CN112968814A - Internet of things data message distribution method and equipment - Google Patents
Internet of things data message distribution method and equipment Download PDFInfo
- Publication number
- CN112968814A CN112968814A CN202110250610.7A CN202110250610A CN112968814A CN 112968814 A CN112968814 A CN 112968814A CN 202110250610 A CN202110250610 A CN 202110250610A CN 112968814 A CN112968814 A CN 112968814A
- Authority
- CN
- China
- Prior art keywords
- data message
- internet
- things
- message
- transmission
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000005540 biological transmission Effects 0.000 claims abstract description 106
- 230000002159 abnormal effect Effects 0.000 claims abstract description 31
- 238000001514 detection method Methods 0.000 claims abstract description 31
- 230000000737 periodic effect Effects 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000035772 mutation Effects 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0823—Errors, e.g. transmission errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Cardiology (AREA)
- Environmental & Geological Engineering (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to the technical field of message distribution, and discloses an Internet of things data message distribution method, which comprises the following steps: detecting a big flow in the data message transmission of the Internet of things by using a big flow detection algorithm based on a transmission rate; for the detected internet of things transmission large flow, the transmission of the internet of things transmission large flow is carried out by utilizing a weighted polling scheduling strategy with length limitation; clustering messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters; constructing a corresponding SVM model by using the Internet of things data message clustering cluster, and detecting abnormal data messages by using an abnormal data message detection algorithm combining KNN and the SVM model; and realizing the distribution of the data message of the Internet of things by using a mixed distribution strategy. The invention also provides equipment for distributing the data message of the Internet of things. The invention realizes the distribution of the data message of the Internet of things.
Description
Technical Field
The invention relates to the technical field of message distribution, in particular to a method and equipment for distributing data messages of an Internet of things.
Background
With the development of cloud computing and big data processing technologies, the scale of the internet of things becomes larger and larger, and the performance and the number of servers of the internet of things are increased in an exponential mode; the traffic mode is mainly changed from the prior longitudinal traffic from a server to a user node into the communication between servers in the internet of things, and the realization of efficient data message transmission in the internet of things becomes a hot topic of current research.
Some links in the internet of things are not high in utilization rate, but some links are frequently congested, so that when congestion occurs, message transmission delay is increased, effective throughput of a network is reduced, and even messages are discarded when the congestion is serious, so that service performance and service quality are influenced; the traditional flow balance strategy is a flow balance strategy taking data flow as granularity, under the condition of high load, due to mutual influence of a plurality of high-speed flows, the route of the data flow is continuously changed, the probability of message out-of-sequence is high, and the network transmission performance is influenced.
In view of this, how to implement more efficient data packet transmission under the condition of avoiding packet out-of-order becomes a problem to be solved urgently by those skilled in the art.
Disclosure of Invention
The invention provides a method for distributing data messages of the Internet of things, which is characterized in that a large flow detection algorithm based on a transmission rate is used for detecting the large flow in the data message transmission of the Internet of things, and a weighted polling scheduling strategy based on length limitation is used for carrying out network resource allocation scheduling; and meanwhile, the abnormal data message of the Internet of things is detected by using an abnormal message detection algorithm based on density, and finally, the data message of the Internet of things is distributed by using a mixed distribution strategy.
In order to achieve the above object, the present invention provides a method for distributing data packets of an internet of things, including:
detecting a big flow in the data message transmission of the Internet of things by using a big flow detection algorithm based on a transmission rate;
for the detected internet of things transmission large flow, the transmission of the internet of things transmission large flow is carried out by utilizing a weighted polling scheduling strategy with length limitation;
clustering messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters;
constructing a corresponding SVM model by using the Internet of things data message clustering cluster, and detecting abnormal data messages by using an abnormal data message detection algorithm combining KNN and the SVM model;
and realizing the distribution of the data message of the Internet of things by using a mixed distribution strategy.
Optionally, the flow of the transmission rate-based large flow detection algorithm is as follows:
the large flow is a flow with more than a certain number of messages transmitted in a period of time;
the flow of the large flow detection algorithm based on the transmission rate is as follows:
when a new message in the Internet of things arrives, extracting a head field of the message to generate a stream identifier, generating d indexes by using d different Hash functions by taking the stream identifier to be searched as input, and searching in corresponding Hash buckets in d sub-tables respectively;
if the corresponding stream entry is found, accumulating the byte number, and judging whether the data volume reaches the data volume threshold of the large stream, if so, calculating whether the stream rate reaches the rate threshold of the large stream;
if the flow item corresponding to the newly received message is not found, judging whether a free storage space exists or not from the Hash barrel with the numbered minimum sub-table;
if the free storage space exists, inserting an entry of a new flow into the free storage space, if the free storage space does not exist, finding out a flow min with the minimum transmission rate in a Hash bucket of d sub-tables corresponding to the current new flow, if the transmission rate of the data flow is smaller than a removal threshold, removing the flow, allocating the released space to the current new flow, and recording the generation time and the number of bytes of the new flow; if the rate of min exceeds the drop threshold, the new flow currently arriving is ignored.
Optionally, the transmission of the internet of things transmission large stream is performed by using a weighted polling scheduling strategy with length limitation.
1) If N internet of things transmission large flows exist at the transmission port of the internet of things, the weight value of the internet of things transmission large flow i is wiThe length of the allocated query share is Li,Lre_iTransmitting the remaining quota length of the big stream i served last time for the Internet of things;
2) divide a poll into T sub-polls, where T ═ max (w)i) In one sub-polling, when the Internet of things transmission stream i starts to send a message, if the total length of the message is less than or equal to Li+Lre_iWhen no new message arrives, all messages in the transmission stream i of the Internet of things are sent, and at the moment Lre_i0 bytes; if the Internet of thingsThe total length of the message in the transmission large flow i is more than Li+Lre_iThen, n messages are sequentially sent from the head, wherein n satisfies:
wherein:
the length of the jth message counted from the head of a queue at the beginning of transmission in the large stream i transmitted by the Internet of things is obtained;
and when the residual share is not enough to send the (n + 1) th message, finishing the polling transmission, wherein the residual share is L're_iSatisfies the following conditions:
3) for the high-priority internet of things to transmit the large flow, the message in the high-priority internet of things is sent preferentially; and only when the transmission large stream of the high-priority Internet of things is empty, the transmission large stream of other Internet of things can be served.
Optionally, the clustering of the packets by using the density-based clustering algorithm includes:
1) for the collected data message sample x (SA, DA, MT, L), wherein SA represents the source address of the data message, DA represents the destination address of the data message, MT represents the type of the data message, and L is the length of the data message;
2) normalizing the data message sample attributes to obtain normalized data message sample attributes (SA, DA, MT, L)*(ii) a Calculating the distance between any two data message samples according to the normalized data message sample attributes, and storing the distance between the samples into an m-order matrix D, wherein m represents the number of the data message samples, and the data message sample distance calculation formula is as follows:
wherein:
xi,xjany two data message samples are obtained;
3) for each data message sample xjAfter sequencing the distance sequences of all other data message samples, combining the distance sequences into a new m-order matrix Ds:
wherein:
ds (i, j) represents data message sample xjDistance from the i-1 th data message sample closest thereto;
ds (1, j) represents data message sample xjDistance from itself;
calculating the difference between adjacent elements in Ds (: j), and selecting the top element Ds (k) from the adjacent element with the largest differencejJ) nearest neighbor radius x as samplejThen data message sample xjAverage distance Ava to all other samples in its neighborhoodjComprises the following steps:
wherein:
kjrepresenting data message samples xjThe number of samples in the neighborhood of (1);
obtaining the average distance from all data message samples to other samples in the neighboring domain, and calculating the average value Ava:
wherein:
m is the number of data message samples;
4) samples corresponding to the first element larger than Ava are searched from small to large in sequence on each column of the matrix Ds, and each data message sample x can be determinedjNearest neighbor sample x outside Avai:
Wherein:
ds (h, j) is a data message sample xjDistance to h-1 sample nearest thereto;
h is dist (x)i,xj) At Ds (: the corresponding serial number in j);
calculating the times that each data message sample becomes the nearest sample outside the radius Ava of other samples, and recording the maximum value of the times as k;
5) sorting the distances between all samples and the nearest kth data message sample to form a sequence { kd }, and calculating the difference value of two adjacent elements in the sequence to form a sequence { kdd }; analyzing the variation amplitude of the element in the { kd } through the { kdd } sequence, and finding out the mutation point sequence number e of the { kd } sequence, wherein the sequence number satisfies the following conditions:
selecting the minimum serial number e from the serial numbers satisfying the formula, taking e as a parameter of the DBSCAN algorithm, and taking the k value as MinPts; the DBSCAN algorithm judges a core object in the sample by using (e, k) and carries out cluster search;
after one cluster search of the DBSCAN algorithm is completed, if the number of the data message samples which are not classified exceeds 10% of the total number of samples, the data message samples which are not classified are taken as a set, clustering is started again from the near-neighbor region, so that misjudgment of noise samples caused by uneven density distribution is avoided, and K cluster clusters are obtained.
Optionally, the detecting the abnormal data packet by using the abnormal data packet detection algorithm combining the KNN and the SVM model includes:
1) calculating the distance between the data message x to be detected and other detected data messages, and finding out K adjacent data messages (x) closest to the data message to be detected1,x2,…,xk) Wherein K is the number of clustering clusters;
2) judging the cluster of the data message x to be detected by adopting a weighted voting mode, namely throwing a vote with a weight value for each adjacent data message as the type of the adjacent data message xiThe weight value possessed during voting is exp (-dist (x, x)i)),dist(x,xi) For neighbor data message xiThe distance from the data message x to be detected is that the vote weight thrown by the adjacent data message closer to the data message x to be detected is larger, and the type with the maximum total weight is the cluster C to which the data message to be detected belongsu;
3) Carrying out anomaly detection on the data message to be detected by utilizing an SVM model corresponding to the cluster, wherein the SVM model is composed of a corresponding cluster CuAll data messages containedDetermining, wherein m is a cluster CuThe number of the contained data messages is obtained by utilizing an SVM model, and the square of the radius is as follows:
wherein:
m is a cluster CuThe number of the contained data messages;
k () represents a gaussian kernel function;
4) calculating the square of the distance from the data message to be detected to the sphere center of the SVM hypersphere as:
if d is2(x) If the data message to be detected is larger than R, the data message to be detected is positioned outside the hypersphere, the data message to be detected is an abnormal data message, and otherwise, the data message to be detected is a normal data message; and sending error information to the source address of the data message for the detected abnormal data message.
Optionally, the hybrid distribution policy is:
initializing a mixed distribution table, filling periodic data messages in the Internet of things, wherein the periodic messages are filled from the position of the upper left corner in the mixed distribution table, the arrangement mode of the messages is filled in sequence from small to large according to the receiving and sending period of the messages, the messages with the smaller receiving and sending period are arranged in the earlier basic time block, and the operation is repeated until all the periodic messages are filled in the mixed distribution table;
the periodic message is sent according to the basic time block specified in the mixed distribution table in advance, and if the event message needs to be sent at the moment, the periodic message can quickly find out the nearest basic time block to fill in the event message and send the event message.
In addition, to achieve the above object, the present invention further provides an internet of things data message distribution device, where the device includes:
the data message acquisition device is used for detecting a large flow in the transmission of the data message of the Internet of things by using a large flow detection algorithm based on the transmission rate, and transmitting the large flow in the transmission of the Internet of things by using a weighted polling scheduling strategy based on length limitation, so as to obtain the data message in the large flow in the transmission of the Internet of things;
the data message processor is used for clustering messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters, constructing a corresponding SVM model by using the Internet of things data message clustering clusters, and detecting abnormal data messages by using an abnormal data message detection algorithm combining the KNN and the SVM models;
and the data message distribution device is used for realizing the distribution of the data message of the Internet of things by utilizing a hybrid distribution strategy.
In addition, to achieve the above object, the present invention also provides a computer readable storage medium, which stores data message distribution program instructions, where the data message distribution program instructions are executable by one or more processors to implement the steps of the implementation method for data message distribution of the internet of things as described above.
Compared with the prior art, the invention provides a data message distribution method of the Internet of things, which has the following advantages:
firstly, for streams which exist in the Internet of things and are transmitted within a period of time and exceed a certain number, the invention detects the stream in the data message transmission of the Internet of things by using a transmission rate-based stream detection algorithm, extracts the head field of the message to generate a stream identifier when a new message in the Internet of things arrives, generates d indexes by using d different Hash functions with the stream identifier to be searched as input, and searches in corresponding Hash buckets in d sub-tables respectively; if the corresponding stream entry is found, accumulating the byte number, and judging whether the data volume reaches the data volume threshold of the large stream, if so, calculating whether the stream rate reaches the rate threshold of the large stream; if the flow item corresponding to the newly received message is not found, judging whether a free storage space exists or not from the Hash barrel with the numbered minimum sub-table; if the free storage space exists, inserting an entry of a new stream into the free storage space, if the free storage space does not exist, finding out a stream min with the minimum transmission rate in a Hash bucket of d sub-tables corresponding to the current new stream, if the transmission rate of the data stream is smaller than a de-selection threshold, de-selecting the stream, allocating the released space to the current new stream, and recording the generation time and the number of bytes of the new stream; if the rate of min exceeds the drop threshold, the new stream currently arriving is ignored. Compared with the prior art, the method and the device have the advantages that the large flow of the Internet of things is detected from the data volume threshold and the rate threshold, and the detected large flow is subjected to targeted processing.
Meanwhile, aiming at the detected Internet of things big flow, the invention provides a weighted polling scheduling strategy based on length limitation to transmit the Internet of things transmission flow, if N Internet of things transmission big flows exist at the transmission port of the Internet of things, the weight value of the Internet of things transmission big flow i is wiThe length of the allocated query share is Li,Lre_iTransmitting the remaining quota length of the big stream i served last time for the Internet of things; divide a poll into T sub-polls, where T ═ max (w)i) In one sub-polling, when the Internet of things transmission stream i starts to send a message, if the total length of the message is less than or equal to Li+Lre_iWhen no new message arrives, all messages in the transmission stream i of the Internet of things are sent, and at the moment Lre_i0 bytes; if the total length of the message in the transmission large flow i of the Internet of things is greater than Li+Lre_iThen, n messages are sequentially sent from the head, wherein n satisfies:
wherein:the length of the jth message counted from the head of a queue at the beginning of transmission in the large stream i transmitted by the Internet of things is obtained; and when the residual share is not enough to send the (n + 1) th message, finishing the polling transmission, wherein the residual share is L're_iSatisfies the following conditions:
for the high-priority internet of things to transmit the large flow, the message in the high-priority internet of things is sent preferentially; only when the high-priority internet of things transmits a large stream to be empty, the high-priority internet of things can be servedOther internet of things transmits large streams. According to the algorithm of the invention, under the condition that all polling large flows are not empty, the condition of sending the message is equal to the condition that the transmission flow j sends a frame of message in each service, and the average length of the message is LjThe allocated bandwidth of transport stream j is:
wherein: r is the total bandwidth value L in the Internet of thingsjIs the length of the message in transport stream j, WjThe total length of each transmission stream for sending the message is set on a fixed numerical value under the condition that all the transmission streams are not empty, the influence of the length change of the message on bandwidth distribution is small, and the network distribution balance in the Internet of things is effectively realized.
Because the traditional DBSCAN algorithm is greatly influenced by neighborhood parameters, the invention improves the traditional DBSCAN algorithm, firstly, the invention carries out normalization processing on the data message sample attributes to obtain normalized data message sample attributes (SA, DA, MT, L)*(ii) a Calculating the distance between any two data message samples according to the normalized data message sample attributes, and storing the distance between the samples into an m-order matrix D, wherein m represents the number of the data message samples, and the data message sample distance calculation formula is as follows:
wherein: x is the number ofi,xjAny two data message samples are obtained; for each data message sample xjAfter sequencing the distance sequences of all other data message samples, combining the distance sequences into a new m-order matrix Ds:
wherein: ds (i)J) represents a data message sample xjDistance from the i-1 th data message sample closest thereto; ds (1, j) represents data message sample xjDistance from itself; by calculating the difference between adjacent elements in Ds (: j), and selecting the top-ranked element Ds (k) from the adjacent element with the largest differencejJ) nearest neighbor radius r as samplejThen data message sample xjAverage distance Ava to all other samples in its neighborhoodjComprises the following steps:
wherein: k is a radical ofjRepresenting data message samples xjThe number of samples in the neighborhood of (1); obtaining the average distance from all data message samples to other samples in the neighboring domain, and calculating the average value Ava:
wherein: m is the number of data message samples; samples corresponding to the first element larger than Ava are searched from small to large in sequence on each column of the matrix Ds, and each data message sample x can be determinedjNearest neighbor sample x outside Avai:
Wherein: ds (h, j) is a data message sample xjDistance to h-1 sample nearest thereto; h is dist (x)i,xj) At Ds (: the corresponding serial number in j); calculating the times that each data message sample becomes the nearest sample outside the radius Ava of other samples, and recording the maximum value of the times as k; sorting the distances between all samples and the nearest kth data message sample to form a sequence { kd }, and calculating the difference value of two adjacent elements in the sequence to form a sequence { kdd }; by { kdd } sequenceAnalyzing the variation amplitude of the elements in the { kd } sequence, and finding out the mutation point sequence number e of the { kd } sequence, wherein the sequence number satisfies the following conditions:
the minimum sequence number e is selected from the sequence numbers meeting the formula, e is used as a parameter of a DBSCAN algorithm, k value is used as MinPts, the distances among samples are sequenced and searched, the distribution density of the samples is judged in a self-adaptive mode according to the sample distances, neighborhood parameters are determined, and therefore more accurate clustering of data messages is achieved.
Drawings
Fig. 1 is a schematic flow chart of a data message distribution method of the internet of things according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of data packet distribution equipment of the internet of things according to an embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Detecting a big flow in the data message transmission of the Internet of things by using a big flow detection algorithm based on a transmission rate, and performing network resource allocation scheduling by using a weighted polling scheduling strategy based on length limitation; and meanwhile, the abnormal data message of the Internet of things is detected by using an abnormal message detection algorithm based on density, and finally, the data message of the Internet of things is distributed by using a mixed distribution strategy. Referring to fig. 1, a schematic diagram of a data packet distribution method of the internet of things according to an embodiment of the present invention is shown.
In this embodiment, the method for distributing the data message of the internet of things includes:
s1, detecting the big flow in the data message transmission of the Internet of things by using a big flow detection algorithm based on the transmission rate.
Firstly, for data messages being transmitted in the Internet of things, the invention detects a large flow in the data message transmission of the Internet of things by using a large flow detection algorithm based on the transmission rate, wherein the large flow is a flow in which the number of the transmitted messages exceeds a certain number within a period of time;
the flow of the transmission rate-based large flow detection algorithm is as follows:
when a new message in the Internet of things arrives, extracting a head field of the message to generate a stream identifier, generating d indexes by using d different Hash functions by taking the stream identifier to be searched as input, and searching in corresponding Hash buckets in d sub-tables respectively;
if the corresponding stream entry is found, accumulating the byte number, and judging whether the data volume reaches the data volume threshold of the large stream, if so, calculating whether the stream rate reaches the rate threshold of the large stream;
if the flow item corresponding to the newly received message is not found, judging whether a free storage space exists or not from the Hash barrel with the numbered minimum sub-table;
if the free storage space exists, inserting an entry of a new flow into the free storage space, if the free storage space does not exist, finding out a flow min with the minimum transmission rate in a Hash bucket of d sub-tables corresponding to the current new flow, if the transmission rate of the data flow is smaller than a removal threshold, removing the flow, allocating the released space to the current new flow, and recording the generation time and the number of bytes of the new flow; if the rate of min exceeds the drop threshold, the new flow currently arriving is ignored.
And S2, for the detected transmission large flow of the Internet of things, carrying out transmission of the transmission large flow of the Internet of things by using a weighted polling scheduling strategy based on the transmission large flow with the length limit.
Further, for the detected internet of things transmission large flow, the invention utilizes a weighted polling scheduling strategy with length limitation to transmit the internet of things transmission large flow, wherein the weighted polling scheduling strategy with length limitation is as follows:
1) if N internet of things transmission large flows exist at the transmission port of the internet of things, the weight value of the internet of things transmission large flow i is wiThe query shares allocated to are longDegree Li,Lre_iTransmitting the remaining quota length of the big stream i served last time for the Internet of things;
2) divide a poll into T sub-polls, where T ═ max (w)i) In one sub-polling, when the Internet of things transmission stream i starts to send a message, if the total length of the message is less than or equal to Li+Lre_iWhen no new message arrives, all messages in the transmission stream i of the Internet of things are sent, and at the moment Lre_i0 bytes; if the total length of the message in the transmission large flow i of the Internet of things is greater than Li+Lre_iThen, n messages are sequentially sent from the head, wherein n satisfies:
wherein:
the length of the jth message counted from the head of a queue at the beginning of transmission in the large stream i transmitted by the Internet of things is obtained;
and when the residual share is not enough to send the (n + 1) th message, finishing the polling transmission, wherein the residual share is L're_iSatisfies the following conditions:
3) for the high-priority internet of things to transmit the large flow, the message in the high-priority internet of things is sent preferentially; and only when the transmission large stream of the high-priority Internet of things is empty, the transmission large stream of other Internet of things can be served.
S3, clustering the messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters.
Further, the invention utilizes a density-based clustering algorithm to cluster the messages, and the density-based clustering algorithm comprises the following processes:
1) for the collected data message sample x (SA, DA, MT, L), wherein SA represents the source address of the data message, DA represents the destination address of the data message, MT represents the type of the data message, and L is the length of the data message;
2) normalizing the data message sample attributes to obtain normalized data message sample attributes (SA, DA, MT, L)*(ii) a Calculating the distance between any two data message samples according to the normalized data message sample attributes, and storing the distance between the samples into an m-order matrix D, wherein m represents the number of the data message samples, and the data message sample distance calculation formula is as follows:
wherein:
xi,xjany two data message samples are obtained;
3) for each data message sample xjAfter sequencing the distance sequences of all other data message samples, combining the distance sequences into a new m-order matrix Ds:
wherein:
ds (i, j) represents data message sample xjDistance from the i-1 th data message sample closest thereto;
ds (1, j) represents data message sample xjDistance from itself;
calculating the difference between adjacent elements in Ds (: j), and selecting the top element Ds (k) from the adjacent element with the largest differencejJ) nearest neighbor radius r as samplejThen data message sample xjAverage distance Ava to all other samples in its neighborhoodjComprises the following steps:
wherein:
kjrepresenting data message samples xjThe number of samples in the neighborhood of (1);
obtaining the average distance from all data message samples to other samples in the neighboring domain, and calculating the average value Ava:
wherein:
m is the number of data message samples;
4) samples corresponding to the first element larger than Ava are searched from small to large in sequence on each column of the matrix Ds, and each data message sample x can be determinedjNearest neighbor sample x outside Avai:
Wherein:
ds (h, j) is a data message sample xjDistance to h-1 sample nearest thereto;
h is dist (x)i,xj) At Ds (: the corresponding serial number in j);
calculating the times that each data message sample becomes the nearest sample outside the radius Ava of other samples, and recording the maximum value of the times as k;
5) sorting the distances between all samples and the nearest kth data message sample to form a sequence { kd }, and calculating the difference value of two adjacent elements in the sequence to form a sequence { kdd }; analyzing the variation amplitude of the element in the { kd } through the { kdd } sequence, and finding out the mutation point sequence number e of the { kd } sequence, wherein the sequence number satisfies the following conditions:
selecting the minimum serial number e from the serial numbers satisfying the formula, taking e as a parameter of the DBSCAN algorithm, and taking the k value as MinPts; the DBSCAN algorithm judges a core object in the sample by using (e, k) and carries out cluster search;
after one cluster search of the DBSCAN algorithm is completed, if the number of the data message samples which are not classified exceeds 10% of the total number of samples, the data message samples which are not classified are taken as a set, clustering is started again from the near-neighbor region, so that misjudgment of noise samples caused by uneven density distribution is avoided, and K cluster clusters are obtained.
S4, constructing a corresponding SVM model by using the Internet of things data message cluster, and detecting abnormal data messages by using an abnormal data message detection algorithm combining the KNN and the SVM model.
Further, the invention utilizes the internet of things data message clustering cluster to construct a corresponding SVM model, thereby utilizing an abnormal data message detection algorithm combining KNN and the SVM model to detect abnormal data messages, wherein the abnormal data message detection algorithm combining KNN and the SVM model has the following flow:
1) calculating the distance between the data message x to be detected and other detected data messages, and finding out K adjacent data messages (x) closest to the data message to be detected1,x2,…,xk) Wherein K is the number of clustering clusters;
2) judging the cluster of the data message x to be detected by adopting a weighted voting mode, namely throwing a vote with a weight value for each adjacent data message as the type of the adjacent data message xiThe weight value possessed during voting is exp (-dist (x, x)i)),dist(x,xi) For neighbor data message xiThe distance from the data message x to be detected is that the vote weight thrown by the adjacent data message closer to the data message x to be detected is larger, and the type with the maximum total weight is the cluster C to which the data message to be detected belongsu;
3) Carrying out anomaly detection on the data message to be detected by utilizing an SVM model corresponding to the cluster, wherein the SVM model is composed of a corresponding cluster CuAll data messages containedDetermining, wherein m is a cluster CuThe number of the contained data messages is obtained by utilizing an SVM model, and the square of the radius is as follows:
wherein:
m is a cluster CuThe number of the contained data messages;
k () represents a gaussian kernel function;
4) calculating the square of the distance from the data message to be detected to the sphere center of the SVM hypersphere as:
if d is2(x) If the data message to be detected is larger than R, the data message to be detected is positioned outside the hypersphere, the data message to be detected is an abnormal data message, and otherwise, the data message to be detected is a normal data message; and sending error information to the source address of the data message for the detected abnormal data message.
And S5, distributing the data message of the Internet of things by using a mixed distribution strategy.
Further, the invention realizes the distribution of the data message of the internet of things by using a mixed distribution strategy, wherein the flow of the mixed distribution strategy is as follows:
initializing a mixed distribution table, filling periodic data messages in the Internet of things, wherein the periodic messages are filled from the position of the upper left corner in the mixed distribution table, the arrangement mode of the messages is filled in sequence from small to large according to the receiving and sending period of the messages, the messages with the smaller receiving and sending period are arranged in the earlier basic time block, and the operation is repeated until all the periodic messages are filled in the mixed distribution table;
the periodic message is sent according to the basic time block specified in the mixed distribution table in advance, and if the event message needs to be sent at the moment, the periodic message can quickly find out the nearest basic time block to fill in the event message and send the event message.
The following describes embodiments of the present invention through an algorithmic experiment and tests of the inventive treatment method. The hardware test environment of the algorithm of the invention is as follows: the operating system is Ubuntu16.04, the computer processor is Inteli5-8500 CPU @3GHZ multiplied by 6, the size of the memory bank is 16G, and the software is matlab; the comparison processing method is an internet of things data message distribution method based on self-adaptive caching and an internet of things data message distribution method based on a Hadoop database.
In the algorithm experiment of the invention, a data set is 10000 collected data messages. In the experiment, the acquired data is input into the method and the comparison method, and the time required for completing the data message distribution is used as an index for evaluating the performance of the algorithm.
According to the experimental result, the data message distribution time of the Internet of things data message distribution method based on the self-adaptive cache is 11.2s, the data message distribution time of the Internet of things data message distribution method based on the Hadoop database is 12.9s, the data message distribution time of the Internet of things data message distribution method is 10.01s, and compared with a comparison method, the Internet of things data message distribution method provided by the invention has lower data message distribution time.
The invention also provides equipment for distributing the data message of the Internet of things. Fig. 2 is a schematic diagram of an internal structure of an internet of things data message distribution device according to an embodiment of the present invention.
In this embodiment, the internet-of-things data message distribution device 1 at least includes a data message obtaining apparatus 11, a data message processor 12, a data message distribution apparatus 13, a communication bus 14, and a network interface 15.
The data packet acquiring device 11 may be a Personal Computer (PC), a terminal device such as a smart phone, a tablet Computer, and a portable Computer, or a server.
The data message processor 12 includes at least one type of readable storage medium including flash memory, hard disks, multi-media cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, and the like. The datagram processor 12 may in some embodiments be an internal storage unit of the internet of things datagram delivery device 1, for example a hard disk of the internet of things datagram delivery device 1. The data message processor 12 may also be an external storage device of the data message distribution device 1 of the internet of things in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are equipped on the data message distribution device 1 of the internet of things. Further, the datagram processor 12 may also include both an internal storage unit of the internet-of-things datagram delivery device 1 and an external storage device. The datagram processor 12 may be used not only to store application software installed in the internet-of-things datagram delivery device 1 and various types of data, but also to temporarily store data that has been output or is to be output.
Data message distribution device 13 may, in some embodiments, be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip that runs program code stored in data message processor 12 or processes data, such as data message distribution program instructions.
The communication bus 14 is used to enable connection communication between these components.
The network interface 15 may optionally comprise a standard wired interface, a wireless interface (e.g. WI-FI interface), typically used for establishing a communication connection between the device 1 and other electronic devices.
Optionally, the device 1 may further comprise a user interface, which may comprise a Display (Display), an input unit such as a Keyboard (Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or a display unit, is used to display information processed in the internet of things data message distribution device 1 and to display a visual user interface.
Fig. 2 shows only the internet of things data message distribution device 1 with components 11-15, and it will be understood by those skilled in the art that the structure shown in fig. 1 does not constitute a limitation of the internet of things data message distribution device 1, and may include fewer or more components than shown, or some components in combination, or a different arrangement of components.
In the embodiment of the device 1 shown in fig. 2, the data message processor 12 has stored therein data message distribution program instructions; the steps of the data packet distribution device 13 executing the data packet distribution program instructions stored in the data packet processor 12 are the same as the implementation method of the data packet distribution method of the internet of things, and are not described here.
Furthermore, an embodiment of the present invention also provides a computer-readable storage medium having stored thereon data packet distribution program instructions executable by one or more processors to implement the following operations:
detecting a big flow in the data message transmission of the Internet of things by using a big flow detection algorithm based on a transmission rate;
for the detected internet of things transmission large flow, the transmission of the internet of things transmission large flow is carried out by utilizing a weighted polling scheduling strategy with length limitation;
clustering messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters;
constructing a corresponding SVM model by using the Internet of things data message clustering cluster, and detecting abnormal data messages by using an abnormal data message detection algorithm combining KNN and the SVM model;
and realizing the distribution of the data message of the Internet of things by using a mixed distribution strategy.
It should be noted that the above-mentioned numbers of the embodiments of the present invention are merely for description, and do not represent the merits of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (8)
1. A method for distributing data messages of the Internet of things is characterized by comprising the following steps:
detecting a big flow in the data message transmission of the Internet of things by using a big flow detection algorithm based on a transmission rate;
for the detected internet of things transmission large flow, the transmission of the internet of things transmission large flow is carried out by utilizing a weighted polling scheduling strategy with length limitation;
clustering messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters;
constructing a corresponding SVM model by using the Internet of things data message clustering cluster, and detecting abnormal data messages by using an abnormal data message detection algorithm combining KNN and the SVM model;
and realizing the distribution of the data message of the Internet of things by using a mixed distribution strategy.
2. The internet of things data message distribution method according to claim 1, wherein the transmission rate-based large flow detection algorithm flow is as follows:
when a new message in the Internet of things arrives, extracting a head field of the message to generate a stream identifier, generating d indexes by using d different Hash functions by taking the stream identifier to be searched as input, and searching in corresponding Hash buckets in d sub-tables respectively;
if the corresponding stream entry is found, accumulating the byte number, and judging whether the data volume reaches the data volume threshold of the large stream, if so, calculating whether the stream rate reaches the rate threshold of the large stream;
if the flow item corresponding to the newly received message is not found, judging whether a free storage space exists or not from the Hash barrel with the numbered minimum sub-table;
if the free storage space exists, inserting an entry of a new flow into the free storage space, if the free storage space does not exist, finding out a flow min with the minimum transmission rate in a Hash bucket of d sub-tables corresponding to the current new flow, if the transmission rate of the data flow is smaller than a removal threshold, removing the flow, allocating the released space to the current new flow, and recording the generation time and the number of bytes of the new flow; if the rate of min exceeds the drop threshold, the new flow currently arriving is ignored.
3. The internet of things data message distribution method according to claim 2, wherein the transmission of the internet of things transmission stream is performed by using a weighted round-robin scheduling strategy based on a length constraint.
1) If N internet of things transmission large flows exist at the transmission port of the internet of things, the weight value of the internet of things transmission large flow i is wiThe length of the allocated query share is Li,Lre_iTransmitting the remaining quota length of the big stream i served last time for the Internet of things;
2) divide a poll into T sub-polls, where T ═ max (w)i) In one sub-polling, when the Internet of things transmission stream i starts to send a message, if the total length of the message is less than or equal to Li+Lre_iWhen no new message arrives, all messages in the transmission stream i of the Internet of things are sent, and at the moment Lre_i0 bytes; if the total length of the message in the transmission large flow i of the Internet of things is greater than Li+Lre_iThen, n messages are sequentially sent from the head, wherein n satisfies:
wherein:
the length of the jth message counted from the head of a queue at the beginning of transmission in the large stream i transmitted by the Internet of things is obtained;
and when the residual share is not enough to send the (n + 1) th message, finishing the polling transmission, wherein the residual share is L're_iSatisfies the following conditions:
3) for the high-priority internet of things to transmit the large flow, the message in the high-priority internet of things is sent preferentially; and only when the transmission large stream of the high-priority Internet of things is empty, the transmission large stream of other Internet of things can be served.
4. The internet of things data message distribution method of claim 3, wherein the clustering of messages by using a density-based clustering algorithm comprises:
1) for the collected data message sample x (SA, DA, MT, L), wherein SA represents the source address of the data message, DA represents the destination address of the data message, MT represents the type of the data message, and L is the length of the data message;
2) normalizing the data message sample attributes to obtain normalized data message sample attributes (SA, DA, MT, L)*(ii) a Calculating the distance between any two data message samples according to the normalized data message sample attributes, and storing the distance between the samples into an m-order matrix D, wherein m represents the number of the data message samples, and the data message sample distance calculation formula is as follows:
wherein:
xi,xjany two data message samples are obtained;
3) for each data message sample xjAfter sequencing the distance sequences of all other data message samples, combining the distance sequences into a new m-order matrix Ds:
wherein:
ds (i, j) represents data message sample xjThe (i-1) th data message sample closest to the (i-1) th data message sampleThe distance of the book;
ds (1, j) represents data message sample xjDistance from itself;
calculating the difference between adjacent elements in Ds (: j), and selecting the top element Ds (k) from the adjacent element with the largest differencejJ) nearest neighbor radius r as samplejThen data message sample xjAverage distance Ava to all other samples in its neighborhoodjComprises the following steps:
wherein:
kjrepresenting data message samples xjThe number of samples in the neighborhood of (1);
obtaining the average distance from all data message samples to other samples in the neighboring domain, and calculating the average value Ava:
wherein:
m is the number of data message samples;
4) samples corresponding to the first element larger than Ava are searched from small to large in sequence on each column of the matrix Ds, and each data message sample x can be determinedjNearest neighbor sample x outside Avai:
Wherein:
ds (h, j) is a data message sample xjDistance to h-1 sample nearest thereto;
h is dist (x)i,xj) The corresponding serial number in Ds (: j);
calculating the times that each data message sample becomes the nearest sample outside the radius Ava of other samples, and recording the maximum value of the times as k;
5) sorting the distances between all samples and the nearest kth data message sample to form a sequence { kd }, and calculating the difference value of two adjacent elements in the sequence to form a sequence { kdd }; analyzing the variation amplitude of the element in the { kd } through the { kdd } sequence, and finding out the mutation point sequence number e of the { kd } sequence, wherein the sequence number satisfies the following conditions:
selecting the minimum serial number e from the serial numbers satisfying the formula, taking e as a parameter of the DBSCAN algorithm, and taking the k value as MinPts; the DBSCAN algorithm judges a core object in the sample by using (e, k) and carries out cluster search;
after one cluster search of the DBSCAN algorithm is completed, if the number of the data message samples which are not classified exceeds 10% of the total number of samples, clustering is started again from the calculation of the near neighborhood by taking the data message samples which are not classified as a set, and K clustering clusters are obtained.
5. The internet of things data message distribution method according to claim 4, wherein the detecting abnormal data messages by using an abnormal data message detection algorithm combining KNN and SVM models comprises:
1) calculating the distance between the data message x to be detected and other detected data messages, and finding out K adjacent data messages (x) closest to the data message to be detected1,x2,...,xk) Wherein K is the number of clustering clusters;
2) judging the cluster of the data message x to be detected by adopting a weighted voting mode, namely throwing a vote with a weight value for each adjacent data message as the type of the adjacent data message xiThe weight value possessed during voting is exp (-dist (x, x)i)),dist(x,xi) For neighbor data message xiThe distance from the data message x to be detected, the closer the adjacent data message to the data message x to be detected is, the vote right thrown by the adjacent data messageThe larger the value is, the type with the maximum total weight is the cluster C to which the data message to be detected belongsu;
3) Carrying out anomaly detection on the data message to be detected by utilizing an SVM model corresponding to the cluster, wherein the SVM model is composed of a corresponding cluster CuAll data messages containedDetermining, wherein m is a cluster CuThe number of the contained data messages is obtained by utilizing an SVM model, and the square of the radius is as follows:
wherein:
m is a cluster CuThe number of the contained data messages;
k () represents a gaussian kernel function;
4) calculating the square of the distance from the data message to be detected to the sphere center of the SVM hypersphere as:
if d is2(x) If the data message to be detected is larger than R, the data message to be detected is positioned outside the hypersphere, the data message to be detected is an abnormal data message, and otherwise, the data message to be detected is a normal data message;and sending error information to the source address of the data message for the detected abnormal data message.
6. The internet of things data message distribution method of claim 5, wherein the hybrid distribution strategy is:
initializing a mixed distribution table, filling periodic data messages in the Internet of things, wherein the periodic messages are filled from the position of the upper left corner in the mixed distribution table, the arrangement mode of the messages is filled in sequence from small to large according to the receiving and sending period of the messages, the messages with the smaller receiving and sending period are arranged in the earlier basic time block, and the operation is repeated until all the periodic messages are filled in the mixed distribution table;
the periodic message is sent according to the basic time block specified in the mixed distribution table in advance, and if the event message needs to be sent at the moment, the periodic message can quickly find out the nearest basic time block to fill in the event message and send the event message.
7. An internet of things data message distribution device, the device comprising:
the data message acquisition device is used for detecting a large flow in the transmission of the data message of the Internet of things by using a large flow detection algorithm based on the transmission rate, and transmitting the large flow in the transmission of the Internet of things by using a weighted polling scheduling strategy based on length limitation, so as to obtain the data message in the large flow in the transmission of the Internet of things;
the data message processor is used for clustering messages by using a density-based clustering algorithm to obtain a plurality of Internet of things data message clustering clusters, constructing a corresponding SVM model by using the Internet of things data message clustering clusters, and detecting abnormal data messages by using an abnormal data message detection algorithm combining the KNN and the SVM models;
and the data message distribution device is used for realizing the distribution of the data message of the Internet of things by utilizing a hybrid distribution strategy.
8. A computer readable storage medium having stored thereon data message distribution program instructions executable by one or more processors to perform the steps of a method for implementing internet of things data message distribution as claimed in any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110250610.7A CN112968814A (en) | 2021-03-08 | 2021-03-08 | Internet of things data message distribution method and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110250610.7A CN112968814A (en) | 2021-03-08 | 2021-03-08 | Internet of things data message distribution method and equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112968814A true CN112968814A (en) | 2021-06-15 |
Family
ID=76276843
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110250610.7A Pending CN112968814A (en) | 2021-03-08 | 2021-03-08 | Internet of things data message distribution method and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112968814A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114724069A (en) * | 2022-04-09 | 2022-07-08 | 北京天防安全科技有限公司 | Video equipment model confirming method, device, equipment and medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6401126B1 (en) * | 1999-03-10 | 2002-06-04 | Microsoft Corporation | File server system and method for scheduling data streams according to a distributed scheduling policy |
CN101834763A (en) * | 2010-06-25 | 2010-09-15 | 山东大学 | Multiple-category large-flow parallel measuring method under high speed network environment |
CN108965160A (en) * | 2018-06-29 | 2018-12-07 | 北京邮电大学 | Dynamic difference Weight Round Robin method suitable for mine Internet of Things multi-service transport |
-
2021
- 2021-03-08 CN CN202110250610.7A patent/CN112968814A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6401126B1 (en) * | 1999-03-10 | 2002-06-04 | Microsoft Corporation | File server system and method for scheduling data streams according to a distributed scheduling policy |
CN101834763A (en) * | 2010-06-25 | 2010-09-15 | 山东大学 | Multiple-category large-flow parallel measuring method under high speed network environment |
CN108965160A (en) * | 2018-06-29 | 2018-12-07 | 北京邮电大学 | Dynamic difference Weight Round Robin method suitable for mine Internet of Things multi-service transport |
Non-Patent Citations (4)
Title |
---|
佟为明等: "基于报文信息标签的智能变电站通信服务策略", 《电力系统自动化》 * |
王洪波等: "基于LRU的大流检测算法", 《电子与信息学报》 * |
王风宇等: "一种高效率的大流提取方法", 《计算机研究与发展》 * |
高吉星: "《基于报文传输的智能变电站通信性能保障方法研究》", 《中国优秀博硕士学位论文全文数据库(博士)工程科技Ⅱ辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114724069A (en) * | 2022-04-09 | 2022-07-08 | 北京天防安全科技有限公司 | Video equipment model confirming method, device, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11122058B2 (en) | System and method for the automated detection and prediction of online threats | |
CN109617996B (en) | File uploading and downloading method, server and computer readable storage medium | |
CN109450900B (en) | Mimicry judgment method, device and system | |
CN112328399A (en) | Cluster resource scheduling method and device, computer equipment and storage medium | |
CN108270618A (en) | Alert the method, apparatus and warning system of judgement | |
CN105956183A (en) | Method and system for multi-stage optimization storage of a lot of small files in distributed database | |
CN105354251B (en) | Electric power cloud data management indexing means based on Hadoop in electric system | |
CN108429701A (en) | network acceleration system | |
CN104618304B (en) | Data processing method and data handling system | |
CN110428124A (en) | Method for allocating tasks, task allocation apparatus, storage medium and computer equipment | |
CN109474542A (en) | Message pushing request flow control method, device and medium based on business rules | |
CN109905328B (en) | Data stream identification method and device | |
He et al. | Priority queue with customer upgrades | |
CN109241084A (en) | Querying method, terminal device and the medium of data | |
CN109766441A (en) | File classification method, apparatus and system | |
CN109492118A (en) | A kind of data detection method and detection device | |
CN104333483A (en) | Identification method, system and identification device for internet application flow | |
CN112968814A (en) | Internet of things data message distribution method and equipment | |
CN108259544B (en) | UR L query method and UR L query server | |
CN112804374B (en) | Threat IP identification method, threat IP identification device, threat IP identification equipment and threat IP identification medium | |
CN110414591A (en) | A kind of data processing method and equipment | |
CN108093047B (en) | Data sending method and device, electronic equipment and middleware system | |
CN117294497A (en) | Network traffic abnormality detection method and device, electronic equipment and storage medium | |
CN108810916B (en) | Wi-Fi hotspot recommendation method and device and storage medium | |
CN109981460A (en) | A kind of service-oriented converged network, calculating, storage integral method and device |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210615 |