CN115378856B - Communication detection method, device and storage medium - Google Patents

Communication detection method, device and storage medium Download PDF

Info

Publication number
CN115378856B
CN115378856B CN202210973887.7A CN202210973887A CN115378856B CN 115378856 B CN115378856 B CN 115378856B CN 202210973887 A CN202210973887 A CN 202210973887A CN 115378856 B CN115378856 B CN 115378856B
Authority
CN
China
Prior art keywords
communication
vector
loss function
data
detected
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210973887.7A
Other languages
Chinese (zh)
Other versions
CN115378856A (en
Inventor
吴嘉澍
王洋
须成忠
叶可江
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Institute of Advanced Technology of CAS
Original Assignee
Shenzhen Institute of Advanced Technology of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Institute of Advanced Technology of CAS filed Critical Shenzhen Institute of Advanced Technology of CAS
Priority to CN202210973887.7A priority Critical patent/CN115378856B/en
Publication of CN115378856A publication Critical patent/CN115378856A/en
Application granted granted Critical
Publication of CN115378856B publication Critical patent/CN115378856B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/50—Testing arrangements
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00—Economic sectors
    • G16Y10/75—Information technology; Communication
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/10—Detection; Monitoring
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/12—Network monitoring probes
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Abstract

The invention discloses a communication detection method, equipment and a storage medium, wherein the method is characterized in that an information source for assisting a target to be detected in communication detection is selected from network equipment by utilizing the similarity among data according to a communication detection task, the network equipment comprises the Internet of things and equipment in a network center of the Internet of things, communication data of the target to be detected and communication data of the information source are input into a communication detection model for communication detection, a communication detection result of the target to be detected aiming at a communication category is obtained, and the communication detection model is trained based on a first loss function representing distribution information and importance of each information source and a second loss function representing spatial gathering of the target to be detected and the information source. The invention realizes the complementation of the information between the information source and the target to be detected by using the information carried by the information source as an aid so as to improve the detection accuracy of the communication data of the target to be detected.

Description

Communication detection method, device and storage medium
Technical Field
The present disclosure relates to the field of internet of things, and in particular, to a communication detection method, device, apparatus, and storage medium.
Background
Along with the progress of internet of things, at present, more and more internet of things equipment is used in people's daily production life, and these abundant internet of things equipment also make many important fields to be transformed and break through, for example wisdom road networking and internet of things equipment have supported smart city and intelligent transportation, and wisdom medical treatment internet of things equipment has supported more accurate and humanized wisdom medical treatment. Because the computing power of the internet of things equipment is relatively weak, the use of the internet of things equipment often requires the cooperation of a network center, and the network center can bear the tasks of large-scale data computation, storage and the like which cannot be easily completed by the internet of things equipment.
The internet of things and the network center thereof all relate to a large amount of network communication, so that detection of the internet of things and the network center communication thereof is necessary. Communication detection for the Internet of things and the network center thereof can be various. If the communication content is detected, the method can be used for content statistics, service condition analysis and the like of the communication of the Internet of things and the network center thereof, and if the method is used for detecting the communication safety, the method can be used for analyzing which communication of the Internet of things and the network center thereof is normal communication and which communication is abnormal communication, so that the safety of the Internet of things equipment and the network center thereof is ensured. Therefore, an accurate and efficient communication detection method for the Internet of things and the network center thereof is necessary, and the method can be used for analyzing the use modes of the Internet of things and the network center thereof and for monitoring and monitoring the safety of the Internet of things and the network center thereof, so that the Internet of things and the network center thereof are ensured to operate in an efficient and reliable mode.
However, the communication faced by the internet of things device and the internet of things device, and the internet of things and its network center in operation, is different. For example, more communication performed by the internet of things device is data transmission, and more communication performed by the internet of things center may be biased to calculate the related communication with the storage. Therefore, communications faced by different devices of the internet of things and communications faced by the network center are all long, but not very comprehensive. For example, if communication data captured by a party to be detected by communication is not abundant enough, a deviation between a detected result and an actual result is larger, which affects the effect of communication detection.
Disclosure of Invention
In view of this, the present application provides a communication detection method, device, apparatus and storage medium, so as to solve the problem of poor communication detection effect of the existing network devices such as the internet of things and the network center thereof.
In order to solve the technical problems, one technical scheme adopted by the application is as follows: provided is a communication detection method, comprising: receiving a communication detection task, wherein the communication detection task comprises a target to be detected, a communication category and network equipment where the target to be detected is located; screening information sources from other targets based on the similarity between the first communication data corresponding to the communication category of the target to be detected and the second communication data corresponding to the communication category of other targets in the network equipment; inputting the first communication data and the third communication data of the information sources into a communication detection model for communication detection to obtain a communication detection result of the target to be detected aiming at the communication category, wherein the communication detection model comprises a feature extraction network and a global public detector, and the feature extraction network and the global public detector are trained and obtained based on a first loss function representing the distribution information and the importance of each information source and a second loss function representing the spatial gathering of the target to be detected and the information sources of the equipment to be detected.
As a further improvement of the present application, screening information sources from other targets based on a similarity between first communication data corresponding to a communication category of a target to be tested and second communication data corresponding to the communication category of other targets in a network device, includes: acquiring first communication data of a target to be detected and second communication data of other targets, extracting the first communication data to obtain a first communication vector by utilizing a characteristic extraction network, and extracting the second communication vector from the second communication data; obtaining statistics according to first data distribution of the first communication data in each communication category and second data distribution of the second communication data in each communication category; sorting all other targets according to the statistics to obtain distribution sorting; calculating Euclidean distance between a first average value vector of the first communication vector in each communication category and a second average value vector of each second communication vector in each communication category, and taking an average value to obtain an average Euclidean distance; sequencing all other targets according to the average Euclidean distance to obtain Euclidean sequencing; and confirming the final ordering of all other targets according to the distribution ordering and the European ordering, and selecting a first preset number of other targets as information sources according to the final ordering.
As a further improvement of the present application, the training communication detection model specifically includes: respectively inputting first communication sample data of a target to be detected and second communication sample data of an information source into a feature extraction network to extract, so as to obtain a first communication sample vector and a second communication sample vector; calculating the weight of each information source by using the first communication sample vector and the second communication sample vector; respectively inputting the first communication sample vector and the second communication sample vector into a global public detector to perform distribution prediction to obtain a first prediction distribution vector of a target to be detected in each communication category and a second prediction distribution vector of an information source in each communication category; calculating a loss function value according to the first prediction distribution vector, the second prediction distribution vector, the weight, the first communication sample vector, the second communication sample vector, the first loss function and the second loss function; and carrying out iterative training on the feature extraction network and the global public detector according to the loss function value and a preset optimization algorithm.
As a further improvement of the present application, the calculating the weight of each information source by using the first communication sample vector and the second communication sample vector includes: calculating Euclidean distance between a first average value vector of the first communication sample vector in each communication category and a second average value vector of each second communication sample vector in each communication category, and taking an average value to obtain an average Euclidean distance; and calculating the weight of the information source according to the average Euclidean distance corresponding to each information source, wherein the calculation formula of the weight is expressed as follows:
Figure BDA0003797961140000041
Wherein omega j Representing the weight of the jth information source, dist j And representing the average Euclidean distance between the jth information source and the target to be detected.
As a further refinement of the present application, calculating a first loss function value for the first loss function includes: calculating a first prediction distribution mean vector according to the first prediction distribution vector; calculating KL divergence between the information source and the target to be detected according to the first prediction distribution mean value vector and the second prediction distribution mean value vector; and calculating a first loss function value according to the first prediction distribution mean vector, the KL divergence and the weight.
As a further improvement of the present application, the calculation formula of the first prediction distribution mean vector is:
Figure BDA0003797961140000042
wherein (1)>
Figure BDA0003797961140000043
Represents a first predictive distribution mean vector, |x (k) I represents the number of communication data belonging to the kth class of communication category, C () represents the global common detector, f (x) represents the feature extraction network, +.>
Figure BDA0003797961140000044
Representing a first predicted distribution vector, T representing a temperature parameter;
the calculation formula of the KL divergence is:
Figure BDA0003797961140000045
Figure BDA0003797961140000046
wherein (1)>
Figure BDA0003797961140000047
Indicating KL divergence between jth information source and distribution information of target to be detected,/for>
Figure BDA0003797961140000048
Representing a second predicted distribution vector;
the calculation formula of the first loss function value is:
Figure BDA0003797961140000049
wherein LI is a first loss function value, k represents the number of communication categories, n represents the number of information sources, ω j Representing the weight of the jth information source, |χ D I represents the number of first communication sample data, L ce () Representing cross entropy loss, y i And a communication detection type label is indicated.
As a further refinement of the present application, calculating a second loss function value for the second loss function includes: combining the second communication sample data of all the information sources into third communication sample data, and combining the first communication sample data of the target to be detected and the second communication sample data of all the information sources into fourth communication sample data; extracting a third communication sample vector and a fourth communication sample vector from the third communication sample data and the fourth communication sample data respectively by utilizing a characteristic extraction network; respectively calculating a first communication sample vector, a third communication sample vector, a fourth communication sample vector, a first average value vector, a third average value vector and a fourth average value vector of each communication class; calculating Euclidean distances between the first mean value vector, the third mean value vector and the fourth mean value vector, and summing to obtain a first Euclidean distance loss function value; respectively selecting a second preset number of first characteristic points from the first communication sample vectors of all the communication categories by using a clustering algorithm, and selecting a second preset number of second characteristic points from the third communication sample vectors of all the communication categories; calculating Euclidean distances between the first characteristic points and the second characteristic points, and averaging to obtain a second Euclidean distance loss function value; and calculating a second loss function value according to the first Euclidean distance loss function value and the second Euclidean distance loss function value.
As a further improvement of the present application, the calculation formula of the first euclidean distance loss function value is:
Figure BDA0003797961140000051
wherein LG denotes a first Euclidean distance loss function value,
Figure BDA0003797961140000052
representing a first mean vector, ">
Figure BDA0003797961140000053
Representing a third mean vector, ">
Figure BDA0003797961140000054
Represents a fourth mean vector, k represents the number of communication categories,/->
Figure BDA0003797961140000055
Representing the Euclidean distance;
the calculation formula of the second Euclidean distance loss function value is as follows:
Figure BDA0003797961140000056
wherein LL represents a second euclidean distance loss function value, R represents a second preset number,
Figure BDA0003797961140000057
n-th, i.e. in the first feature points representing the selected R i-th communication classes in the first communication sample vector>
Figure BDA0003797961140000058
Representing the mth of the second feature points of the selected R ith communication categories in the third communication sample vector.
In order to solve the technical problem, another technical scheme adopted by the application is as follows: provided is a communication detection device including: the receiving module is used for receiving a communication detection task, wherein the communication detection task comprises a target to be detected, a communication category and network equipment where the target to be detected is located; the selecting module is used for screening information sources from other targets based on the similarity between the first communication data corresponding to the communication category of the target to be detected and the second communication data corresponding to the communication category of other targets in the network equipment; the prediction module is used for inputting the first communication data and the third communication data of the information sources into the communication detection model to carry out communication detection, so as to obtain a communication detection result of the target to be detected aiming at the communication category, the communication detection model comprises a feature extraction network and a global public detector, and the feature extraction network and the global public detector are obtained by training based on a first loss function representing the distribution information and the importance of each information source and a second loss function representing the spatial gathering of the target to be detected and the information source of the equipment to be detected.
In order to solve the technical problem, a further technical scheme adopted by the application is as follows: there is provided a computer device comprising a processor, a memory coupled to the processor, the memory having stored therein program instructions which, when executed by the processor, cause the processor to perform the steps of the communication detection method as described in any of the above.
In order to solve the technical problem, a further technical scheme adopted by the application is as follows: a storage medium is provided that stores program instructions capable of implementing any one of the communication detection methods described above.
The beneficial effects of this application are: according to the communication detection method, the information sources for assisting the to-be-detected targets to carry out communication detection are selected from the similarity network equipment among the data according to the communication detection task, the network equipment comprises the Internet of things and equipment in the network center of the Internet of things, communication data of the to-be-detected targets and communication data of the information sources are input into the communication detection model to carry out communication detection, communication detection results of the to-be-detected targets aiming at communication categories are obtained, the communication detection model is trained based on first loss functions representing distribution information of each information source and importance of the information sources and second loss functions representing the to-be-detected targets and the information sources on space gathering, and therefore when the communication data of the to-be-detected targets are detected, the information sources are used as assistance, information complementation between the information sources and the to-be-detected targets is carried out, more refined detection of the communication data of the to-be-detected targets is facilitated, detection effects of the to-be-detected targets are improved, and detection accuracy is improved.
Drawings
FIG. 1 is a flow chart of a communication detection method according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a functional module of a communication detection device according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a computer device according to an embodiment of the present invention;
fig. 4 is a schematic structural view of a storage medium according to an embodiment of the present invention.
Detailed Description
The following description of the technical solutions in the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The terms "first," "second," "third," and the like in this application are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first", "a second", and "a third" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise. All directional indications (such as up, down, left, right, front, back … …) in the embodiments of the present application are merely used to explain the relative positional relationship, movement, etc. between the components in a particular gesture (as shown in the drawings), and if the particular gesture changes, the directional indication changes accordingly. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The communication detection method of the embodiment is applied to network equipment in the internet of things and a network center thereof, wherein the network equipment comprises, but is not limited to, a computer, a mobile phone, a tablet and the like. Because the computing power of the network equipment in the internet of things is relatively weak, the use of the network equipment in the internet of things often requires the cooperation of a network center, and the network center can bear the tasks of large-scale data computation, storage and the like which cannot be easily completed by the network equipment of the internet of things. It should be understood that, the internet of things and network devices in the network center thereof can be complementary in advantages under different scenes, and are regarded as information sources with rich communication detection information or targets to be detected with insufficient communication detection information, and the targets to be detected are not limited to one device, but can be one device cluster with identical configuration, or be a network center of the internet of things, and similarly, the information sources can be one device, or one device cluster with identical configuration, or be a network center of the internet of things. The complementation in this embodiment refers to the network device in the internet of things or the network center of the internet of things possibly becoming an information source in some communication detection scenarios, i.e. a party providing assistance in the communication detection process, and possibly becoming a target to be detected in other communication detection scenarios, i.e. a party requiring assistance in the communication detection process. Therefore, an advantage complementary relationship is formed between the Internet of things network equipment and between the Internet of things network equipment and the network center. Based on the advantage complementary relation, the invention provides a communication detection method to realize more accurate communication detection of a target to be detected.
Fig. 1 is a flow chart of a communication detection method according to an embodiment of the invention. It should be noted that, if there are substantially the same results, the method of the present invention is not limited to the flow sequence shown in fig. 1. As shown in fig. 1, the communication detection method includes the steps of:
step S101: and receiving a communication detection task, wherein the communication detection task comprises a target to be detected, a communication category and network equipment where the target to be detected is located.
Specifically, it should be noted that, in this embodiment, the network device refers to a device in the internet of things and a network center thereof, where the communication detection task refers to detecting data on the device in the internet of things, and based on different communication detection tasks, the data on the network device may be divided into a plurality of communication categories, for example, when the communication detection task is intrusion data detection, the communication category is divided into communication intrusion categories, communication detection needs to be performed on intrusion data in the device, and when the communication detection task is access data detection, the communication category is divided into communication access categories, and communication detection needs to be performed on access data in the device.
Step S102: and screening information sources from other targets based on the similarity between the first communication data corresponding to the communication category of the target to be detected and the second communication data corresponding to the communication category of other targets in the network equipment.
Specifically, after receiving the communication detection task, in order to improve the detection effect of the target to be detected, an information source needs to be selected from the internet of things to assist in detecting the communication data of the target to be detected. And screening information sources through data similarity between communication data of the target to be detected and other targets in the Internet of things.
Further, step S102 specifically includes:
1. the method comprises the steps of obtaining first communication data of a target to be detected and second communication data of other targets, extracting the first communication data to obtain a first communication vector by utilizing a feature extraction network, and extracting the second communication vector from the second communication data.
Specifically, the data of the object to be detected is input to the feature extraction network, a first communication vector is extracted, all network devices except the object to be detected in the internet of things and the network center of the internet are taken as other targets, the communication data of each other target are obtained, the communication data are input to the feature extraction network, and a second communication vector is extracted.
2. And calculating statistics according to the first data distribution of the first communication data in each communication category and the second data distribution of the second communication data in each communication category.
It should be noted that, the communication data generated by the network device generally includes a plurality of communication categories, and each communication category includes a plurality of data. Specifically, after the first communication data is obtained, the number of the data of each communication category of the object to be measured is counted, so that the distribution of the communication data of the object to be measured can be obtained, for example, 100 communication categories A and 600 communication categories B in the communication data of the object to be measured, and the like. Therefore, the first data distribution of the first communication data in each communication category and the second data distribution of the second communication data in each communication category are obtained, and statistics are calculated based on the chi-square fitting goodness-of-fit detection method. Note that, the communication category in this embodiment refers to a communication category shared by the first communication data and the second communication data.
The calculation formula of the statistic is as follows:
Figure BDA0003797961140000101
wherein χ is 2 Represents statistics, k is the number of communication categories, O i The number E of the communication data of the information source is the number E of the communication data in the ith communication category i The number of the communication data of the i-th type communication category of the target communication data to be detected is determined.
3. And sorting all other targets according to the statistics to obtain distribution sorting.
Specifically, after obtaining the statistics, other targets corresponding to each statistic are confirmed, and then all other targets are sorted according to the ascending order of the statistics from low to high, so that the distribution sorting is obtained.
4. And calculating the Euclidean distance between the first average value vector of the first communication vector in each communication category and the second average value vector of each second communication vector in each communication category, and taking the average value to obtain the average Euclidean distance.
Specifically, the calculation formula of the average euclidean distance is:
Figure BDA0003797961140000102
wherein l represents the average Euclidean distance, k represents the number of communication categories, < >>
Figure BDA0003797961140000103
Mean value of communication data representing the i-th communication category of the information source,/->
Figure BDA0003797961140000104
Mean value of communication data representing ith communication category of target to be detected
5. And sorting all other targets according to the average Euclidean distance to obtain Euclidean sorting.
Specifically, after the average Euclidean distance corresponding to each information source is obtained, the information sources are sorted in ascending order according to the average Euclidean distance, and Euclidean sorting is obtained.
6. And confirming the final ordering of all other targets according to the distribution ordering and the European ordering, and selecting a first preset number of other targets as information sources according to the final ordering.
Specifically, for each other target, confirming a first sequence number corresponding to the other target in the distributed sequencing, simultaneously confirming a second sequence number corresponding to the other target in the European sequencing, taking the average value of the first sequence number and the second sequence number as the sequence number of the final sequencing, thus obtaining the final sequencing of each other target, and selecting a first preset number of other targets from the final sequencing as information sources. It should be noted that the first preset number is preset, for example, 2, 5, etc.
In this embodiment, by adopting an information source relatively similar to the target to be detected, the target to be detected can be better assisted to perform more accurate communication detection.
Step S103: inputting the first communication data and the third communication data of the information sources into a communication detection model for communication detection to obtain a communication detection result of the target to be detected aiming at the communication category, wherein the communication detection model comprises a feature extraction network and a global public detector, and the feature extraction network and the global public detector are trained and obtained based on a first loss function representing the distribution information and the importance of each information source and a second loss function representing the spatial gathering of the target to be detected and the information sources of the equipment to be detected.
Specifically, after the first communication data and the third communication data of the information source are obtained, the first communication data and the third communication data are input into a communication detection model to carry out communication detection, and a communication detection result of the target to be detected for the communication category is obtained.
It should be noted that, the communication detection model includes a feature extraction network and a global public detector, and in order to ensure the detection effect of the communication detection model, the feature extraction network and the global public detector are obtained by training based on a first loss function representing the distribution information and the importance of each information source and a second loss function representing the spatial aggregation of the target to be detected and the information source of the device to be detected.
Further, training the communication detection model specifically includes:
1. and respectively inputting the first communication sample data of the target to be detected and the second communication sample data of the information source into a feature extraction network to extract, thereby obtaining a first communication sample vector and a second communication sample vector.
In this embodiment, it should be understood that the communication data generated by different devices in the internet of things may have different dimensions, for example, the information source is a vehicle-mounted internet of things device, the communication data is a 10-dimensional vector, and the communication data of the internet of things network center (object to be tested) may be a 160-dimensional vector. Further, in this embodiment, the number of feature extraction networks is the same as the total number of targets to be tested and information sources, and each feature extraction network is responsible for extracting a communication vector from communication data of one device.
Specifically, before communication detection is performed on a certain network device in the internet of things, the network device is used as a target to be detected, the network device specified based on a preset rule is used as an information source, and communication data of the information source is used as an aid to improve the detection effect of the communication data of the target to be detected. After first communication sample data of a target to be detected and second communication sample data of an information source are obtained, a feature extraction network is utilized to extract first communication sample vectors and second communication sample vectors from the first communication sample data and the second communication sample data respectively.
2. And calculating the weight of each information source by using the first communication sample vector and the second communication sample vector.
It should be noted that, in order to weight the information sources to reflect the importance of each information source, that is, the transmission degree of the information source transmitting the communication detection information to the target to be detected, the algorithm performs weight assignment according to the feature space difference between the information source and the target to be detected.
Specifically, the calculating the weight of each information source by using the first communication sample vector and the second communication sample vector includes:
2.1, calculating the Euclidean distance between the first average value vector of the first communication sample vector in each communication category and the second average value vector of each second communication sample vector in each communication category, and obtaining the average Euclidean distance.
Specifically, please refer to the above for the calculation method of the average euclidean distance, and the description thereof is omitted herein.
2.2, calculating the weight of the information source according to the average Euclidean distance corresponding to each information source, wherein the calculation formula of the weight is expressed as follows:
Figure BDA0003797961140000131
wherein omega j Representing the weight of the jth information source, dist j And representing the average Euclidean distance between the jth information source and the target to be detected.
In this embodiment, the weight calculated according to the weight calculation formula described above is within the [0.75,1.25] interval, so that even if the information source is located closer to the target to be measured, the information source does not have a weight of 0 and thus loses the influence on the target to be measured. In addition, when the information source is relatively close to the target to be detected, the weight is relatively small, which means that the target to be detected has sufficiently acquired the communication detection information transmitted by the information source, so that the information source is given a relatively small weight to reflect the importance of the relatively small weight; on the contrary, when the information source is far away from the target to be measured, the weight is larger, which means that the target to be measured does not sufficiently acquire the communication detection information transmitted by the information source, so that the information source is endowed with a relatively larger weight to enhance the importance of the communication detection information, and the target to be measured can better grasp the communication detection information of the information source.
3. And respectively inputting the first communication sample vector and the second communication sample vector into a global public detector to perform distribution prediction, so as to obtain a first prediction distribution vector of the target to be detected in each communication category and a second prediction distribution vector of the information source in each communication category.
Specifically, after the original communication data of the information source and the target to be detected are mapped to the global feature space through the feature extraction network, the global public detector is utilized for detection.
4. And calculating a loss function value according to the first prediction distribution vector, the second prediction distribution vector, the weight, the first communication sample vector, the second communication sample vector, the first loss function and the second loss function.
Specifically, the first loss function is used for representing distribution information of different information sources and importance of the distribution information, and the second loss function represents that communication data of the information sources and communication data of a target to be detected are gathered at the angle of space characteristics.
Further, calculating a first loss function value of the first loss function includes:
4.11 calculating a first predicted distribution mean vector from the first predicted distribution vector.
The calculation formula of the first prediction distribution mean vector is as follows:
Figure BDA0003797961140000141
Wherein (1)>
Figure BDA0003797961140000142
Represents a first predictive distribution mean vector, |χ (k) I represents the number of communication data belonging to the kth class of communication category, C () represents the global common detector, f (x) represents the feature extraction network, +.>
Figure BDA0003797961140000143
Representing a first predicted distribution vector, T representing a temperature parameter.
And 4.12, calculating the KL divergence between the information source and the target to be measured according to the first prediction distribution mean value vector and the second prediction distribution vector.
Wherein, the calculation formula of KL divergence is:
Figure BDA0003797961140000144
Figure BDA0003797961140000145
wherein (1)>
Figure BDA0003797961140000146
Indicating KL divergence between jth information source and distribution information of target to be detected,/for>
Figure BDA0003797961140000147
Representing a second predicted distribution vector.
4.13, calculating a first loss function value according to the first prediction distribution mean value vector, the KL divergence and the weight.
The calculation formula of the first loss function value is as follows:
Figure BDA0003797961140000148
Figure BDA0003797961140000149
wherein LI is a first loss function value, k represents a communication classOther number, n, represents the number of information sources, ω j Representing the weight of the jth information source, |χ D I represents the number of first communication sample data, L ce () Representing cross entropy loss, y i And a communication detection type label is indicated.
In this embodiment, the distributed information can be transferred from the information source with rich information to the target to be detected in the communication detection process, and weights are added to dynamically reflect the importance of the information in the information source, so that accurate communication detection is better realized.
Further, calculating a second loss function value of the second loss function includes:
and 4.21, merging the second communication sample data of all the information sources into third communication sample data, and merging the first communication sample data of the target to be detected and the second communication sample data of all the information sources into fourth communication sample data.
Specifically, communication detection information transfer between an information source and a target to be detected is assisted from the angle of the characteristic space distance.
And 4.22, respectively extracting the third communication sample vector and the fourth communication sample vector from the third communication sample data and the fourth communication sample data by utilizing the characteristic extraction network.
And 4.23, respectively calculating a first average value vector, a third average value vector and a fourth average value vector of the first communication sample vector, the third communication sample vector and the fourth communication sample vector in each communication category.
Specifically, each communication category corresponds to at least one communication data, and the average value of vectors corresponding to all the communication data is obtained to obtain the average value vector of each communication category.
4.24, calculating Euclidean distance between the first mean value vector, the third mean value vector and the fourth mean value vector, and summing to obtain a first Euclidean distance loss function value.
Wherein, the formula of calculation of the first Euclidean distance loss function value is:
Figure BDA0003797961140000151
wherein LG denotes a first Euclidean distance loss function value,
Figure BDA0003797961140000152
representing a first mean vector, ">
Figure BDA0003797961140000153
Representing a third mean vector, ">
Figure BDA0003797961140000154
Represents a fourth mean vector, k represents the number of communication categories,/->
Figure BDA0003797961140000155
Representing the euclidean distance.
And 4.25, respectively selecting a second preset number of first characteristic points from the first communication sample vectors of the communication categories by using a clustering algorithm, and selecting a second preset number of second characteristic points from the third communication sample vectors of the communication categories.
It should be noted that, from the point of view of a local representative point, the information source and the target to be measured need to be spatially gathered. Wherein the clustering algorithm includes, but is not limited to, a Kmeans++ clustering algorithm.
And 4.26, calculating Euclidean distances between the first characteristic points and the second characteristic points, and averaging to obtain a second Euclidean distance loss function value.
Wherein, the formula of calculation of the second Euclidean distance loss function value is:
Figure BDA0003797961140000161
wherein LL represents a second euclidean distance loss function value, R represents a second preset number,
Figure BDA0003797961140000162
representing a first communication sample vectorN-th,/-th in the first feature points of the selected R i-th communication categories >
Figure BDA0003797961140000163
Representing the mth of the second feature points of the selected R ith communication categories in the third communication sample vector.
And 4.27, calculating a second loss function value according to the first Euclidean distance loss function value and the second Euclidean distance loss function value.
Specifically, the second loss function value=the first euclidean distance loss function value+the second euclidean distance loss function value.
5. And carrying out iterative training on the feature extraction network and the global public detector according to the loss function value and a preset optimization algorithm.
Specifically, the preset optimization algorithm includes, but is not limited to, an optimization algorithm such as a random gradient descent method.
According to the communication detection method, the information sources for assisting the to-be-detected targets in communication detection are selected from the network equipment according to the communication detection task by utilizing the similarity among the data, the network equipment comprises the Internet of things and equipment in the network center of the Internet of things, communication data of the to-be-detected targets and communication data of the information sources are input into the communication detection model for communication detection, communication detection results of the to-be-detected targets aiming at communication categories are obtained, the communication detection model is trained based on a first loss function representing distribution information and importance of each information source and a second loss function representing the to-be-detected targets and the information sources on space gathering, and therefore when the communication data of the to-be-detected targets are detected, the information sources are used as assistance, information complementation between the information sources and the to-be-detected targets is utilized, more refined detection of the communication data of the to-be-detected targets is facilitated, detection effects of the to-be-detected targets are improved, and detection accuracy is improved.
Fig. 2 is a schematic diagram of a functional module of a communication detection device according to an embodiment of the invention. As shown in fig. 2, the communication detection device 20 includes a receiving module 21, a selecting module 22, and a predicting module 23.
The receiving module 21 is configured to receive a communication detection task, where the communication detection task includes a target to be detected, a communication category, and a network device where the target to be detected is located; the selecting module 22 is configured to screen and obtain an information source from other targets based on the similarity between the first communication data corresponding to the communication category of the target to be tested and the second communication data corresponding to the communication category of other targets in the network device; the prediction module 23 is configured to input the first communication data and the third communication data of the information sources into a communication detection model for communication detection, so as to obtain a communication detection result of the to-be-detected target for the communication category, where the communication detection model includes a feature extraction network and a global public detector, and the feature extraction network and the global public detector are obtained by training based on a first loss function representing distribution information and importance of each information source and a second loss function representing spatial aggregation of the to-be-detected target and the information source of the to-be-detected device.
Optionally, the selecting module 22 performs screening to obtain an information source from other targets based on the similarity between the first communication data corresponding to the communication category of the target to be tested and the second communication data corresponding to the communication category of other targets in the network device, including: acquiring first communication data of a target to be detected and second communication data of other targets, extracting the first communication data to obtain a first communication vector by utilizing a feature extraction network, and extracting the second communication vector from the second communication data; obtaining statistics according to first data distribution of the first communication data in each communication category and second data distribution of the second communication data in each communication category; sorting all other targets according to the statistics to obtain distribution sorting; calculating Euclidean distance between a first average value vector of the first communication vector in each communication category and a second average value vector of each second communication vector in each communication category, and taking an average value to obtain an average Euclidean distance; sequencing all other targets according to the average Euclidean distance to obtain Euclidean sequencing; and confirming the final ordering of all other targets according to the distribution ordering and the European ordering, and selecting a first preset number of other targets as information sources according to the final ordering.
Optionally, the communication detection device 20 further includes a training module, configured to train a communication detection model, and specifically includes: respectively inputting first communication sample data of a target to be detected and second communication sample data of an information source into a feature extraction network to extract, so as to obtain a first communication sample vector and a second communication sample vector; calculating the weight of each information source by using the first communication sample vector and the second communication sample vector; respectively inputting the first communication sample vector and the second communication sample vector into a global public detector to perform distribution prediction to obtain a first prediction distribution vector of a target to be detected in each communication category and a second prediction distribution vector of an information source in each communication category; calculating a loss function value according to the first prediction distribution vector, the second prediction distribution vector, the weight, the first communication sample vector, the second communication sample vector, the first loss function and the second loss function; and carrying out iterative training on the feature extraction network and the global public detector according to the loss function value and a preset optimization algorithm.
Optionally, the training module performs a calculation using the first communication sample vector and the second communication sample vector to obtain a weight of each information source, including: calculating Euclidean distance between a first average value vector of the first communication sample vector in each communication category and a second average value vector of each second communication sample vector in each communication category, and taking an average value to obtain an average Euclidean distance; and calculating the weight of the information source according to the average Euclidean distance corresponding to each information source, wherein the calculation formula of the weight is expressed as follows:
Figure BDA0003797961140000181
Wherein omega j Representing the weight of the jth information source, dist j And representing the average Euclidean distance between the jth information source and the target to be detected.
Optionally, the operation of the training module calculating the first loss function value of the first loss function specifically includes: calculating a first prediction distribution mean vector according to the first prediction distribution vector; calculating KL divergence between the information source and the target to be detected according to the first prediction distribution mean value vector and the second prediction distribution mean value vector; and calculating a first loss function value according to the first prediction distribution mean vector, the KL divergence and the weight.
Optionally, the calculation formula of the first prediction distribution mean vector is:
Figure BDA0003797961140000191
wherein (1)>
Figure BDA0003797961140000192
Represents a first predictive distribution mean vector, |χ (k) I represents the number of communication data belonging to the kth class of communication category, C () represents the global common detector, f (x) represents the feature extraction network, +.>
Figure BDA0003797961140000193
Representing a first predicted distribution vector, T representing a temperature parameter;
the calculation formula of the KL divergence is:
Figure BDA0003797961140000194
Figure BDA0003797961140000195
wherein (1)>
Figure BDA0003797961140000196
Indicating KL divergence between jth information source and distribution information of target to be detected,/for>
Figure BDA0003797961140000197
Representing a second predicted distribution vector;
the calculation formula of the first loss function value is:
Figure BDA0003797961140000198
wherein LI is a first loss function value, k represents the number of communication categories, n represents the number of information sources, ω j Representing the weight of the jth information source, |χ D I represents the number of first communication sample data, L ce () Representing cross entropy loss, y i Communication detection class labelAnd (5) signing.
Optionally, the operation of the training module to perform calculating the second loss function value of the second loss function specifically includes: combining the second communication sample data of all the information sources into third communication sample data, and combining the first communication sample data of the target to be detected and the second communication sample data of all the information sources into fourth communication sample data; extracting a third communication sample vector and a fourth communication sample vector from the third communication sample data and the fourth communication sample data respectively by utilizing a characteristic extraction network; respectively calculating a first communication sample vector, a third communication sample vector, a fourth communication sample vector, a first average value vector, a third average value vector and a fourth average value vector of each communication class; calculating Euclidean distances between the first mean value vector, the third mean value vector and the fourth mean value vector, and summing to obtain a first Euclidean distance loss function value; respectively selecting a second preset number of first characteristic points from the first communication sample vectors of all the communication categories by using a clustering algorithm, and selecting a second preset number of second characteristic points from the third communication sample vectors of all the communication categories; calculating Euclidean distances between the first characteristic points and the second characteristic points, and averaging to obtain a second Euclidean distance loss function value; and calculating a second loss function value according to the first Euclidean distance loss function value and the second Euclidean distance loss function value.
Optionally, the calculation formula of the first euclidean distance loss function value is:
Figure BDA0003797961140000201
wherein LG denotes a first Euclidean distance loss function value,
Figure BDA0003797961140000202
representing a first mean vector, ">
Figure BDA0003797961140000203
Representing a third mean vector, ">
Figure BDA0003797961140000204
Represents a fourth mean vector, k represents the number of communication categories,/->
Figure BDA0003797961140000205
Representing the Euclidean distance;
the calculation formula of the second Euclidean distance loss function value is as follows:
Figure BDA0003797961140000206
wherein LL represents a second euclidean distance loss function value, R represents a second preset number,
Figure BDA0003797961140000207
n-th, i.e. in the first feature points representing the selected R i-th communication classes in the first communication sample vector>
Figure BDA0003797961140000208
Representing the mth of the second feature points of the selected R ith communication categories in the third communication sample vector.
For further details of the implementation of the communication detection device according to the foregoing embodiments, reference may be made to the description of the communication detection method according to the foregoing embodiments, which is not repeated herein.
It should be noted that, in the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described as different from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other. For the apparatus class embodiments, the description is relatively simple as it is substantially similar to the method embodiments, and reference is made to the description of the method embodiments for relevant points.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the invention. As shown in fig. 3, the computer device 30 includes a processor 31 and a memory 32 coupled to the processor 31, where the memory 32 stores program instructions that, when executed by the processor 31, cause the processor 31 to perform the communication detection method steps described in any of the embodiments above.
The processor 31 may also be referred to as a CPU (Central Processing Unit ). The processor 31 may be an integrated circuit chip with signal processing capabilities. The processor 31 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Referring to fig. 4, fig. 4 is a schematic structural diagram of a storage medium according to an embodiment of the present invention. The storage medium according to the embodiment of the present invention stores the program instructions 41 capable of implementing the communication detection method, where the program instructions 41 may be stored in the storage medium in the form of a software product, and include several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random-access Memory (RAM, random Access Memory), a magnetic disk, an optical disk, or other various media capable of storing program codes, or a computer device such as a computer, a server, a mobile phone, a tablet, or the like.
In the several embodiments provided in this application, it should be understood that the disclosed computer apparatus, device, and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units. The foregoing is only the embodiments of the present application, and not the patent scope of the present application is limited by the foregoing description, but all equivalent structures or equivalent processes using the contents of the present application and the accompanying drawings, or directly or indirectly applied to other related technical fields, which are included in the patent protection scope of the present application.

Claims (10)

1. A method of communication detection, the method comprising:
receiving a communication detection task, wherein the communication detection task comprises a target to be detected, a communication category and network equipment where the target to be detected is located;
screening information sources from other targets based on the similarity between the first communication data corresponding to the communication category of the target to be detected and the second communication data corresponding to the communication category of the other targets in the network equipment;
inputting the first communication data and the third communication data of the information sources into a communication detection model for communication detection to obtain a communication detection result of the target to be detected aiming at the communication category, wherein the communication detection model comprises a feature extraction network and a global public detector, and the feature extraction network and the global public detector are trained and obtained based on a first loss function representing the distribution information of each information source and the importance of the information sources and a second loss function representing the spatial gathering of the target to be detected and the information sources.
2. The communication detection method according to claim 1, wherein the screening the information source from the other targets based on the similarity between the first communication data corresponding to the communication category of the target to be detected and the second communication data corresponding to the communication category of the other targets in the network device includes:
Acquiring the first communication data of the target to be detected and second communication data of other targets, extracting a first communication vector from the first communication data by utilizing the characteristic extraction network, and extracting a second communication vector from the second communication data;
obtaining statistics according to first data distribution of the first communication data in each communication category and second data distribution of the second communication data in each communication category;
sorting all other targets according to the statistics to obtain distribution sorting;
calculating Euclidean distance between a first average value vector of the first communication vector in each communication category and a second average value vector of each second communication vector in each communication category, and taking an average value to obtain an average Euclidean distance;
sorting all other targets according to the average Euclidean distance to obtain Euclidean sorting;
and confirming the final ordering of all other targets according to the distribution ordering and the European ordering, and selecting a first preset number of other targets as the information sources according to the final ordering.
3. The communication detection method according to claim 1, wherein training the communication detection model specifically comprises:
Respectively inputting the first communication sample data of the target to be detected and the second communication sample data of the information source into the feature extraction network for extraction to obtain a first communication sample vector and a second communication sample vector;
calculating the weight of each information source by using the first communication sample vector and the second communication sample vector;
the first communication sample vector and the second communication sample vector are respectively input to the global public detector to conduct distribution prediction, so that a first prediction distribution vector of the target to be detected in each communication category and a second prediction distribution vector of the information source in each communication category are obtained;
calculating a loss function value according to the first prediction distribution vector, the second prediction distribution vector, the weight, the first communication sample vector, the second communication sample vector, the first loss function and the second loss function;
and carrying out iterative training on the feature extraction network and the global public detector according to the loss function value and a preset optimization algorithm.
4. The communication detection method according to claim 3, wherein the calculating the weight of each information source using the first communication sample vector and the second communication sample vector includes:
Calculating Euclidean distance between a first average value vector of the first communication sample vector in each communication category and a second average value vector of each second communication sample vector in each communication category, and taking an average value to obtain an average Euclidean distance;
calculating the weight of each information source according to the average Euclidean distance corresponding to the information source, wherein the calculation formula of the weight is expressed as follows:
Figure QLYQS_1
wherein omega j Representing the weight of the jth information source, dist j And representing the average Euclidean distance between the jth information source and the target to be detected.
5. The communication detection method of claim 3, wherein calculating a first loss function value of the first loss function comprises:
calculating a first prediction distribution mean vector according to the first prediction distribution vector;
calculating KL divergence between the information source and the target to be detected according to the first prediction distribution mean value vector and the second prediction distribution vector;
and calculating the first loss function value according to the first prediction distribution mean vector, the KL divergence and the weight.
6. The communication detection method according to claim 5, wherein the calculation formula of the first prediction distribution mean vector is:
Figure QLYQS_2
Wherein (1)>
Figure QLYQS_3
Representing said first predictive distribution mean vector,/->
Figure QLYQS_4
Representing the number of communication data belonging to the k-th communication class, C () representing said global common detector, f (x) representing said feature extraction network,/->
Figure QLYQS_5
Representing the first predicted distribution vector, T representing a temperature parameter;
the calculation formula of the KL divergence is as follows:
Figure QLYQS_6
Figure QLYQS_7
wherein (1)>
Figure QLYQS_8
Indicating KL divergence between jth information source and distribution information of target to be detected,/for>
Figure QLYQS_9
Representing the second predicted distribution vector;
the calculation formula of the first loss function value is as follows:
Figure QLYQS_10
Figure QLYQS_11
wherein LI is the first loss function value, k represents the number of communication categories, n represents the number of information sources, ω j Weight representing the jth information source, < ->
Figure QLYQS_12
Indicating the number of the first communication sample data, L ce () Representing cross entropy loss, y i And a communication detection type label is indicated.
7. The communication detection method according to claim 3, wherein calculating a second loss function value of the second loss function comprises:
merging the second communication sample data of all the information sources into third communication sample data, and merging the first communication sample data of the target to be detected and the second communication sample data of all the information sources into fourth communication sample data;
Extracting a third communication sample vector and a fourth communication sample vector from the third communication sample data and the fourth communication sample data respectively by utilizing the characteristic extraction network;
respectively calculating a first average value vector, a third average value vector and a fourth average value vector of the first communication sample vector, the third communication sample vector and the fourth communication sample vector in each communication category;
calculating Euclidean distances between the first mean value vector, the third mean value vector and the fourth mean value vector, and summing to obtain a first Euclidean distance loss function value;
a clustering algorithm is utilized to respectively select a second preset number of first characteristic points from the first communication sample vectors of all the communication categories, and select the second preset number of second characteristic points from the third communication sample vectors of all the communication categories;
calculating Euclidean distances between the first characteristic points and the second characteristic points, and averaging to obtain a second Euclidean distance loss function value;
and calculating the second loss function value according to the first Euclidean distance loss function value and the second Euclidean distance loss function value.
8. The communication detection method according to claim 7, wherein the first euclidean distance loss function value is calculated by the following formula:
Figure QLYQS_13
wherein LG represents the first Euclidean distance loss function value,
Figure QLYQS_14
representing the first mean vector, +.>
Figure QLYQS_15
Representing the third mean vector, +.>
Figure QLYQS_16
Representing the fourth mean vector, k representing the number of communication categories, +.>
Figure QLYQS_17
Representing the Euclidean distance;
the calculation formula of the second Euclidean distance loss function value is as follows:
Figure QLYQS_18
wherein LL represents the second euclidean distance loss function value, R represents the second preset number,
Figure QLYQS_19
n-th,/in the first feature points representing the selected R i-th communication categories in the first communication sample vector>
Figure QLYQS_20
And representing the mth of the second characteristic points of the R ith communication categories selected from the third communication sample vector.
9. A computer device comprising a processor, a memory coupled to the processor, the memory having stored therein program instructions that, when executed by the processor, cause the processor to perform the steps of the communication detection method of any of claims 1-8.
10. A storage medium storing program instructions capable of implementing the communication detection method according to any one of claims 1 to 8.
CN202210973887.7A 2022-08-15 2022-08-15 Communication detection method, device and storage medium Active CN115378856B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210973887.7A CN115378856B (en) 2022-08-15 2022-08-15 Communication detection method, device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210973887.7A CN115378856B (en) 2022-08-15 2022-08-15 Communication detection method, device and storage medium

Publications (2)

Publication Number Publication Date
CN115378856A CN115378856A (en) 2022-11-22
CN115378856B true CN115378856B (en) 2023-07-14

Family

ID=84066571

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210973887.7A Active CN115378856B (en) 2022-08-15 2022-08-15 Communication detection method, device and storage medium

Country Status (1)

Country Link
CN (1) CN115378856B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113344692A (en) * 2021-04-24 2021-09-03 大连理工大学 Method for establishing network loan credit risk assessment model with multi-information-source fusion
CN113392359A (en) * 2021-08-18 2021-09-14 腾讯科技（深圳）有限公司 Multi-target prediction method, device, equipment and storage medium
WO2021258480A1 (en) * 2020-06-22 2021-12-30 网宿科技股份有限公司 Web crawler detection method, system, and device employing graph neural network

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003271656A (en) * 2002-03-19 2003-09-26 Fujitsu Ltd Device and method for related candidate generation, related system, program for related candidate generation and readable recording medium recorded with the same program
US8572107B2 (en) * 2011-12-09 2013-10-29 International Business Machines Corporation Identifying inconsistencies in object similarities from multiple information sources

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021258480A1 (en) * 2020-06-22 2021-12-30 网宿科技股份有限公司 Web crawler detection method, system, and device employing graph neural network
CN113344692A (en) * 2021-04-24 2021-09-03 大连理工大学 Method for establishing network loan credit risk assessment model with multi-information-source fusion
CN113392359A (en) * 2021-08-18 2021-09-14 腾讯科技（深圳）有限公司 Multi-target prediction method, device, equipment and storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于KL距离的卷积神经网络人脸特征提取模型;罗可;周安众;;长沙理工大学学报(自然科学版)(02);全文 *
检测信息源的群集人员监测及预警系统研究;肖宾杰;;电气自动化(01);全文 *

Also Published As

Publication number Publication date
CN115378856A (en) 2022-11-22

Similar Documents

Publication Publication Date Title
CN108280477B (en) Method and apparatus for clustering images
CN112085952B (en) Method and device for monitoring vehicle data, computer equipment and storage medium
CN109583332B (en) Face recognition method, face recognition system, medium, and electronic device
WO2021253510A1 (en) Bidirectional interactive network-based pedestrian search method and system, and device
CN102254394A (en) Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis
WO2022142903A1 (en) Identity recognition method and apparatus, electronic device, and related product
CN112770265B (en) Pedestrian identity information acquisition method, system, server and storage medium
CN110909222A (en) User portrait establishing method, device, medium and electronic equipment based on clustering
WO2021190470A1 (en) Electronic device positioning method and apparatus, and server, system and storage medium
CN113781767A (en) Traffic data fusion method and system based on multi-source perception
CN112712102A (en) Recognizer capable of simultaneously recognizing known radar radiation source individuals and unknown radar radiation source individuals
CN110348516B (en) Data processing method, data processing device, storage medium and electronic equipment
CN108304852B (en) Method and device for determining road section type, storage medium and electronic device
CN115378856B (en) Communication detection method, device and storage medium
CN115114329A (en) Method and device for detecting data stream abnormity, electronic equipment and storage medium
CN116935083A (en) Image clustering method and device
CN115392351A (en) Risk user identification method and device, electronic equipment and storage medium
CN114333997A (en) Data processing method, data processing model training method, data processing device, data processing equipment and data processing medium
CN113762037A (en) Image recognition method, device, equipment and storage medium
US20230409422A1 (en) Systems and Methods for Anomaly Detection in Multi-Modal Data Streams
CN117478434B (en) Edge node network traffic data processing method, device, equipment and medium
CN116776160B (en) Data processing method and related device
CN111401224B (en) Target detection method and device and electronic equipment
CN116578874B (en) Satellite signal attribute appraising method and device based on network protocol
CN116737932A (en) Self-adaptive text classification method in scientific and technological information field

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant