Predicting wireless channel data, electronic device, computer readable medium

Abstract

The present disclosure provides a method for predicting wireless channel data, which includes: obtaining multiple sample groups; each of the sample groups includes multiple sample wireless channel data generated by a wireless channel modeling process; determining the group confidence of each sample group; the group confidence represents the credibility of the sample wireless channel data in the sample group; and determining target wireless channel data according to the multiple sample wireless channel data and their corresponding group confidence. The present disclosure also provides an electronic device and a computer-readable medium.

Description

Technical Field

[0001] The present disclosure relates to the technical field of wireless channel modeling, and in particular to a method for predicting wireless channel data, an electronic device, and a computer-readable medium. Background Art

[0002] Wireless channel modeling can provide an important basis for the planning, design, testing, finalization, adjustment, and optimization of wireless networks (wireless communication networks).

[0003] However, all existing wireless channel modeling technologies have inevitable shortcomings. They cannot adapt to the evolution of wireless networks, cannot perform full life cycle modeling, or have low accuracy, high cost, or affect normal communication services. Summary of the invention

[0004] The present disclosure provides a method for predicting wireless channel data, an electronic device, and a computer-readable medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for predicting wireless channel data, comprising:

[0006] Acquire multiple sample groups; each of the sample groups includes multiple sample wireless channel data generated by a wireless channel modeling process;

[0007] Determining a group confidence of each of the sample groups; the group confidence represents the credibility of the sample wireless channel data in the sample group;

[0008] Target wireless channel data is determined according to the plurality of sample wireless channel data and the corresponding group confidences.

[0009] In a second aspect, an embodiment of the present disclosure provides an electronic device, comprising a memory and a processor; the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, it implements any one of the methods for predicting wireless channel data of the embodiments of the present disclosure.

[0010] In a third aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the computer program is executed by a processor, any one of the methods for predicting wireless channel data of the embodiments of the present disclosure is implemented.

[0011] In the disclosed embodiment, target wireless channel data is obtained by synthesizing (fusion) the group confidence of sample wireless channel data (multiple data sources) from multiple wireless channel modeling processes, that is, wireless channel modeling is implemented, so that it can fully utilize the advantages of various data sources while avoiding their defects, adapt to the evolution of wireless networks, and implement full life cycle modeling, while ensuring high accuracy, low cost, and no impact on normal communication services. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In the accompanying drawings of the embodiments of the present disclosure:

[0013] Figure 1 A flowchart of a method for predicting wireless channel data provided by an embodiment of the present disclosure;

[0014] Figure 2 A flowchart of another method for predicting wireless channel data provided by an embodiment of the present disclosure;

[0015] Figure 3 A block diagram of an electronic device provided in an embodiment of the present disclosure;

[0016] Figure 4 A block diagram of a computer-readable medium provided for an embodiment of the present disclosure;

[0017] Figure 5 A schematic diagram of another method for predicting wireless channel data provided by an embodiment of the present disclosure;

[0018] Figure 6 A process diagram of another method for predicting wireless channel data provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the method for predicting wireless channel data, electronic device, and computer-readable medium provided by the embodiments of the present disclosure are described in detail below in conjunction with the accompanying drawings.

[0020] The present disclosure will be described more fully below with reference to the accompanying drawings, but the embodiments shown may be embodied in different forms, and the present disclosure should not be construed as being limited to the embodiments set forth below. On the contrary, the purpose of providing these embodiments is to make the present disclosure thorough and complete, and will enable those skilled in the art to fully understand the scope of the present disclosure.

[0021] The accompanying drawings of the embodiments of the present disclosure are used to provide a further understanding of the embodiments of the present disclosure, and constitute a part of the specification, and are used together with the detailed embodiments to explain the present disclosure, and do not constitute a limitation of the present disclosure. By describing the detailed embodiments with reference to the accompanying drawings, the above and other features and advantages will become more apparent to those skilled in the art.

[0022] The present disclosure may be described with reference to plan views and / or cross-sectional views by means of ideal schematic views of the present disclosure. Therefore, the exemplary illustrations may be modified according to manufacturing techniques and / or tolerances.

[0023] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0024] The terms used in the present disclosure are only used to describe specific embodiments and are not intended to limit the present disclosure. The term "and / or" as used in the present disclosure includes any and all combinations of one or more related enumerated items. The singular forms "one" and "the" as used in the present disclosure are also intended to include plural forms, unless the context clearly indicates otherwise. The terms "including", "made of..." as used in the present disclosure specify the presence of the features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof.

[0025] Unless otherwise defined, all terms (including technical and scientific terms) used in this disclosure have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning unless explicitly defined in this disclosure.

[0026] The present disclosure is not limited to the embodiments shown in the drawings, but includes modifications of the configurations formed based on the manufacturing process. Therefore, the regions illustrated in the drawings have schematic properties, and the shapes of the regions shown in the drawings illustrate the specific shapes of the regions of the elements, but are not intended to be limiting.

[0027] Wireless channel modeling is a technology that uses certain means to establish an abstract model of a wireless channel (a channel for wireless communication), and studies and simulates the wireless channel based on the established model to predict the wireless channel data. The predicted wireless channel data can be used for wireless network planning, design, testing, finalization, adjustment, and optimization to improve the performance and efficiency of the wireless network.

[0028] In some cases, data-driven wireless channel modeling may be adopted, that is, instead of using a generalized unified model, dedicated modeling for a specific usage scenario (such as a target area) is performed based on data in the specific usage scenario.

[0029] For example, data-driven wireless channel modeling can include the following categories:

[0030] (1) Ray tracing wireless channel modeling

[0031] This method pre-constructs a spatial scene model (three-dimensional model) of the target area to be predicted. The spatial scene model includes information about structures such as buildings and terrain in the target area that will affect the propagation of wireless signals (beams or rays of wireless channels), such as their positions, shapes, surface properties (such as reflectivity to electromagnetic waves), etc.; and then the transmission of wireless signals can be simulated in the spatial scene model, such as calculating what structures the wireless signals will encounter, how they will be reflected / scattered on the structures, etc., and then calculating the situation of the wireless signals reaching a specific location, that is, predicting the wireless channel data, such as the signal-to-noise ratio, multipath amplitude, phase, etc. of the wireless channel.

[0032] Ray tracing wireless channel modeling does not require the use of actual detection devices. It can be achieved based on digital maps, photos, etc. of the target area. It does not require the actual existence of a wireless network and can be performed even when there is no wireless network.

[0033] However, the accuracy of the spatial scene model is limited. For example, the surface properties of different buildings, and even the surface properties of different locations of a building, may be different. In addition, there are moving objects such as vehicles in the target area, and variable conditions such as climate will also affect the propagation of wireless signals. Therefore, it is difficult to establish an accurate spatial scene model that includes all factors that affect the propagation of wireless signals. The wireless channel data obtained based on an inaccurate spatial scene model is naturally not very accurate. At the same time, due to the complexity of the environment, a slight difference in the initial situation of the wireless signal may cause a large change in its propagation situation. Therefore, considering the computational complexity, this method can only simulate the conditions of several main wireless channels, and cannot obtain rich wireless channel data. Therefore, the quantity and accuracy of wireless channel data obtained by ray tracing wireless channel modeling are limited.

[0034] (2) Channel Probing Wireless Channel Modeling

[0035] This method deploys multiple dedicated detection devices (such as Sounder detectors) in the target area that needs to be modeled. These Sounder detectors are interconnected through a wireless network; each Sounder detector obtains the measured data of its own wireless signal through the antenna and the transmitting and receiving circuit, such as channel response, channel loss, transmission time delay, RSSI (received signal strength indicator), RTT (round trip time), etc., and then uploads and stores these measured data to the central server; the central server performs denoising, channel parameter estimation, propagation path analysis and other processing on the measured data, and establishes a model based on the processing results to predict future wireless channel conditions (wireless channel data).

[0036] Channel sounding wireless channel modeling relies on measured data, so it has higher accuracy.

[0037] However, channel detection wireless channel modeling requires the use of a large number of dedicated detection devices, and detection devices must be deployed and detected separately for each target area, which is costly, time-consuming, and expensive, and cannot be used in the stage where the wireless network has not yet been deployed.

[0038] (3) Online wireless channel modeling

[0039] This method uses communication devices (such as mobile phones, base stations, etc.) actually deployed and running in wireless networks as detection devices, records the measured data generated during the communication process, and uses it to model and predict wireless channel data.

[0040] Online network wireless channel modeling also relies on measured data, so it is also highly accurate, and it does not require the use and deployment of dedicated detection devices, making it relatively easy to implement.

[0041] However, online wireless channel modeling of the network relies on communication equipment that is working normally, and consumes bandwidth and resources in normal communication, which will affect normal business, and it cannot be used in the stage where the wireless network has not yet been deployed.

[0042] It can be seen that various wireless channel modeling technologies in related technologies all use data from a single source, so each type of wireless channel modeling has unavoidable inherent defects, cannot adapt to the evolution of wireless networks, cannot perform full life cycle modeling, or has low accuracy, high cost, or affects normal communication services.

[0043] In a first aspect, an embodiment of the present disclosure provides a method for predicting wireless channel data.

[0044] The disclosed embodiments are used to predict (wireless channel modeling) wireless channel conditions in a wireless network (such as a wireless network in a target area) to obtain target wireless channel data, such as the signal-to-noise ratio, multipath amplitude, phase, etc. of the wireless channel.

[0045] Reference Figure 1 , the method for predicting wireless channel data in an embodiment of the present disclosure includes:

[0046] S101. Obtain multiple sample groups.

[0047] Each sample group includes a plurality of sample wireless channel data generated by a wireless channel modeling process.

[0048] S102: Determine the group confidence of each sample group.

[0049] The group confidence represents the credibility of the sample wireless channel data in the sample group.

[0050] S103: Determine target wireless channel data according to a plurality of sample wireless channel data and their corresponding group confidences.

[0051] The disclosed embodiment first obtains multiple sample groups as a basis for calculation, each sample group includes multiple sample wireless channel data of the wireless network to be predicted, and all sample wireless channel data in the same sample group are generated in a wireless channel modeling process, that is, the result of a wireless channel modeling process.

[0052] Among them, "sample wireless channel data generated by a wireless channel modeling process" refers to multiple data detected by a detection device (such as a Sounder detector) through the wireless channel modeling technology, or multiple data detected by multiple detection devices, or multiple data obtained in a prediction process according to this type of wireless channel modeling technology, etc., so that they have the same degree of credibility.

[0053] Each sample wireless channel data may be a channel multipath component, each channel multipath component includes multiple specific parameters, and the j-th channel multipath component of the k-th (k=1,...,K) sample group in a total of K sample groups may be recorded as x k,j , and there is The parameters represent the power, delay, transmitting horizontal azimuth, transmitting vertical azimuth, receiving transmitting horizontal azimuth, and receiving vertical azimuth of the wireless channel respectively.

[0054] It should be understood that the specific parameters of the channel multipath components are not limited to the above examples, and may also include other parameters such as whether the wireless channel is a line-of-sight wireless channel.

[0055] Therefore, each sample group is equivalent to a "set" composed of some channel multipath components. For example, the kth sample group can be recorded as Indicates that the sample group is J k The number of channel multipath components in all sample groups is

[0056] As before, the multiple sample wireless channel data in each sample group are generated through the same wireless channel modeling process, so their credibility should be the same. Therefore, the group confidence of each sample group can be calculated, that is, the credibility of each sample wireless channel data (which must belong to a sample group) can be obtained.

[0057] Furthermore, multiple sample wireless channel data may be fused according to their corresponding group confidences (equivalent to weights) to obtain target wireless channel data as prediction results, thereby realizing wireless channel modeling.

[0058] In the disclosed embodiment, target wireless channel data is obtained by synthesizing (fusion) the group confidence of sample wireless channel data (multiple data sources) from multiple wireless channel modeling processes, that is, wireless channel modeling is implemented, so that it can fully utilize the advantages of various data sources while avoiding their defects, adapt to the evolution of wireless networks, and implement full life cycle modeling, while ensuring high accuracy, low cost, and no impact on normal communication services.

[0059] In some embodiments, the type of wireless channel modeling process includes at least one of the following: ray tracing wireless channel modeling, channel probing wireless channel modeling, and network online wireless channel modeling.

[0060] As one method of an embodiment of the present disclosure, the specific type of the wireless channel modeling process of each sample group may be the above-mentioned ray tracing wireless channel modeling, channel detection wireless channel modeling, network online wireless channel modeling, etc.

[0061] It should be understood that the specific types of wireless channel modeling processes in the embodiments of the present disclosure are not limited to the above examples, and any wireless channel modeling technology that can obtain sample wireless channel data is applicable.

[0062] In some embodiments, the types of wireless channel modeling processes of at least some of the sample groups are different;

[0063] and / or,

[0064] The types of wireless channel modeling processes of at least some sample groups are the same, and the wireless channel modeling processes of any two sample groups are performed in two time periods respectively, the time difference between the two time periods exceeds a preset time threshold, and the duration of each time period does not exceed the time threshold.

[0065] As one method of an embodiment of the present disclosure, sample wireless channel data obtained by different types of wireless channel modeling technologies (such as data collected by different detection devices) belong to different sample groups.

[0066] As another embodiment of the present disclosure, sample wireless channel data obtained by the same wireless channel modeling technology at two times with a long interval (exceeding the time threshold, such as an interval of 1 month) (such as data collected by a collection device on January 1 and February 2) can also be used as different sample groups. This is because when the time interval is long, the situation of the wireless network will change significantly, and thus the credibility of the data obtained by the same wireless channel modeling technology will also change, that is, the group confidence of the sample wireless channel data is different, so it should be regarded as a different sample group. Correspondingly, the acquisition time of multiple sample wireless channel data in a sample group cannot differ too much (such as not exceeding the time threshold).

[0067] In some embodiments, reference Figure 2 , determining the group confidence of each sample group (S102) includes:

[0068] S1021. Determine a group confidence of each sample group according to at least one of a type sub-confidence, a device sub-confidence, and a data sub-confidence of the sample group.

[0069] Among them, the type sub-confidence represents the credibility of the type of the wireless channel modeling process of the sample group, the device sub-confidence represents the credibility of the detection device used in the wireless channel modeling process of the sample group, and the data sub-confidence represents the credibility of the sample wireless channel data itself in the sample group.

[0070] As one method of an embodiment of the present disclosure, the group confidence of each sample group may be calculated based on three components (or some of the components) of type sub-confidence, device sub-confidence, and data sub-confidence.

[0071] The category sub-confidence indicates the credibility of the wireless channel modeling technology used by the sample group, which may be a value preset based on experience. For example, corresponding category sub-confidences may be preset for various wireless channel modeling technologies.

[0072] In the same type of wireless channel modeling technology, different acquisition devices (such as acquisition devices of different grades, different manufacturers, and different models) have different performances, and the credibility of the sample wireless channel data they generate is also different. Therefore, corresponding confidence components, namely, device sub-confidence, can be pre-set for various acquisition devices based on experience.

[0073] The multiple sample wireless channel data obtained in each wireless channel modeling process can also indicate their credibility, so the corresponding data sub-confidence can be directly calculated based on the sample wireless channel data in the sample group.

[0074] It should be understood that there are various specific ways to determine the group confidence according to the type sub-confidence, device sub-confidence, and data sub-confidence. For example, the group confidence w of the kth sample group is k It can be calculated by the following formula:

[0075] w k =q k,1 ·q k,2 ·q k,3 ;

[0076] Among them, q k,1 Indicates the confidence level of the type of the k-th sample group, such as q corresponding to the three methods of ray tracing wireless channel modeling, channel detection wireless channel modeling, and network online wireless channel modeling. k,1 They are 0.6, 1 and 0.8 respectively.

[0077] q k,2 Indicates the confidence level of the device in the kth sample group, such as the q of the high-precision detection device k,2 is 1, while the q of conventional detection devices is k,2 is 0.8.

[0078] qk,3 represents the data sub-confidence of the kth sample group, which is calculated based on the sample wireless channel data in the sample group. For example, M related parameters s1,...s can be obtained based on the sample wireless channel data. M , and set the function q k,3 =f(s1,...s M ) is used to calculate q k,3 .

[0079] It should be understood that the specific calculation method of the group confidence in the embodiment of the present disclosure is not limited to the above example. For example, q k,1 ,q k,2 ,q k,3 Add as group confidence, or just based on q k,1 ,q k, 2.q k,3 The group confidence may be calculated based on one or two sub-confidences in the above method, or the sub-confidences may be selected based on other factors.

[0080] It should be understood that when the types of wireless channel modeling for all data groups are the same, or the models of the detection devices are the same, the values ​​of the type sub-confidence and device sub-confidence used in calculating the group confidence may be the same, or the components of the type sub-confidence and device sub-confidence may not be used in the calculation.

[0081] In some embodiments, the data sub-confidence is calculated in a negative correlation manner based on the dispersion and / or variance of the sample wireless channel data in the sample group.

[0082] As one method of an embodiment of the present disclosure, the discreteness, variance and other parameters of the sample wireless channel data can be first calculated, and the data sub-confidence can be calculated according to an algorithm in which the larger the discreteness and variance, the smaller the data sub-confidence (i.e., a negative correlation method); generally speaking, the sample wireless channel data generated in a wireless channel modeling process should be relatively similar, and the greater the difference, the lower the credibility of the sample wireless channel data itself, and the discreteness, variance, etc. can indicate the similarity of the data, so the data sub-confidence can be calculated in a negative correlation method based on the discreteness, variance, etc.

[0083] It should be understood that the specific algorithm of the data sub-confidence in the embodiment of the present disclosure is not limited to the above example. For example, the data sub-confidence may also be calculated based on parameters such as the signal-to-noise ratio in the sample wireless channel data.

[0084] In some embodiments, reference Figure 2 , determining target wireless channel data (S103) according to the plurality of sample wireless channel data and their corresponding group confidences includes:

[0085] S1031. Cluster multiple sample wireless channel data into multiple clusters.

[0086] S1032: Determine, based on sample wireless channel data in the clusters, that at least some of the clusters are credible clusters.

[0087] S1033: Determine corresponding target wireless channel data respectively according to the sample wireless channel data in each credible cluster and its corresponding group confidence.

[0088] As a method of an embodiment of the present disclosure, the process of determining the target wireless channel data based on the sample wireless channel data can be specifically as follows: first, all the sample wireless channel data are clustered, that is, all the sample wireless channel data are divided into multiple classes (clusters), each cluster includes multiple relatively similar sample wireless channel data; then, based on the situation of the sample wireless channel data in each cluster, it is determined whether the credibility of the cluster itself is high enough, and the relatively credible cluster is determined as a "credible cluster"; further, for each credible cluster determined above, a target wireless channel data corresponding to the credible cluster can be calculated based on the sample wireless channel data therein (that is, multiple relatively similar and highly credible sample wireless channel data) and the group confidence.

[0089] In some embodiments, reference Figure 2 Clustering the plurality of sample wireless channel data into a plurality of clusters (S1031) includes:

[0090] S10311. Unsupervised clustering of multiple sample wireless channel data into multiple clusters.

[0091] As one method of an embodiment of the present disclosure, clustering of sample wireless channel data may specifically be “unsupervised clustering”, such as k-means clustering, k-median clustering, and the like.

[0092] Among them, the basic process of unsupervised clustering includes:

[0093] (1) The number of clusters N to be generated is determined in advance, and N are randomly selected as cluster centers from all sample wireless channel data.

[0094] (2) Each sample wireless channel data is classified into the cluster with the closest cluster center.

[0095] (3) Calculate the center position of all sample wireless channel data in each cluster as a new cluster center, and return to the above step (2) to re-determine the cluster to which each sample wireless channel data belongs based on the new cluster center.

[0096] (4) This cycle continues until the cluster to which all sample wireless channel data belongs no longer changes.

[0097] Thus, N clusters (i.e., the set of channel multipath components) A can be obtained. n ,n=1,…,N, where A n Contains I n Sample wireless channel data (channel multipath component) {x k′,j′}, these sample wireless channel data may come from the same or different sample groups.

[0098] It should be understood that the specific method of unsupervised clustering is not limited to k-means, k-median, etc., and the clustering available in the embodiments of the present disclosure is not limited to unsupervised clustering.

[0099] Each sample wireless channel data (each channel multipath component) can be clustered as a multidimensional vector (the dimension is equal to the number of parameters), or each sample wireless channel data can be converted into a feature vector before clustering, such as the channel multipath component x k,j The characteristic vector of can be defined as:

[0100]

[0101] Among them, β is the normalized weight coefficient;

[0102] Furthermore, in the above clustering process, the two sample wireless channel data x k1,j1 , and x k2,j2 The distance between (or the distance between the sample wireless channel data and the cluster center) can be defined as:

[0103] D(x k1,j1 ,x k2,j2 )=||F(x k1,j1 )-F(x k2,j2 )||;

[0104] Here, ||·|| represents the inner product induced norm (it may also be other functions or operators with similar properties).

[0105] In some embodiments, reference Figure 2 , determining at least part of the clusters as credible clusters according to the sample wireless channel data in the clusters (S1032) includes:

[0106] S10321. Determine at least part of the clusters as credible clusters according to the group confidences corresponding to the sample wireless channel data in the cluster and / or according to the number of sample groups to which the sample wireless channel data in the cluster belongs.

[0107] As one method of an embodiment of the present disclosure, the higher the group confidence of the sample wireless channel data in the cluster, the higher the credibility of the cluster. Therefore, the group confidence of the sample wireless channel data in the cluster can be used as one of the bases for determining a credible cluster.

[0108] As another method of the embodiment of the present disclosure, if the sample wireless channel data in the cluster comes from more different sample groups, the data diversity is good and the inherent defects of a single sample group can be more effectively avoided. Therefore, the number of sample groups from which the sample wireless channel data in the cluster comes can be used as one of the bases for determining a credible cluster.

[0109] It should be understood that the specific algorithm for determining the above trusted clusters may be various.

[0110] For example, for each cluster It can be assumed that there are M sample wireless channel data, and each sample wireless channel data comes from a sample group {k′ m}, then the decision function T({k′ n}) to determine whether the cluster is a credible cluster, that is, T({k′ m )}=1 represents a credible cluster, T({k′ m )}=0 represents that it is not a trusted cluster.

[0111] Furthermore, T({k′ n}) can be specifically expressed as:

[0112]

[0113] Among them, Threshold is a preset threshold value; that is, it can be considered as the sum of the group confidences of all sample wireless channel data in a cluster. If it exceeds the threshold value, the cluster is a credible cluster, and if it does not exceed the threshold value, the cluster is not a credible cluster.

[0114] Or, T({k′ n}) can also be expressed as:

[0115]

[0116] That is, it can be considered that if a cluster includes sample wireless channel data from all sample groups, it is a credible cluster; and if a cluster lacks sample wireless channel data from one or more sample groups, it is not a credible cluster.

[0117] Or, T({k′ n}) The above two judgment criteria can also be combined at the same time, such as:

[0118] T3({k′ m})=T1({k′ m})T2({k′ m}).

[0119] It should be understood that the specific method of determining a credible cluster is not limited to the above examples. For example, when the number of source sample groups of sample wireless channel data in a cluster exceeds a predetermined value (not all), it can be considered a credible cluster; or, a sub-value can be calculated based on the group confidence of the sample wireless channel data in the cluster and the number of source sample groups, and then whether it is a credible cluster is determined based on whether the sum of the sub-values ​​reaches a predetermined value; or, other parameters of the cluster (such as the number of sample wireless channel data in the cluster) can be used as one of the parameters for determining a credible cluster, etc.

[0120] In some embodiments, determining the corresponding target wireless channel data (S1033) respectively according to the sample wireless channel data in each trusted cluster and its corresponding group confidence includes:

[0121] S10331. Multiply each sample wireless channel data in each trusted cluster by its corresponding group confidence and then sum them up. Divide the sum by the sum of the group confidences corresponding to all sample wireless channel data in the trusted cluster to obtain the target wireless channel data corresponding to the trusted cluster.

[0122] As one method of an embodiment of the present disclosure, for each credible cluster, the group confidence therein may be used as the "weight" of the corresponding sample wireless channel data, and the target wireless channel data may be calculated based on the sample wireless channel data in a weighted average manner.

[0123] For example, suppose from N clusters A n ,n=1,...,N select L credible clusters B l ,n=1,...,L, where the lth credible cluster Including V sample wireless channel data, the target wireless channel data including multiple parameters can be calculated by the following formula:

[0124]

[0125] Furthermore, for L credible clusters, a total of L y l , that is, l target wireless channel data as prediction results are obtained.

[0126] It should be understood that the method for predicting wireless channel data according to the embodiment of the present disclosure may further include other steps.

[0127] For example, various detectors (such as Sounder detectors, etc.) required for wireless channel modeling techniques may be deployed in the target area in advance, data (sample wireless channel data) acquired by these detectors may be collected, and data from the same detector may be used to form a sample group.

[0128] For example, the sample wireless channel data collected by the detection device, or the sample wireless channel data in the sample group, can also be pre-processed (i.e., data cleaning) to delete and exclude obvious erroneous data, such as excessive power values ​​and delays, and scheduling that does not conform to the value range.

[0129] Example 1:

[0130] The following is an exemplary introduction to a specific method for predicting wireless channel data in an embodiment of the present disclosure, referring to Figure 5 , which may include the following steps:

[0131] (1) Wireless channel data (sample wireless channel data) collection

[0132] Ray tracing channel generators (detection devices used in ray tracing wireless channel modeling technology), Sounder detectors (detection devices used in channel detection wireless channel modeling technology), and communication equipment (detection devices used in network online wireless channel modeling technology) are deployed in the target area.

[0133] Each detection device can collect data of multiple wireless channels, that is, multiple channel multipath components, and output them as a data set (sample group).

[0134] There are three detectors in total, and the data set output by the kth (k=1,...,K; K=3) detector is recorded as It contains J k channel multipath components, each channel multipath component is counted as Among them, the parameters respectively represent the power, delay, horizontal transmission angle, vertical transmission angle, horizontal receiving angle, and vertical receiving angle of the wireless channel.

[0135] (2) Wireless channel data preprocessing

[0136] For the above obtained The multipath components of each channel are filtered to exclude obviously erroneous data, such as excessive power values ​​and delays, and scheduling that does not conform to the value range.

[0137] (3) Determine the confidence coefficient of the data set (group confidence)

[0138] Assign the corresponding confidence coefficient w to the filtered wireless channel data from the kth data set k, the confidence coefficient is calculated as follows:

[0139] w k =q k,1 ·q k,2 ·q k,3 ;

[0140] Among them, q k,1 The confidence level (type sub-confidence level) corresponding to the type of wireless channel modeling technology for the kth data set is specifically set to q corresponding to the ray tracing channel generator, Sounder detector, and communication device. k,1 are 0.6, 1, and 0.8 respectively; and q k,2 represents the confidence of the specific model of the detection device used in the kth data set (device sub-confidence), all set to 1; q k,3 represents the confidence of the wireless channel data itself in the kth data set (data sub-confidence), set q k,3 It is equal to the inverse of the variance of the wireless channel data in the data set.

[0141] (4) Wireless Channel Data Clustering

[0142] Will a total of The multipath components of the channels are taken as sample points for unsupervised clustering, where the sample point x k,j The characteristic vector of is defined as:

[0143]

[0144] Among them, β is the inverse of the maximum delay of all sample points, and then the two sample points x in the cluster k1,j1 , and x k2,j2 The distance between can be defined as:

[0145] D(x k1,j1 ,x k2,j2 )=||F(x k1,j1 )-F(x k2,j2 )||;

[0146] Among them, ||·|| is the inner product induced norm.

[0147] The unsupervised clustering specifically uses the K-Means algorithm, and sets the number of clusters obtained to N, where N is the maximum value of all channel multipath components.

[0148] The sample points {x k′,j} as an initial classification set (cluster) output, generating a total of N initial classification sets A n ,n=1,...,N, where A n Contains I nchannel multipath components, which may come from one or more data sets.

[0149] (5) Reliability determination of initial classification set

[0150] For each initial classification set Assuming that it includes M channel multipath components, the source data set {k′ m}, using the decision function T({k′ n}) to determine whether the initial classification set is reliable, and T({k′ m The initial classification set with})=1 is set as the credible classification set (credible cluster).

[0151] Among them, T({k′ n}) is represented as:

[0152]

[0153] Thus, L credible classification sets B are determined l ,n=1,…,L.

[0154] (6) Obtain estimated wireless channel multipath components (target wireless channel data)

[0155] Assume that the lth credible classification set B l Contains V channel multipath components, that is, Then for the Perform the following operations (specifically, perform them on each parameter separately):

[0156]

[0157] And get the new estimated channel multipath component

[0158] Thus, a total of L channel multipath components y are obtained l As the final prediction result.

[0159] Example 2:

[0160] The following is an exemplary introduction to a specific method for predicting wireless channel data in an embodiment of the present disclosure, referring to Figure 6 , which may include the following steps:

[0161] (1) Wireless channel data (sample wireless channel data) collection

[0162] In the target area, the wireless channel data is collected using communication equipment (detection equipment used in online wireless channel modeling technology).

[0163] The communication equipment collects multiple wireless channel data every day on May 1, June 1, and July 1, that is, collects multiple channel multipath components, where the multiple channel multipath components collected on the same day are output as a data set (sample group), while the channel multipath components collected on different days belong to different data sets.

[0164] The data are collected on 3 days respectively, and the data set collected on the kth day (k = 1, ..., K; K = 3) is recorded as It contains J k channel multipath components, each channel multipath component is counted as Among them, the parameters respectively represent the power, delay, horizontal transmission angle, vertical transmission angle, horizontal receiving angle, and vertical receiving angle of the wireless channel.

[0165] (2) Wireless channel data preprocessing

[0166] For the above obtained The multipath components of each channel are filtered to exclude obviously erroneous data, such as excessive power values ​​and delays, and scheduling that does not conform to the value range.

[0167] (3) Determine the confidence coefficient of the data set (group confidence)

[0168] Assign the corresponding confidence coefficient w to the filtered wireless channel data from the kth data set k , the confidence coefficient is calculated as follows:

[0169] w k =q k,1 ·q k,2 ·q k,3 ;

[0170] Among them, q k,1 represents the confidence level (type sub-confidence level) corresponding to the type of wireless channel modeling technology of the kth data set, q k,2 represents the confidence of the specific model of the detection device used in the kth data set. Since the same communication equipment is used for data collection in Example 2, q k,1 ,q k,2 are all set to 1; q k,3 represents the confidence of the wireless channel data itself in the kth data set (data sub-confidence), set q k,3 is the exponential function of the signal-to-noise ratio of the wireless channel data in the data set, that is:

[0171] q k,3 =γexp(SNR k );

[0172] Among them, γ is the normalization parameter.

[0173] (4) Wireless Channel Data Clustering

[0174] Will a total of The multipath components of the channels are taken as sample points for unsupervised clustering, where the sample point x k,j The characteristic vector of is defined as:

[0175]

[0176] Among them, β is the inverse of the maximum delay of all sample points, and then the two sample points x in the cluster k1,j1 , and x k2,j2 The distance between can be defined as:

[0177] D(x k1,j1 ,x k2,j2 )=||F(x k1,j1 )-F(x k2,j2 )||;

[0178] Here, ||·|| is a function with the same properties as the inner product induced norm.

[0179] The unsupervised clustering specifically uses the k-medeian algorithm, and sets the number of obtained clusters to N, where N is the maximum value of all channel multipath components.

[0180] The sample points {x k′,j} as an initial classification set (cluster) output, generating a total of N initial classification sets A n ,n=1,...,N, where A n Contains I n channel multipath components, which may come from one or more data sets.

[0181] (5) Reliability determination of initial classification set

[0182] For each initial classification set Assuming that it includes M channel multipath components, the source data set {k′ m}, using the decision function T({k′ n}) to determine whether the initial classification set is reliable, and T({k′ m The initial classification set with})=1 is set as the credible classification set (credible cluster).

[0183] Among them, T({k′ n}) is represented as:

[0184]

[0185] Among them, Threshold is a preset threshold value.

[0186] Thus, L credible classification sets B are determined l ,n=1,…,L.

[0187] (6) Obtain estimated wireless channel multipath components (target wireless channel data)

[0188] Assume that the lth credible classification set B l Contains V channel multipath components, that is, For Perform the following operations (specifically, perform them on each parameter separately):

[0189]

[0190] And get the new estimated channel multipath component

[0191] Thus, a total of L channel multipath components y are obtained l As the final prediction result.

[0192] Second, refer to Figure 3 An embodiment of the present disclosure provides an electronic device, which includes a memory and a processor; the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, any method for predicting wireless channel data in the embodiments of the present disclosure is implemented.

[0193] Thirdly, refer to Figure 4 An embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon. When the computer program is executed by a processor, any method for predicting wireless channel data in the embodiments of the present disclosure is implemented.

[0194] Among them, the processor is a device with data processing capabilities, including but not limited to the central processing unit (CPU); the memory is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) is connected between the processor and the memory, and can realize information exchange between the memory and the processor, including but not limited to the data bus (Bus), etc.

[0195] Those skilled in the art will appreciate that all or some of the steps, systems, and functional modules / units in the apparatus disclosed above may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0196] In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed by several physical components in cooperation.

[0197] Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit (CPU), a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-temporary medium) and a communication medium (or temporary medium). As known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH) or other disk storage; read-only compact disk (CD-ROM), digital versatile disk (DVD) or other optical disk storage; magnetic cassettes, magnetic tapes, disk storage or other magnetic storage; any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0198] The present disclosure has disclosed example embodiments, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly stated, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. A method for predicting wireless channel data, comprising: Acquire multiple sample groups; each of the sample groups includes multiple sample wireless channel data generated by a wireless channel modeling process; Determining a group confidence level for each of the sample groups; The group confidence represents the credibility of the sample wireless channel data in the sample group; Determining target wireless channel data according to the plurality of sample wireless channel data and the corresponding group confidences; Determining target wireless channel data according to the plurality of sample wireless channel data and the corresponding group confidences includes: Clustering a plurality of the sample wireless channel data into a plurality of clusters; Determining, based on the sample wireless channel data in the cluster, that at least part of the cluster is a credible cluster; The corresponding target wireless channel data is determined respectively according to the sample wireless channel data in each of the trusted clusters and the corresponding group confidences.

2. The method according to claim 1, wherein: The types of the wireless channel modeling processes of at least some of the sample groups are different; and / or, The types of wireless channel modeling processes of at least some of the sample groups are the same, and the wireless channel modeling processes of any two of the sample groups are performed in two time periods respectively, the time difference between the two time periods exceeds a preset time threshold, and the duration of each time period does not exceed the time threshold.

3. The method according to claim 1, wherein: The type of the wireless channel modeling process includes at least one of the following: Ray tracing wireless channel modeling, channel detection wireless channel modeling, network online wireless channel modeling.

4. The method according to claim 1, wherein: Determining the group confidence of each sample group includes: The group confidence of the sample group is determined based on at least one of the type sub-confidence, device sub-confidence, and data sub-confidence of each sample group; the type sub-confidence represents the credibility of the type of the wireless channel modeling process of the sample group, the device sub-confidence represents the credibility of the detection device used in the wireless channel modeling process of the sample group, and the data sub-confidence represents the credibility of the sample wireless channel data itself in the sample group.

5. The method according to claim 4, wherein: The data sub-confidence is calculated in a negative correlation manner according to the discreteness and / or variance of the sample wireless channel data in the sample group.

6. The method according to claim 1, wherein: The clustering of the plurality of sample wireless channel data into a plurality of clusters comprises: The plurality of sample wireless channel data are unsupervisedly clustered into a plurality of clusters.

7. The method according to claim 1, wherein: The determining, based on the sample wireless channel data in the cluster, that at least part of the cluster is a credible cluster comprises: At least part of the clusters are determined to be credible clusters according to the group confidence corresponding to the sample wireless channel data in the cluster and / or according to the number of sample groups to which the sample wireless channel data in the cluster belongs.

8. An electronic device comprising a memory and a processor; the memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the method for predicting wireless channel data according to any one of claims 1 to 7 is implemented.

9. A computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for predicting wireless channel data according to any one of claims 1 to 7 is implemented.

Images