the short-wave propagation prediction model; and / or collect the plurality of pieces of sample communication data from historical communication records according to the preset communication parameter set.

[0114] In an embodiment, the training module is further configured to analyze the plurality of pieces of sample communication data by using the communication frequency model to be trained to obtain a prior probability corresponding to the SNR; the prior probability is used to represent the occurrence probability of each SNR value in the plurality of pieces of sample communication data; determine a conditional probability corresponding to the communication parameter set according to the plurality of pieces of sample communication data and the prior probability by using the communication frequency model to be trained, the conditional probability is used to represent the occurrence probability of each piece of sample data corresponding to each communication characteristic parameter under each SNR value in the plurality of pieces of sample communication data; and determine a joint probability density between the communication parameter set and the SNR according to the prior probability and the conditional probability.

[0115] In the embodiments of the present application, the communication frequency model can be trained according to a large amount of sample communication data, so that the communication frequency model has the ability to accurately obtain the available probability of each candidate communication frequency value under the input communication demand, thereby being able to quickly and accurately select the communication frequency.

[0116] In an embodiment, the communication frequency determination apparatus 600 described above further includes an updating module.

[0117] The updating module is configured to perform service communication based on the target frequency value, record a communication log corresponding to the service communication, the communication log including communication data corresponding to the service communication and an actual SNR value, and update the communication frequency prediction model according to the communication log.

[0118] In the embodiments of the present application, the communication frequency prediction model can be dynamically updated by using the communication data and the actual SNR value in the actual service communication, and the communication frequency prediction model is provided with data support for next time determination of the target communication frequency value in an incremental learning manner, thereby improving the accuracy of the communication frequency prediction model.

[0119] Figure 7 FIG. 7 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 7 As shown in FIG. 7, the electronic device 700 can include one or more of the following components: a processor 710, a memory 720 coupled with the processor 710, and a bus 730. The memory 720 can store one or more computer programs, which can be configured to be executed by the processor 710 to implement the method described in the above embodiments.

[0120] The processor 710 can include one or more processing cores. The processor 710 connects various parts within the entire electronic device 700 by running or executing instructions, programs, code sets or instruction sets stored in the memory 720, and calling data stored in the memory 720, to perform various functions and process data of the electronic device 700. Optionally, the processor 710 can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), a programmable logic array (PLA). The processor 710 can integrate a combination of one or several of a central processing unit (CPU), a graphics processor (GPU), and a modem, etc. Among them, the CPU mainly processes operating systems, user interfaces, and application programs, etc.; the GPU is responsible for rendering and drawing display content; and the modem is used for processing wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 710, but be implemented by a separate communication chip.

[0121] The memory 720 can include a random access memory (RAM) and can also include a read-only memory (ROM). The memory 720 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 720 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. The data storage area can also store data created by the electronic device 700 in use, etc.

[0122] It can be understood that the electronic device 700 can include more or less structural elements than those in the above-mentioned structural block diagram, for example, including a power module, a physical key, a WiFi (Wireless Fidelity) module, a speaker, a Bluetooth module, a sensor, etc., which are not limited herein.

[0123] The embodiments of the present application disclose a computer readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the method described in the above-mentioned embodiments.

[0124] The embodiment of the present application discloses a computer program product, which comprises a non-transitory computer-readable storage medium storing a computer program, and the computer program can be executed by a processor to implement the method described in the above embodiments.

[0125] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. The storage medium can be a magnetic disc, an optical disc, a ROM, and the like.

[0126] As used herein, any reference to memory, storage, database, or other medium can include non-volatile and / or volatile storage. Suitable non-volatile storage can include ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), or flash memory. Volatile storage can include random access memory (RAM) used as external cache memory. As an illustration and not a limitation, RAM can be in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus DRAM (RDRAM), and direct Rambus dynamic RAM (DRDRAM).

[0127] It should be understood that every feature, structure, or characteristic described herein is within a preferred embodiment of the present application. It should be understood that "comprising," "including," "has," and any variations thereof are not intended to exclude other features, structures, or characteristics from the present application. Further, unless otherwise specified, "first," "second," or "third" and any variations thereof are not intended to imply that a feature, structure, or characteristic described herein is preferred over another feature, structure, or characteristic. Moreover, unless otherwise specified, the use of relative terms, such as "about," "approximately," "substantially," or the like, typically denote that a feature, structure, or characteristic is within some acceptable limit or range.

[0128] In various embodiments of the present application, it should be understood that the magnitude of the serial number of the above processes does not mean the inevitable sequence of execution, and the execution sequence of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0129] The units described as separate components above can or can not be physically separate, and the components shown as units can or can not be physical units, i.e. they can be located in one place, or they can be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0130] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0131] The above describes in detail a communication frequency determination method, device, electronic equipment and storage medium disclosed by the embodiments of the present application. The principles and implementation modes of the present application are described by applying specific examples. The above embodiment descriptions are only used to help understand the method and its core idea of the present application. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method of determining a communication frequency, characterized by, The method comprises: obtaining communication data corresponding to at least one communication condition parameter; inputting the communication data into a communication frequency prediction model, and calculating, by the communication frequency prediction model, a posterior probability corresponding to each candidate communication frequency value according to the communication data, the posterior probability being used to represent a probability that the candidate communication frequency value meets a communication quality requirement, the communication frequency prediction model being a Naive Bayes classifier model; determining a target communication frequency value from the candidate communication frequency values according to the posterior probabilities corresponding to the candidate communication frequency values; the communication frequency prediction model is obtained by training the following steps: collecting multiple pieces of sample communication data according to a preset communication parameter set, wherein the communication parameter set comprises multiple communication characteristic parameters, the multiple communication characteristic parameters comprising a communication frequency and at least one communication condition parameter, and each piece of sample communication data comprises sample data corresponding to each of the communication characteristic parameters and an SNR value corresponding to the sample communication data; calculating, by a to-be-trained communication frequency prediction model, a joint probability density between the communication parameter set and the SNR according to the multiple pieces of sample communication data; the calculation of the joint probability density between the communication parameter set and the SNR by the to-be-trained communication frequency prediction model according to the multiple pieces of sample communication data comprises: analyzing, by the to-be-trained communication frequency model, the multiple pieces of sample communication data to obtain a prior probability corresponding to the SNR, wherein the prior probability is used to represent a probability of occurrence of each SNR value in the multiple pieces of sample communication data; determining, by the to-be-trained communication frequency model, a conditional probability corresponding to the communication parameter set according to the multiple pieces of sample communication data and the prior probability, the conditional probability being used to represent a probability of occurrence of each sample data corresponding to each of the communication characteristic parameters in the multiple pieces of sample communication data under each of the SNR values; determining the joint probability density between the communication parameter set and the SNR according to the prior probability and the conditional probability.

2. The method of claim 1, wherein, the determination of the target communication frequency value from the candidate communication frequency values according to the posterior probabilities corresponding to the candidate communication frequency values comprises: determining, in a descending order of the posterior probabilities corresponding to the candidate communication frequency values, the candidate communication frequency values in the first N positions as the target communication frequency values, N being a positive integer.

3. The method of claim 1, wherein, the posterior probabilities corresponding to the candidate communication frequency values comprise probabilities corresponding to the candidate communication frequency values and multiple SNR values respectively; the determination of the target communication frequency value from the candidate communication frequency values according to the posterior probabilities corresponding to the candidate communication frequency values comprises: obtaining a maximum probability corresponding to each of the candidate communication frequency values and a target SNR value corresponding to the maximum probability according to the probabilities corresponding to the candidate communication frequency values and multiple SNR values respectively; and The target communication frequency value is determined from the first N candidate communication frequency values in descending order of the target SNR values corresponding to the candidate communication frequency values, where N is a positive integer.

4. The method of claim 1, wherein, The collecting multiple pieces of sample communication data according to the preset communication parameter set comprises: The multiple pieces of sample communication data conforming to the preset communication parameter set are generated by a short-wave propagation prediction model; and / or The multiple pieces of sample communication data are collected from historical communication records according to the preset communication parameter set.

5. The method according to any one of claims 1 to 4, characterized in that, The at least one communication condition parameter comprises at least one of a communication time, a channel bandwidth, a service type, a calling geographic location, a calling station power level, a calling antenna type, a called geographic location, a called station power level, and a called antenna type.

6. The method according to any one of claims 1 to 4, characterized in that, The method comprises: The service communication is performed based on the target communication frequency value, and a communication log corresponding to the service communication is recorded, the communication log comprising communication data and an actual SNR value corresponding to the service communication; The communication frequency prediction model is updated according to the communication log.

7. A communication frequency determining apparatus characterized by comprising: The method comprises: The data acquisition module is configured to acquire communication data corresponding to at least one communication condition parameter; The probability calculation module is configured to input the communication data into a communication frequency prediction model, and calculate a posterior probability corresponding to each candidate communication frequency value by the communication frequency prediction model based on the communication data, the posterior probability being used to represent a probability that the candidate communication frequency value meets a communication quality requirement, the communication frequency prediction model being a naive Bayes classifier model; The frequency determination module is configured to determine a target communication frequency value from the candidate communication frequency values based on the posterior probabilities corresponding to the candidate communication frequency values. The probability calculation module is specifically configured to train the communication frequency prediction model by the following steps: The multiple pieces of sample communication data are collected according to a preset communication parameter set; wherein the communication parameter set comprises multiple communication characteristic parameters, the multiple communication characteristic parameters comprising a communication frequency and at least one communication condition parameter; each piece of sample communication data comprises sample data corresponding to each communication characteristic parameter, and comprises an SNR value corresponding to the sample communication data; A joint probability density between the communication parameter set and the SNR is calculated by a to-be-trained communication frequency prediction model based on the multiple pieces of sample communication data; The probability calculation module is specifically configured to calculate the joint probability density between the communication parameter set and the SNR by the to-be-trained communication frequency prediction model based on the multiple pieces of sample communication data, which comprises: The multiple pieces of sample communication data are analyzed by the to-be-trained communication frequency model to obtain a prior probability corresponding to the SNR; wherein the prior probability is used to represent an occurrence probability of each SNR value in the multiple pieces of sample communication data. determining, by the communication frequency model to be trained, a conditional probability corresponding to the communication parameter set according to the multiple pieces of sample communication data and the prior probability, the conditional probability being used to represent an occurrence probability of each sample data corresponding to each communication characteristic parameter in the multiple pieces of sample communication data under each SNR value; determining a joint probability density between the communication parameter set and SNR according to the prior probability and the conditional probability.

8. An electronic device, comprising: The computer program is executed by the processor to implement the method according to any one of claims 1-6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method according to any one of claims 1-6.

Images