CN115052309A

CN115052309A - Interference detection method, device, equipment and storage medium

Info

Publication number: CN115052309A
Application number: CN202110257673.5A
Authority: CN
Inventors: 席志成; 徐晓景; 芮华; 林伟
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2021-03-09
Filing date: 2021-03-09
Publication date: 2022-09-13
Also published as: WO2022188467A1

Abstract

The application provides an interference detection method, an interference detection device, interference detection equipment and a storage medium. The method comprises the following steps: under the condition that the interference type of the data to be detected is not detected by adopting a pre-established first time granularity detection model, filtering the combination of the data to be detected and a predetermined first interference detection probability to obtain a first filtering result; determining the interference detection probability of the first filtering result in a pre-established current time granularity detection model as a second interference detection probability, wherein the current time granularity detection model is a detection model with coarser time granularity than the first time granularity detection model; and determining the interference type of the data to be detected according to the comparison result of the second interference detection probability and a first preset detection probability threshold value. By adopting the multi-stage time granularity detection model, the interference of different time granularities is adapted, the capability of identifying the interference type is improved, and the interference identification precision under a complex network scene is further improved.

Description

Interference detection method, device, equipment and storage medium

Technical Field

The present application relates to communications, and in particular, to an interference detection method, apparatus, device, and storage medium.

Background

Interference is one of the important problems faced by wireless networks, and with the continuous construction of various wireless networks, various potential interference sources are continuously generated at an incredible speed, and the wireless networks are faced with complex interference environments. The communication products occupy the existing frequency resources of other networks, the network configuration of operators is not proper, the problems of transmitters, the overlapping of frequency spectrum resources, special interference and the like are all the reasons for the generation of the wireless network interference. Network operators want to optimize network performance and improve communication quality through interference identification. The existing interference identification method has low accuracy and cannot adapt to the existing complex network environment. Therefore, how to increase the interference detection rate to adapt to a complex network environment is a problem to be solved at present.

Disclosure of Invention

In view of this, embodiments of the present application provide an interference detection method, apparatus, device, and storage medium, which adapt to interference of different time granularities by using a multi-level time granularity detection model, thereby improving the ability of identifying the interference type and improving the accuracy of identifying the interference in a complex network scenario.

The embodiment of the application provides an interference detection method, which comprises the following steps:

under the condition that the interference type of the data to be detected is not detected by adopting a pre-established first time granularity detection model, filtering the combination of the data to be detected and a predetermined first interference detection probability to obtain a first filtering result;

determining an interference detection probability of the first filtering result in a pre-established current time granularity detection model as a second interference detection probability, wherein the current time granularity detection model is a detection model with coarser time granularity than the first time granularity detection model;

and determining the interference type of the data to be detected according to the comparison result of the second interference detection probability and a first preset detection probability threshold value.

The embodiment of the application provides an interference detection device, includes:

the first filter is configured to filter a combination of the data to be detected and a first interference detection probability which is predetermined under the condition that the interference type of the data to be detected is not detected by adopting a first time granularity detection model which is established in advance, so as to obtain a first filtering result;

a first determining module configured to determine, as a second interference detection probability, an interference detection probability of the first filtering result in a pre-created current time granularity detection model, wherein the current time granularity detection model is a detection model with a coarser time granularity than that of the first time granularity detection model;

and the second determining module is configured to determine the interference type of the data to be detected according to a comparison result of the second interference detection probability and a first preset detection probability threshold value.

The embodiment of the application provides an interference detection device, which comprises: a communication module, a memory, and one or more processors;

the communication module is configured to perform communication interaction between the communication nodes;

the memory configured to store one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the method of any of the embodiments described above.

The embodiment of the present application provides a storage medium, wherein the storage medium stores a computer program, and the computer program realizes the method of any one of the above embodiments when being executed by a processor.

According to the technical scheme, under the condition that the interference type of the data to be detected is not detected by the pre-established first time granularity detection model, the data to be detected is subjected to interference detection in the pre-established current time granularity detection model until the interference type of the data to be detected is detected, the interference type of the data to be detected is detected by the multi-stage time granularity detection model, so that the interference of different time granularity characteristics is adapted, the interference type identification capability is improved, and the interference identification precision under a complex network scene is improved.

Drawings

Fig. 1 is a flowchart of an interference detection method according to an embodiment of the present application;

fig. 2 is a block diagram of an interference detection apparatus according to an embodiment of the present disclosure;

FIG. 3 is a block diagram of a training unit according to an embodiment of the present disclosure;

fig. 4 is a block diagram of a detection unit according to an embodiment of the present disclosure;

FIG. 5 is a schematic diagram illustrating creation of a time granularity detection model according to an embodiment of the present application;

fig. 6 is a schematic diagram of interference detection provided in an embodiment of the present application;

FIG. 7 is a schematic diagram of another time granularity detection model creation provided by an embodiment of the present application;

fig. 8 is a schematic diagram of another interference detection provided in the embodiment of the present application;

fig. 9 is a schematic diagram of another interference detection provided in the embodiment of the present application;

fig. 10 is a block diagram of another interference detection apparatus according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of an interference detection device according to an embodiment of the present application.

Detailed Description

Hereinafter, embodiments of the present application will be described with reference to the drawings. The present application is described below with reference to the accompanying drawings of embodiments, which are provided for illustration only and are not intended to limit the scope of the present application.

In an embodiment, fig. 1 is a flowchart of an interference detection method provided in the embodiment of the present application. The present embodiment may be performed by an interference detection device. The interference detection device may be a terminal side (e.g., a user equipment), among others. As shown in fig. 1, the present embodiment includes: S110-S130.

S110, under the condition that the interference type of the data to be detected is not detected by adopting the pre-established first time granularity detection model, filtering the combination of the data to be detected and the predetermined first interference detection probability to obtain a first filtering result.

Wherein, the first time granularity detection model refers to a pre-created first time granularity level training model. In the actual operation process, a pre-established AI model is trained by using a first time granularity characteristic value until a satisfactory training effect is obtained, and then a first time granularity training model is output and is used as a first time granularity detection model. Illustratively, the first time-granular detection model may include one of: a time slot level detection model; a minute-scale detection model; a day-level detection model; and (4) a week-level detection model.

In the embodiment, when the first time granularity detection model is not used to detect the interference type of the data to be detected, it indicates that the interference characteristic of the data to be detected is not at the first time granularity, and at this time, a time granularity detection model corresponding to a previous time granularity of the time granularity corresponding to the first time granularity detection model may be used to perform further interference detection on the data to be detected. Firstly, filtering a combination of data to be detected and a predetermined first interference detection probability to obtain a first filtering result. The first interference detection probability refers to the probability of outputting the data to be detected in the first time granularity detection model.

In an embodiment, since different time granularity detection models correspond to different time lengths, time-domain filtering is performed on a combination of data to be detected and a predetermined first interference detection probability to obtain a filtering result.

And S120, determining the interference detection probability of the first filtering result in a pre-created current time granularity detection model as a second interference detection probability.

And the current time granularity detection model is a detection model with coarser time granularity than the first time granularity detection model. In an embodiment, the time granularity corresponding to the current time granularity detection model is coarser than the time granularity corresponding to the first time granularity detection model. Exemplarily, in case that the first time granularity detection model is a slot-level detection model, the current time granularity detection model is a minute-level detection model; under the condition that the first time granularity detection model is a minute-level detection model, the current time granularity detection model is a day-level detection model; and under the condition that the first time granularity detection model is the day-level detection model, the current time granularity detection model is the week-level detection model.

In an embodiment, the interference detection probability may be understood as a frequency of occurrence of the first filtered result at the current time granularity, i.e. the interference detection probability may be directly calculated from a frequency of occurrence of the first filtered result at the current time granularity. Exemplarily, it is assumed that the current time granularity detection model is a timeslot-level detection model, and the total detection duration is 10 timeslots, where 2 timeslots have interference 1 and 4 timeslots have interference 2 within the total detection duration, the interference detection probability of the interference 1 is 0.2, and the interference detection probability of the interference 2 is 0.4.

S130, determining the interference type of the data to be detected according to the comparison result of the second interference detection probability and the first preset detection probability threshold value.

The first preset detection probability threshold value is a preset detection probability threshold value. In the embodiment, when the second interference detection probability is greater than the first preset detection probability threshold value, it indicates that the interference in the current time granularity in the data to be detected is detected; and under the condition that the second interference detection probability is smaller than the first preset detection probability threshold value, indicating that no interference is detected in the current time granularity, and not marking the filtering result as any interference. The scheme can be used for intelligently detecting the interference type in the data to be detected by combining AI, thereby being suitable for complex network scenes.

In an embodiment, when the second interference detection probability is smaller than the corresponding first preset detection probability threshold, the method further includes: and switching the current time granularity detection model to a time granularity detection model with coarser time granularity, and returning to the step of filtering the combination of the data to be detected and the predetermined first interference detection probability to obtain a first filtering result until the interference type of the data to be detected is detected. In the embodiment, under the condition that the interference type of the data to be detected is not detected by adopting the first time granularity detection model, the current time granularity detection model is switched to the time granularity detection model with the coarser time granularity, the interference detection is carried out on the data to be detected by utilizing the next time granularity detection model with the coarser time granularity until the interference type in the data to be detected is detected, so that the interference detection is carried out on the data to be detected on different time granularities by utilizing the multi-stage time granularity detection model respectively, and the detection results of the time granularities at different levels are given out, thereby improving the interference detection precision under the complex network scene.

In an embodiment, the filtering a combination of data to be detected and a predetermined first interference detection probability to obtain a first filtering result includes: combining the data to be detected and a predetermined first interference detection probability to obtain combined data; and filtering the time length of the current time granularity of the combined data to obtain a first filtering result. In the embodiment, the process of combining the data to be detected and the first interference detection probability refers to combining the first interference detection probability and the data to be detected together as combined data. Illustratively, it is assumed that the data to be detected character _slotincludes: [ RSSI ', NI ', Cha ' _space ,Cha’ _time ,Cha’ _freq1 ,Cha’ _freq2 ]The first interference detection probability includes P1 '_ slot, P2' _ slot, P6 '_ slot, and the combined data is [ RSSI', NI ', Cha' _space ,Cha’ _time ,Cha’ _freq1 ,Cha’ _freq2 ，P1′_slot,P2′_slot,...,P6′_slot]。

In an embodiment, the filtering the combined data for a time length of the current time granularity to obtain a first filtering result includes: and filtering the combined data according to the instantaneous combined data of the current time granularity, the combined data after the previous current time granularity is filtered, the first preset weight coefficient and the second preset weight coefficient to obtain a first filtering result. In an embodiment, the first preset weight coefficient and the second preset weight coefficient are empirical values obtained in advance through a large number of experiments. It is understood that different preset weighting factors are set according to different time granularities. It should be noted that the first preset weight coefficientThe second preset weighting factor may be equal to or different from the first preset weighting factor, which is not limited herein. Of course, in the actual operation process, the first preset weight coefficient and the second preset weight coefficient may also be adjusted according to the actual situation, which is not limited herein. Exemplarily, assuming that the current time granularity is minutes, the instantaneous combination data of the current time granularity refers to the instantaneous combination data of the current 1 minute; the previous current time-granularity filtered combined data refers to the previous 1-minute filtered combined data. For example, let's train index _ minute' _t Represents the combined data, train data _ minute ', obtained after the current 1-minute filtering' _t-1 Represents the combined data after the previous 1 minute filtering, traindata _ minute _t Represents the instantaneous combined data of the current 1 minute, the first preset weight coefficient is 0.1, the second preset weight coefficient is 0.9, then train data _ minute' _t ＝0.9*traindata_minute′ _t-1 +0.1*traindata_minute _t 。

In an embodiment, before the interference type of the data to be detected is not detected by using the pre-created first time granularity detection model, the method further includes:

inputting a predetermined first time granularity characteristic value into a first AI training model which is created in advance until the detection rate reaches a first preset detection rate threshold value, outputting the first AI training model and a first time granularity interference probability distribution value, and taking the first AI training model as a first time granularity detection model;

combining the first time granularity characteristic value and the first time granularity interference probability distribution value to serve as a second time granularity characteristic value;

performing time domain filtering on the second time granularity characteristic value to obtain second time granularity level training data; inputting second time granularity level training data into a second pre-established AI training model until the detection rate reaches a second preset detection rate threshold value, outputting the second AI training model and a second time granularity interference probability distribution value, and taking the second AI training model as a second time granularity detection model; and the time granularity corresponding to the first time granularity detection model and the second time granularity detection model is successively coarser.

In an embodiment, after the using the second AI training model as the second time granularity detection model, the method further includes:

merging the second time granularity level training data and the second time granularity interference probability distribution value to serve as a third time granularity characteristic value;

performing time domain filtering on the third time granularity characteristic value to obtain third time granularity training data;

inputting third time granularity level training data into a third pre-established AI training model until the detection rate reaches a third preset detection rate threshold value, outputting the third AI training model and a third time granularity interference probability distribution value, and taking the third AI training model as a third time granularity detection model; and the time granularity corresponding to the first time granularity detection model, the second time granularity detection model and the third time granularity detection model is successively thickened.

For example, when the first time granularity detection model is a time slot level detection model, the time granularity corresponding to the second time granularity detection model is coarser than the time granularity corresponding to the first time granularity detection model, for example, the second time granularity detection model may be a minute level detection model or a day level detection model; accordingly, the time granularity corresponding to the third time granularity detection model is coarser than the time granularity corresponding to the second time granularity detection model, for example, the third time granularity detection model may be a cycle-level detection model.

It should be noted that, in the process of training the time granularity detection models, it refers to training at least two time granularity detection models corresponding to time granularity, that is, the time granularity detection models corresponding to two time granularities may be trained, or the time granularity detection models corresponding to three time granularities may also be trained, or even the time granularity detection models corresponding to four or five … … N time granularities may be trained. Of course, if the time granularity detection model corresponding to the more time granularities is trained, the higher the computing power requirement of the device.

In one embodiment, the first time-granular characteristic value includes one of: time domain RSSI; frequency domain NI; spatial domain matching filtering; a time domain correlation value; a frequency domain correlation value; intermediate RB energy distribution.

In one embodiment, the temporal granularity detection model includes one of: a time slot level detection model; a minute-scale detection model; a day-level detection model; and (4) a week-level detection model.

In an embodiment, an interference detection process is described by taking an example in which the interference detection apparatus includes two processing units. Fig. 2 is a block diagram of an interference detection apparatus according to an embodiment of the present application. As shown in fig. 2, the interference detection apparatus in this embodiment includes: a training unit 210 and a detection unit 220.

Fig. 3 is a block diagram of a training unit according to an embodiment of the present disclosure. As shown in fig. 3, the training unit 210 is composed of several stages of subunits, namely a data preprocessing unit 2101 and a neural network training unit 2102. The data preprocessing unit 2101 is configured to perform operations such as feature value extraction and data filtering on data; the neural network training unit 2102 is configured to send the obtained training sample data to a neural network for training, adjust training parameters, and finally obtain a trained detection model.

Fig. 4 is a block diagram of a detection unit according to an embodiment of the present disclosure. As shown in fig. 4, the detection unit 220 is composed of several stages of sub-units, namely a data preprocessing unit 2201, a model selection unit 2202, a decision unit 2203 and a result collection unit 2204. The data preprocessing unit 2201 is configured to perform operations such as feature value extraction and data filtering on data to generate data with corresponding time granularity; a model selection unit 2202 is used for selecting a corresponding trained model; a judgment unit 2203 for judging the data type using the model selected by the model selection unit; the result collecting unit 2204 is used for collecting and sorting the output results of each level of model.

In one embodiment, the first time-granular characteristic value comprises: time domain RSSI, frequency domain NI, spatial domain matched filtering, time domain correlation values, frequency domain correlation values and middle RB energy distribution are taken as examples to explain the interference detection process in the 5G uplink system. Table 1 is a schematic table of an initial training data structure provided in an embodiment of the present application. As shown in table 1, in this embodiment, the combination of the measurement quantities in the 5G uplink system is: time domain RSSI, frequency domain NI, spatial domain matched filtering, time domain correlation values, frequency domain correlation values and intermediate RB energy distribution. Table 2 is an exemplary table of interference types in a 5G NR uplink system according to an embodiment of the present disclosure. As shown in table 2, the interference types of the 5G NR uplink system may include: atmospheric waveguide interference, frame out-of-sync interference, D1D2 interference, D4D5 interference, sideband interference, and narrowband interference.

TABLE 1 initial training data structure schematic table

TABLE 2 interference type indication table in 5G NR uplink system

Interference sequence number	Interference class
		1	Atmospheric waveguide interference
2	Frame out-of-step interference
		3	D1D2 interference
4	D4D5 interference
		5	Sideband interference
6	Narrow-band interference

Before the interference detection is carried out on the data to be detected by adopting the detection unit, a time granularity detection model is established by utilizing the training unit. Fig. 5 is a schematic diagram of creating a time granularity detection model according to an embodiment of the present application. For example, in this embodiment, a training process of a time granularity detection model corresponding to three time granularities is described. For example, the first time granularity detection model, the second time granularity detection model, and the third time granularity detection model sequentially include: the time slot level detection model, the minute level detection model and the day level detection model are taken as examples, and the creation process of the time granularity detection model is explained. As shown in fig. 5, the creation process of the time granularity detection model includes the following steps:

step one, RSSI values of 14 symbols in a slot (slot) are obtained _symbol (also referred to as time domain RSSI), where symbol is 0, 1.

Step two, calculating a correlation coefficient Coef between adjacent symbols _{symbol1,symbol2} (also called time domain correlation value), wherein symbol1 is 0, 1.. 12, symbol2 is symbol2+1, so that 13 time domain features Cha are finally obtained _time 。

Step three: obtaining NI value NI of 273 RBs in slot _rb (may also be referred to as frequency domain NI) where rb is 0, 1.

Step four: calculating a correlation coefficient Coef between adjacent RBs _rb1,rb2 (also called frequency domain correlation value), where rb1 ═ 0, 1.. 271, rb2 ═ rb1+1, finally resulting in 272 frequency domain features.

Step five: obtaining the receiving Power value Power of all sub-carriers on the middle RB, namely 137 th RB _sc (also called intermediate RB energy division)Cloth), sc ═ 0,1,. multidot.11, resulting in 12 frequency domain features Cha1 _freq 。

Step six: obtaining the matched filtering results (also called spatial matched filtering) of different wave beams, assuming that the number of the wave beams is n, obtaining n spatial domain characteristics Cha _space 。

Step seven: the characteristics to be finally extracted are [ RSSI, NI, Cha ] _space ,Cha _time ,Cha _freq1 ,Cha _freq2 ]And is denoted as character _ slot.

Step eight: sending the character _ slot into a pre-established AI model for training to obtain a satisfactory training effect (whether the interference detection accuracy reaches a first preset detection rate threshold value can be judged, and when the interference detection accuracy reaches the first preset detection rate threshold value, the satisfactory training effect is considered to be obtained), then outputting two results, namely a slot-level training model (AI _ slot, namely the first AI training model in the above embodiment), and storing the slot-level training model; second, a preset interference probability distribution value (Pn _ slot, i.e., the first time granularity interference probability distribution value in the above embodiment) of the slot, where Pn _ slot is [ P1_ slot, P2_ slot, P3_ slot, P4_ slot, P5_ slot, P6_ slot ].

Step nine: and a data preprocessing unit of the training unit merges the character _ slot output in the step seven and the Pn _ slot output in the step eight into the tranndata _ slot, wherein the tranndata _ slot is [ RSSI, NI, Chaspace, Chatime, Chafreq1, Chafreq2, P1_ slot, P2_ slot, P3_ s lot, P4_ slot, P5_ slot, P6_ slot ], performs time-domain filtering on the tranndata _ slot for 1 minute to obtain minute-level training data which is recorded as character _ slot.

The filtering mode of the traindata _ slot is as follows:

traindata_slot′ _t ＝0.9*traindata_slot′ _t-1 +0.1*traindata_slot _t wherein, traindata _ slot' _t Represents the training data, train _ slot ', obtained after the current slot filtering' _t-1 Represents the previously slot filtered training data, thinndata _ slot _t The instantaneous training data representing the current slot.

Step ten: sending the character _ minimum into a pre-established AI model for training to obtain a satisfactory training effect (whether the interference detection accuracy reaches a second preset detection rate threshold value can be judged, and when the interference detection accuracy reaches the second preset detection rate threshold value, the satisfactory training effect is considered to be obtained), and then outputting two results, namely, an AI _ minimum (i.e. the second AI training model in the above embodiment) and storing the minimum training model; second is a preset interference probability distribution value (Pn _ min, i.e. the second time granularity interference probability distribution value in the above embodiment) of the min, where Pn _ min ═ P1_ min, P2_ min, P3_ min, P4_ min, P5_ min, P6_ min.

Step eleven: and a data preprocessing unit of the training unit merges the character _ minute output in the step nine and the Pn _ minute output in the step ten into a train _ minute, wherein the train _ minute is [ RSSI, NI, chasspace, Chatime, Chafreq1, Chafreq2, P1_ minute, P2_ minute, P3_ minute, P4_ minute, P5_ minute, P6_ minute ], performs time-domain filtering on the train _ minute for 1 day to obtain day-level training data, and the day-level training data is recorded as a train _ day.

The filtering mode of traindata _ min ute is as follows:

traindata_minute′ _t ＝0.9*traindata_minute′ _t-1 +0.1*traindata_minute _t wherein, traindata _ minute' _t Represents the training data, train _ minute ', obtained after the current 1-minute filtering' _t-1 Represents the previous 1 minute filtered training data, train _ minute _t Representing the current 1 minute of instantaneous training data.

Step twelve: sending the character _ day into a pre-created AI model for training to obtain a satisfactory training effect (whether the interference detection accuracy reaches a third preset detection rate threshold value can be judged, and when the interference detection accuracy reaches the third preset detection rate threshold value, the satisfactory training effect is considered to be obtained), then outputting two results, namely a day-level training model (AI _ day, namely the third AI training model in the embodiment), and storing the day-level training model; second, the preset interference probability distribution value (Pn _ day, i.e. the third time granularity interference probability distribution value in the above embodiment) of this day, where Pn _ day is [ P1_ day, P2_ day, P3_ day, P4_ day, P5_ day, P6_ day ].

At this point, in the training phase, a total of three time granularity detection models are generated, namely a time slot level detection model, a minute level detection model and a day level detection model.

It should be noted here that steps one to seven, nine, and eleven in the creation process of the time-granularity detection model are performed by the data preprocessing unit in the training unit, and steps eight, ten, and twelve are performed by the neural network training unit in the training unit.

Fig. 6 is a schematic diagram of interference detection according to an embodiment of the present application. Exemplarily, in the embodiment, the time granularity detection model created as shown in fig. 5 is used to detect the interference type of the data to be detected, that is, three time granularity detection models are used to detect the interference type of the data to be detected. The process of interference detection may be performed by the detection unit in the above embodiments. As shown in fig. 6, the interference detection process includes the following steps:

the method comprises the following steps: the data preprocessing unit extracts a feature value consistent with the extracted feature value of the training set from the original data to obtain a character' _ slot (i.e., the data to be detected in the above embodiment).

Step two: the model selecting unit calls a slot level detection model (AI _ slot), the pilot 'slot to be detected enters the slot level detection model, and the interference detection probabilities P1', P2 ', and P6' slot (i.e., the first interference detection probability in the above embodiment) are output.

Step three: the preset time slot level threshold Thr is configured in the judgment unit _slot Judging whether the detection probability in the step two exceeds a preset time slot level threshold Thr or not _slot If so, the corresponding interference is considered to appear, the current time slot data is marked as the corresponding interference, and the judgment result is set to be 1 and is output to the result collection unit; if not, the current time slot is considered to be not detected with interference, the data of the current time slot is not marked with any interference, the data are divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Illustratively, taking frame out-of-sync interference in this embodiment as an example, such interference may be significantly characterized within one time slot,it can be successfully detected under the slot-level model.

Step four: the data preprocessing unit merges the unknown branch data generated in the third step with the Pn '_ slot output in the second step, and performs filtering, and obtains a filter' minimum (i.e., the first filtering result in the above embodiment) after 1 minute of filtering time.

Step five: the model selecting unit calls the minute-level detection model AI _ minu (as the current time granularity detection model), the data char 'actor to be detected enters the minute-level model, and the interference detection probabilities P1' _ minu, P2 '_ minu, P6' _ minu (i.e., the second interference detection probability in the above embodiment) are output.

Step six: the preset minute-level threshold Thr is configured in the judgment unit _minu Judging whether the detection probability in the step five exceeds a preset minute-level threshold Thr or not _minu If so, the corresponding interference is considered to appear, the data in the current minute is marked as the corresponding interference, and the judgment result is set to be 1 and is output to the result collection unit; if not, the interference is not detected in the current minute, the data in the current minute is not marked as any interference, the data are divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Illustratively, the narrowband interference in this embodiment is taken as an example, and the narrowband interference may exhibit a distinct characteristic within one minute, but may not be detected at the time slot level, but may be successfully detected under the minute level model.

Step seven: and the data preprocessing unit combines the data of the unknown branch generated in the step six with the Pn' _ min u output in the step five, and respectively performs filtering, and a filtering result character _dayis obtained after the filtering time is 1 day.

Step eight: the model selection unit calls an antenna-level detection model AI _ day, the data to be detected character _ day enters the antenna-level model, and interference detection probabilities P1 ' _ day, P2 ' _ day, P6 ' _ day are output.

Step nine: the preset antenna threshold Thr is configured in the judgment unit _day Judging whether the detection probability in the step eight exceeds a preset day threshold Thr or not _day If there isIf the interference is judged to occur, marking the current day-level data as the interference corresponding to the current day-level data, setting the judgment result as 1 and outputting the judgment result to a result collection unit; if not, the current day level is considered to be not detected with interference, the current day level data is not marked with any interference, the current day level data is divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Taking the atmospheric waveguide interference in the present embodiment as an example, the interference will be obviously characterized in one day, and although it cannot be detected in the time slot level and the minute level, it can be successfully detected in the antenna level model.

Step ten: and the result collecting unit gives the final interference type of the data to be detected according to the judgment result of the three-level model.

It should be noted that in this embodiment, when the first time granularity detection model (i.e., the timeslot level detection model) does not detect the interference type of the data to be detected, the current time granularity detection model (i.e., the minute level detection model) is used to continuously detect the interference type of the data to be detected, and if the interference type is not detected yet, the current time granularity detection model is switched to the time granularity detection model with a coarser time granularity (i.e., the day level detection model), and a combination of the data to be detected and the first interference detection probability is returned to be filtered, so as to obtain a new first filtering result, and the interference type of the data to be detected is continuously detected until the interference type of the data to be detected is detected. Of course, in the actual operation process, when the interference type of the data to be detected is not detected by using the day-level detection model, the interference detection of the data to be detected can be continued by using the time granularity detection model with the time granularity coarser than that of the day-level detection model.

According to the technical scheme of the embodiment, the multi-stage time granularity detection model is generated by adapting to data with different time granularities in the detection model training stage. Then, a detection model with multi-stage time granularity is used to adapt to the interference of different time granularities, so that the interference type identification capability is improved, and the interference type identification accuracy is further improved.

In one embodiment, the first time-granular characteristic value comprises: the 15-minute granularity frequency domain NI is taken as an example to describe the interference detection process in the 5G NR uplink system. Table 3 is another schematic table of an initial training data structure provided in the embodiment of the present application. Table 4 is a schematic table of interference types in another 5G NR uplink system provided in this embodiment. As shown in table 3, one measurement combination that can be obtained by the 5G NR uplink system is: [15 minute granularity frequency domain NI ], and as shown in Table 4, the interference types for the 5G NR uplink system may include: narrow-band interference, D4D5 interference, full-band interference, and zero-frequency interference.

TABLE 3 schematic diagram of another initial training data structure

TABLE 4 schematic representation of interference types in another 5G NR uplink system

Interference sequence number	Interference class
		1	Narrow-band interference
2	D4D5 interference
		3	Full frequency band interference
4	Zero frequency interference

Before the interference detection is carried out on the data to be detected by adopting the detection unit, a time granularity detection model is established by utilizing the training unit. Fig. 7 is a schematic diagram of creation of another time granularity detection model provided in an embodiment of the present application. In this embodiment, three time granularity detection models are created as an example, and a first time granularity detection model, a second time granularity detection model, and a third time granularity detection model are sequentially: the minute-level detection model, the day-level detection model and the week-level detection model are taken as examples to explain the creation process of the time granularity detection model. As shown in fig. 7, the creation process of the temporal granularity detection model includes the following steps:

the method comprises the following steps: obtaining the NI value NI of 273 RB in 15 minutes _rb (also referred to as 15-minute granularity frequency domain NI), where rb is 0, 1.

Step two: calculating a correlation coefficient Coef between adjacent RBs _rb1,rb2 (i.e., the frequency-domain correlation values in the above embodiments), where rb1 is 0,1,. 271, rb2 is rb1+1, and finally 272 frequency-domain features Cha are obtained _freq 。

Step three: the characteristics to be finally extracted are [15 minute granularity frequency domain NI, Cha _freq ]And is denoted as character _15 m.

Step four: sending the character _15m into a pre-created AI model for training, obtaining a satisfactory training effect (whether the interference detection accuracy reaches a first preset detection rate threshold value can be judged, and when the interference detection accuracy reaches the first preset detection rate threshold value, the satisfactory training effect is considered to be obtained), outputting two results, namely a 15-minute training model (AI _15m, namely the first AI training model in the embodiment), and storing the 15-minute training model AI _15 m; secondly, the preset interference probability distribution value (Pn _15m, i.e. the first time granularity interference probability distribution value in the above embodiment) of 15 minutes, where Pn _15m is [ P1_15m, P2_15m, P3_15m, P4_15m ].

Step five: and a preprocessing unit of the training unit merges the character _15m output in the step three and the Pn _15m output in the step four into the train _15m, wherein the train _15m is [ NI, P1_15m, P2_15m, P3_15m and P4_15m ], and filters the train _15m to obtain the antenna-level training data which is recorded as the character _ day.

The filtering mode of traindata _15m is as follows:

traindata_15m′ _t ＝0.9*traindata_15m′ _t-1 +0.1*traindata_15m _t wherein, traindata _15 m' _t Represents the training data, train data _15m ', obtained after current 15-minute filtering' _t-1 Represents the previous 15 minutes of filtered training data, train data _15m _t Representing the current 15 minutes of instantaneous training data.

Step six: sending the character _ day into a pre-created AI model for training to obtain a satisfactory training effect (whether the interference detection accuracy reaches a second preset detection rate threshold value can be judged, and when the interference detection accuracy reaches the second preset detection rate threshold value, the satisfactory training effect is considered to be obtained), and then outputting two results, namely a day-level training model (AI _ day, namely the second AI training model in the above embodiment), and storing the day-level training model; secondly, a preset interference probability distribution value (Pn _ day, the second time granularity interference probability distribution value in the above embodiment) of this day, where Pn _ day is [ P1_ day, P2_ day, P3_ day, P4_ day ].

Step seven: and the preprocessing unit of the training unit combines the character _ day output in the step five and the Pn _ day output in the step six into the traindata _ day, wherein the traindata _ day is [ NI, P1_ day, P2_ day, P3_ day, P4_ day ], and the traindata _ day is subjected to time-domain filtering for 1 week to obtain week-level training data which is recorded as character _ week.

The filtering mode of the traindata _ day is as follows:

traindata_day′ _t ＝0.9*traindata_day′ _t-1 +0.1*traindata_day _t wherein, traindata _ day' _t Represents the training data, train _ day ', obtained after current 1-day filtering' _t-1 Represents the training data after the first 1 day filtering, train _ day _t Representing the instantaneous training data of the current day.

Sending the character _ week into an AI model for training to obtain a satisfactory training effect (whether the interference detection accuracy reaches a third preset detection rate threshold value can be judged, and when the interference detection accuracy reaches the third preset detection rate threshold value, the satisfactory training effect is considered to be obtained), outputting two results, namely, a week-level training model (AI _ week, namely the third AI training model in the embodiment), and storing the week-level training model; second, the preset interference probability distribution value of week (Pn _ week, i.e., the third time-granularity interference probability distribution value in the above embodiment) is set, where Pn _ week is [ P1_ week, P2_ week, P3_ week, P4_ week ].

So far, in the training stage, a total of three time granularity detection models are generated, namely a 15-minute detection model, a daily detection model and a weekly detection model.

It should be noted here that steps one to three, step five, and step seven in the creation process of the time-granularity detection model shown in fig. 7 are executed by the data preprocessing unit in the training unit; and step four, step six and step eight are executed by a neural network training unit in the training unit.

Fig. 8 is a schematic diagram of another interference detection provided in the embodiment of the present application. In an embodiment, the time granularity detection model created as shown in fig. 7 is used to detect the interference type of the data to be detected. The process of interference detection may be performed by the detection unit in the above embodiments. Illustratively, the interference detection process is described by taking the first time granularity detection model as a 15-minute level detection model, the current time granularity detection model as a day level detection model, and the week level detection model as examples. As shown in fig. 8, the interference detection process includes the following steps:

the method comprises the following steps: the data preprocessing unit extracts feature values of the original data, which are consistent with the extracted feature values of the training set, to obtain character' _15m (i.e., the data to be detected in the above embodiment).

Step two: the model selecting unit calls a pre-created 15-minute level detection model, the data to be detected character '_ 15m enters the 15-minute level detection model, and interference detection probabilities P1', P2 ', P4' (i.e., the first interference detection probability in the above embodiment) are output.

Step three: the preset 15-minute threshold Thr is configured in the judgment unit _15m Judging whether the detection probability in the step two isHas Thr exceeding the preset 15-minute threshold _15m If so, the corresponding interference is considered to appear, the current 15-minute data is marked as the corresponding interference, and the judgment result is set to be 1 and is output to the result collection unit; if not, the interference is not detected in the current 15 minutes, the data in the current 15 minutes are not marked as any interference, the data are divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. By taking the narrowband interference in the implementation as an example, the narrowband interference can present a remarkable characteristic within 15 minutes, and thus can be successfully detected under a 15-minute detection model.

Step four: the data preprocessing unit combines the data of the unknown branch generated in the third step with Pn' output from the second step, and then performs filtering, and obtains a filter result character _ day (i.e. the first filtering result in the above embodiment) after 1 day of filtering time.

Step five: the model selection unit calls a pre-created day-level detection model (as the current time granularity detection model), the data to be detected character _dayenters the day-level detection model, and interference detection probabilities P1 ', P2 ', P4 ' (i.e., the second interference detection probability in the above embodiment) are output.

Step six: the preset antenna threshold Thr is configured in the judgment unit _day Judging whether the detection probability in the step five exceeds an antenna threshold Thr or not _day If so, the corresponding interference is considered to appear, the current day data is marked as the corresponding interference, and the judgment result is set as 1 and output to the result collection unit; if not, the current day is considered to be not detected with interference, the data of the current day is not marked as any interference, the data are divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Taking the full bandwidth interference in this implementation as an example, the interference will present obvious characteristics in one day, so that the interference can be successfully detected under an antenna-level detection model.

Step seven: and the data preprocessing unit combines the data of the unknown branch generated in the step six with the Pn 'output in the step five, and then filters the data, wherein the filtering result, namely the character' _ week, is obtained after the filtering time is 1 week.

Step eight: the model selection unit calls a weekly detection model, the data to be detected character _weekenters the weekly detection model, and interference detection probabilities P1 ', P2 ', P4 ' are output.

Step nine: the preset cycle threshold Thr is configured in the judgment unit _week Judging whether the detection probability in the step eight exceeds a cycle threshold Thr or not _week If so, considering that corresponding interference occurs, marking the current cycle data as the corresponding interference, setting a judgment result as 1 and outputting the judgment result to a result collection unit; if not, the current cycle detection model is considered to not detect the interference, the current cycle data is not marked as any interference, the current cycle data is divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Taking the D4D5 interference in this embodiment as an example, the interference will show obvious features in one week, so it can be successfully detected under the week-level detection model.

It should be noted that in this embodiment, when the first time granularity detection model (i.e., the 15-minute-level detection model) does not detect the interference type of the data to be detected, the current time granularity detection model (i.e., the day-level detection model) is used to continuously detect the interference type of the data to be detected, and if the interference type is not detected yet, the current time granularity detection model is switched to the time granularity detection model with a coarser time granularity (i.e., the week-level detection model), and a combination of the data to be detected and the first interference detection probability is returned to be filtered, so as to obtain a new first filtering result, and the interference type of the data to be detected is continuously detected until the interference type of the data to be detected is detected. Of course, in the actual operation process, when the interference type of the data to be detected is not detected by using the periodic detection model, the interference detection of the data to be detected can be continued by using the time granularity detection model with the coarser time granularity than that of the periodic detection model.

In one embodiment, the detection unit may be used independently if a trained detection model and a training set feature value extraction approach have been obtained. Table 5 is an obtained interference characteristic schematic table provided in this embodiment, and as shown in table 5, it is assumed that the characteristic extraction manner of the training set is [ time domain RSSI, frequency domain NI ]. Table 6 is a schematic interference type table provided in the embodiment of the present application. As shown in table 6, the interference types include: frame out-of-step interference, narrow band interference and full band interference. The trained detection models are a time slot level detection model AI _ slot, a 15 minute level detection model AI _15m and an hour level detection model AI __ hour respectively.

TABLE 5A schematic representation of interference signatures obtained

Time domain raw data	Frequency domain raw data
		RSSI0,RSSI1,...RSSI13	NI0,NI1,...,NI272

TABLE 6 interference type schematic diagram

Interference sequence number	Interference class
		1	Frame out-of-sync interference
2	Narrow-band interference
		3	Full band interference

Fig. 9 is a schematic diagram of another interference detection provided in the embodiment of the present application. The process of interference detection may be performed by the detection unit in the above embodiments. Illustratively, the interference detection process is described by taking the first time granularity detection model as a time slot level detection model, the current time granularity detection model as a 15-minute level detection model, and the hour level detection model as an example. As shown in fig. 9, the interference detection process includes the following steps:

Step two: the model selecting unit calls a slot level detection model AI _ slot, the data to be detected character _ slot enters the slot level model, and the interference detection probabilities P1 ' _ slot, P2 ' _ slot, and P3 ' _ slot (i.e. the first interference detection probability in the above embodiment) are output.

Step three: the preset time slot level threshold in the decision unit is Thr _slot Judging whether the detection probability in the step two exceeds a time slot level threshold Thr or not _slot If so, the corresponding interference is considered to appear, the current time slot data is marked as the corresponding interference, and the judgment result is set to be 1 and is output to the result collection unit; if not, the current time slot is considered to be not detected with interference, the data of the current time slot is not marked as any interference, the data is divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Book withFrame-out-of-sync interference in the implementation is an example, and the interference is obviously characterized in one time slot, so that the interference can be successfully detected under a time slot level model.

Step four: the data preprocessing unit merges the data of the unknown branch generated in step three with the Pn '_ slot output in step two, performs filtering, and obtains a filter result character' _15m (i.e., the first filtering result in the above embodiment) after 15 minutes of filtering time.

Step five: the model selecting unit calls the 15-minute level detection model AI _15m (as the current time granularity detection model), the data to be detected character '_ 15m enters the 15-minute level detection model, and the interference detection probabilities P1' _15m, P2 '_ 15m, and P3' _15m (i.e., the second interference detection probability in the above embodiment) are output.

Step six: the preset 15-minute threshold in the decision unit is Thr _15m Judging whether the detection probability in the step five exceeds a 15-minute threshold Thr _15m If so, the corresponding interference is considered to appear, the current 15-minute data is marked as the corresponding interference, and the judgment result is set to be 1 and is output to the result collection unit; if not, the interference is not detected in the current 15 minutes, the data in the current 15 minutes are not marked as any interference, the data are divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Taking the narrowband interference in the present embodiment as an example, the narrowband interference may exhibit a distinct characteristic within 15 minutes, and although the narrowband interference cannot be detected at the timeslot level, the narrowband interference may be successfully detected under a 15-minute level detection model.

Step seven: and the data preprocessing unit combines the data of the unknown branch generated in the step six with the Pn '_ 15m output in the step five, and respectively performs filtering, and after the filtering time is 1 hour, a filtering result character' _ hour is obtained.

Step eight: the model selection unit calls an hour-level detection model AI _ hour, the data to be detected character 'hour enters the hour-level detection model, and interference detection probabilities P1', P2 ', P3' hour are output.

Step nine: the preset hour-level threshold in the decision unit is Thr _hour Judgment ofWhether the detection probability in the step eight exceeds a small time threshold Thr or not _hour If so, the corresponding interference is considered to appear, the current hour-level data is marked as the corresponding interference, and the judgment result is set to be 1 and is output to the result collection unit; if not, the interference is not detected in the current hour, the current hour-level data is not marked as any interference, the current hour-level data is divided into unknown branches, and the judgment result is set to be 0 and is output to the result collection unit. Taking the full-band interference in this embodiment as an example, the interference will be obviously characterized within one hour, and although it cannot be detected at the time slot level and 15 minutes level, it can be successfully detected under the hour level detection model.

It should be noted that in this embodiment, when the first time granularity detection model (i.e., the timeslot level detection model) does not detect the interference type of the data to be detected, the current time granularity detection model (i.e., the 15-minute level detection model) is used to continuously detect the interference type of the data to be detected, and if the interference type is not detected yet, the current time granularity detection model is switched to the time granularity detection model (i.e., the hour level detection model) with a coarser time granularity, and a combination of the data to be detected and the first interference detection probability is returned to be filtered, so as to obtain a new first filtering result, and the interference type of the data to be detected is continuously detected until the interference type of the data to be detected is detected. Of course, in the actual operation process, when the small-scale detection model is used to not detect the interference type of the data to be detected, the time granularity detection model with coarser time granularity than that of the small-scale detection model can be used to continue the interference detection of the data to be detected.

According to the technical scheme, the detection model of the multi-stage time granularity is utilized, the interference of different time granularities is adapted, the interference type recognition capability is improved, and the interference type recognition accuracy is further improved.

In the embodiment, the multi-stage time granularity detection model is adopted to adapt to the interference of different time granularities, and the interference type identification capability is greatly improved. When the interference presents short-term time granularity characteristics, such as time slot level characteristics, symbol level characteristics and the like, the detection model can successfully detect the interference in the initial detection stage and give a detection result; when the interference presents medium and long time granularity characteristics, such as minute-level characteristics, hour-level characteristics, day-level characteristics and the like, the detection model can also successfully detect the interference in the medium and later detection periods and give a detection result. When the data to be detected only contains one dimension of characteristics, the multistage time particle size detection model can also carry out interference detection, so that interference can be effectively avoided and eliminated.

In an embodiment, fig. 10 is a block diagram of another interference detection apparatus provided in the embodiment of the present application. The embodiment is applied to the interference detection device. As shown in fig. 10, the present embodiment includes: a first filter 310, a first determination module 320, and a second determination module 330.

The first filter 310 is configured to filter a combination of data to be detected and a predetermined first interference detection probability to obtain a first filtering result, when the interference type of the data to be detected is not detected by using a pre-created first time granularity detection model;

a first determining module 320 configured to determine, as a second interference detection probability, an interference detection probability of the first filtering result in a pre-created current time granularity detection model, wherein the current time granularity detection model is a detection model with a coarser time granularity than that of the first time granularity detection model;

the second determining module 330 is configured to determine the interference type of the data to be detected according to a comparison result between the second interference detection probability and the first preset detection probability threshold.

In an embodiment, when the second interference detection probability is smaller than the first preset detection probability threshold, the method further includes:

and switching the current time granularity detection model to a time granularity detection model with coarser time granularity, and returning to the step of filtering the combination of the data to be detected and the predetermined first interference detection probability to obtain a filtering result until the interference type of the data to be detected is detected.

In an embodiment, the filtering a combination of data to be detected and a predetermined first interference detection probability to obtain a first filtering result includes:

combining the data to be detected and a predetermined first interference detection probability to obtain combined data;

and filtering the time length of the current time granularity of the combined data to obtain a first filtering result.

In an embodiment, the filtering the combined data for the time length of the current time granularity to obtain a first filtering result includes:

and filtering the combined data according to the instantaneous combined data of the current time granularity, the combined data after the previous current time granularity is filtered, a first preset weight coefficient and a second preset weight coefficient to obtain a first filtering result.

In one embodiment, the interference detection apparatus further includes:

the first creation module is configured to input a predetermined first time granularity characteristic value into a first pre-created AI training model before the interference type of the data to be detected is not detected by using a pre-created first time granularity detection model until the detection rate reaches a first preset detection rate threshold value, output the first AI training model and a first time granularity interference probability distribution value, and use the first AI training model as a first time granularity detection model;

a first combiner configured to combine the first time-granularity characteristic value and the first time-granularity interference probability distribution value as a second time-granularity characteristic value;

the second filter is configured to perform time-domain filtering on the second time granularity characteristic value to obtain second time granularity training data;

the second creating module is configured to input second time granularity training data into a second pre-created AI training model until the detection rate reaches a second preset detection rate threshold, output the second AI training model and a second time granularity interference probability distribution value, and use the second AI training model as a second time granularity detection model; and the time granularity corresponding to the first time granularity detection model and the second time granularity detection model is successively coarser.

In an embodiment, the interference detection apparatus further includes:

a second combiner configured to combine the second time-granularity-level training data and the second time-granularity interference probability distribution value as a third time-granularity characteristic value after the second AI training model is used as a second time-granularity detection model;

the third filter is configured to perform time-domain filtering on the third time granularity characteristic value to obtain third time granularity training data;

the third creating module is configured to input third time granularity training data into a third pre-created AI training model until the detection rate reaches a third preset detection rate threshold, output the third AI training model and a third time granularity interference probability distribution value, and use the third AI training model as a third time granularity detection model; and the time granularity corresponding to the first time granularity detection model, the second time granularity detection model and the third time granularity detection model is successively thickened.

In an embodiment, the first time-granular characteristic value comprises one of: time domain Received Signal Strength Indication (RSSI); frequency domain noise indication NI; performing spatial domain matched filtering; a time domain correlation value; a frequency domain correlation value; middle resource block RB energy distribution.

In an embodiment, the first time granularity detection model and the current time granularity detection model each include one of: a time slot level detection model; a minute-scale detection model; a day-level detection model; and (4) a weekly detection model.

The interference detection apparatus provided in this embodiment is configured to implement the interference detection method in the embodiment shown in fig. 1, and the implementation principle and the technical effect of the interference detection apparatus provided in this embodiment are similar, which are not described herein again.

Fig. 11 is a schematic structural diagram of an interference detection device according to an embodiment of the present application. As shown in fig. 10, the present application provides an apparatus comprising: a processor 410, a memory 420, and a communication module 430. The number of the processors 410 in the device may be one or more, and one processor 410 is taken as an example in fig. 10. The number of the memory 420 in the device may be one or more, and one memory 420 is taken as an example in fig. 10. The processor 410, memory 420 and communication module 430 of the device may be connected by a bus or other means, as exemplified by the bus connection in fig. 10. In this embodiment, the device may be a terminal side (e.g., a user equipment).

The memory 420, which is a computer-readable storage medium, may be configured to store software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the apparatus of any embodiment of the present application (e.g., the first filter 310, the first determining module 320, and the second determining module 330 in the interference detection device). The memory 420 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the device, and the like. Further, the memory 420 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 420 may further include memory located remotely from processor 410, which may be connected to devices through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

A communication module 430 configured for communication interaction between the respective communication nodes.

The interference detection device provided above may be configured to perform the interference detection method provided in any of the embodiments above, and has corresponding functions and effects.

Embodiments of the present application also provide a storage medium containing computer-executable instructions that when executed by a computer processor are configured to perform a method of interference detection, the method comprising: under the condition that the interference type of the data to be detected is not detected by adopting a pre-established first time granularity detection model, filtering the combination of the data to be detected and a predetermined first interference detection probability to obtain a first filtering result; determining the interference detection probability of the first filtering result in a pre-established current time granularity detection model as a second interference detection probability, wherein the current time granularity detection model is a detection model with coarser time granularity than that of the first time granularity detection model; and determining the interference type of the data to be detected according to the comparison result of the second interference detection probability and a first preset detection probability threshold value.

It will be clear to a person skilled in the art that the term user equipment covers any suitable type of wireless user equipment, such as mobile phones, portable data processing devices, portable web browsers or vehicle-mounted mobile stations.

In general, the various embodiments of the application may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. For example, some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device, although the application is not limited thereto.

Embodiments of the application may be implemented by a data processor of a mobile device executing computer program instructions, for example in a processor entity, or by hardware, or by a combination of software and hardware. The computer program instructions may be assembly instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or target code written in any combination of one or more programming languages.

Any logic flow block diagrams in the figures of this application may represent program steps, or may represent interconnected logic circuits, modules, and functions, or may represent a combination of program steps and logic circuits, modules, and functions. The computer program may be stored on a memory. The Memory may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as, but not limited to, Read-Only Memory (ROM), Random Access Memory (RAM), optical storage devices and systems (Digital Video Disc (DVD) or Compact Disc (CD)), etc. The computer readable medium may include a non-transitory storage medium. The data processor may be of any type suitable to the local technical environment, such as but not limited to general purpose computers, special purpose computers, microprocessors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Programmable logic devices (FGPAs), and processors based on a multi-core processor architecture.

The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An interference detection method, comprising:

determining an interference detection probability of the first filtering result in a pre-established current time granularity detection model as a second interference detection probability, wherein the current time granularity detection model is a detection model with coarser time granularity than that of the first time granularity detection model;

2. The method of claim 1, wherein if the second interference detection probability is smaller than a first preset detection probability threshold, the method further comprises:

and switching the current time granularity detection model to a time granularity detection model with coarser time granularity, and returning to the step of filtering the combination of the data to be detected and the predetermined first interference detection probability to obtain a first filtering result until the interference type of the data to be detected is detected.

3. The method according to claim 1, wherein the filtering the combination of the data to be detected and the predetermined first interference detection probability to obtain a first filtering result comprises:

4. The method of claim 3, wherein the filtering the combined data for a time duration of a current time granularity to obtain a first filtering result comprises:

5. The method according to claim 1, before the detecting, with the pre-created first time-granular detection model, the interference type of the data to be detected, further comprising:

merging the first time granularity characteristic value and the first time granularity interference probability distribution value to serve as a second time granularity characteristic value;

performing time domain filtering on the second time granularity characteristic value to obtain second time granularity level training data;

inputting the second time granularity level training data into a second pre-established AI training model until the detection rate reaches a second preset detection rate threshold, outputting the second AI training model and a second time granularity interference probability distribution value, and taking the second AI training model as a second time granularity detection model; and the time granularity corresponding to the first time granularity detection model and the second time granularity detection model is successively thickened.

6. The method of claim 5, further comprising:

inputting the third time granularity level training data into a third pre-established AI training model until the detection rate reaches a third preset detection rate threshold, outputting the third AI training model and a third time granularity interference probability distribution value, and taking the third AI training model as a third time granularity detection model; and the time granularity corresponding to the first time granularity detection model, the second time granularity detection model and the third time granularity detection model is successively thickened.

7. The method of claim 5 or 6, wherein the first time-granularity feature value comprises one of: time domain Received Signal Strength Indication (RSSI); frequency domain noise indication NI; performing spatial domain matched filtering; a time domain correlation value; a frequency domain correlation value; middle resource block RB energy distribution.

8. The method of any of claims 1-4, wherein the first time-granularity detection model and the current time-granularity detection model each comprise one of: a time slot level detection model; a minute-scale detection model; a day-level detection model; and (4) a week-level detection model.

9. An interference detection device, comprising:

the first filter is configured to filter a combination of the data to be detected and a predetermined first interference detection probability to obtain a first filtering result under the condition that the interference type of the data to be detected is not detected by adopting a pre-established first time granularity detection model;

10. An interference detection device, comprising: a communication module, a memory, and one or more processors;

the communication module is configured to perform communication interaction among the communication nodes;

the memory configured to store one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-8 above.

11. A storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method according to any one of claims 1-8.