CN117411806A - Power communication network performance evaluation method, system, equipment and storage medium - Google Patents
Power communication network performance evaluation method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN117411806A CN117411806A CN202311705506.8A CN202311705506A CN117411806A CN 117411806 A CN117411806 A CN 117411806A CN 202311705506 A CN202311705506 A CN 202311705506A CN 117411806 A CN117411806 A CN 117411806A
- Authority
- CN
- China
- Prior art keywords
- packet loss
- matrix
- link
- network
- network state
- 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.)
- Granted
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 59
- 238000004891 communication Methods 0.000 title claims abstract description 58
- 239000011159 matrix material Substances 0.000 claims abstract description 187
- 238000011084 recovery Methods 0.000 claims abstract description 154
- 238000000034 method Methods 0.000 claims abstract description 69
- 230000000694 effects Effects 0.000 claims abstract description 47
- 238000001514 detection method Methods 0.000 claims abstract description 45
- 230000004927 fusion Effects 0.000 claims abstract description 24
- 238000005516 engineering process Methods 0.000 claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 239000000523 sample Substances 0.000 claims description 25
- 230000005540 biological transmission Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 230000000977 initiatory effect Effects 0.000 claims description 12
- 230000001010 compromised effect Effects 0.000 claims description 4
- 238000012549 training Methods 0.000 description 16
- 238000004422 calculation algorithm Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 12
- 238000012544 monitoring process Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000004043 dyeing Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000013256 Gubra-Amylin NASH model Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
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
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0663—Performing the actions predefined by failover planning, e.g. switching to standby network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0677—Localisation of faults
-
- 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
- H04L43/04—Processing captured monitoring data, e.g. for logfile generation
-
- 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
- H04L43/0829—Packet loss
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/31—Flow control; Congestion control by tagging of packets, e.g. using discard eligibility [DE] bits
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Mining & Analysis (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a power communication network performance evaluation method, a system, equipment and a storage medium, wherein the method is that telemetry information acquired by an INT detection packet initiated by a fusion terminal is uploaded to a master station platform for analysis statistics, after a full-network queue depth matrix and a link packet loss statistics result are obtained, packet loss link data recovery is carried out on an anti-network according to the link packet loss statistics result, the full-network queue depth matrix and a pre-built generation type anti-network to obtain a network state recovery matrix, recovery effect evaluation is carried out on each packet loss link according to the network state recovery matrix to generate a link set to be detected, each packet loss link in the link set to be detected is detected according to a source routing technology, and the network state recovery matrix is updated according to the acquired link detection data to obtain a network state evaluation result. The invention can reflect the depth condition of the whole network queue in real time and effectively recover lost data, thereby providing guarantee for the stable and reliable operation of the communication network.
Description
Technical Field
The present invention relates to the field of wireless communications technologies, and in particular, to a method, a system, a computer device, and a storage medium for evaluating performance of a power communication network based on current-following monitoring and active detection.
Background
The power communication network is an important component of the power system, and the operation, control and management of the power system provide necessary information transmission means. With the complexity and scale of the power system, the performance requirements of the power communication network are also higher and higher. In order to ensure stable and reliable operation of the power system, real-time monitoring and management of the power communication network becomes particularly important.
Network state monitoring methods of existing power communication networks can be divided into two categories: passive monitoring techniques and active probe techniques. In-band network telemetry (In-band Network Telemetry, INT) is the most widely used passive monitoring technique, and by embedding telemetry information In a data packet, network equipment is allowed to collect network status information when forwarding the data packet, and real-time information about network delay, packet loss rate and other key performance indicators can be provided; the active probe technology collects network performance information by sending specific data packets to the network and observing the behavior of the data packets, and can provide more detailed network performance information while increasing network load. Although the prior art can perform a certain monitoring function on the power communication network, there are still some application limitations of the prior art: in-band network telemetry may not provide queue depth information for the whole network and exists depending on traffic flows, and when packet loss occurs in traffic flows, the INT technique is prone to misleading; active probe technology may generate additional load on the network, and the lost data cannot be effectively recovered when the network packet loss occurs. That is, the existing network state monitoring method still cannot truly meet the real-time detection and feedback requirements of the power grid communication system, and fault cause diagnosis and fault location positioning cannot be performed when a network fails.
Disclosure of Invention
The invention aims to provide a power communication network performance evaluation method, which combines in-band network telemetry technology with active probe technology, realizes lost data recovery of different types of packet loss links by using a generated countermeasure network (GAN) while comprehensively perceiving the state of a power communication network in real time, and performs active packet sending detection under the condition that the in-band network telemetry data recovery is not reliable enough, so that the application defects that the current power communication network state monitoring cannot meet the real-time detection and lost data recovery of a power network communication system are overcome, the depth condition of a whole network queue can be reflected in real time, various lost data can be effectively recovered, the real-time reliable feedback tracking of the whole network state is realized, and the guarantee is provided for the stable and reliable operation of a communication network.
In order to achieve the above object, it is necessary to provide a power communication network performance evaluation method, system, computer device, and storage medium in order to solve the above technical problems.
In a first aspect, an embodiment of the present invention provides a method for evaluating performance of a power communication network, the method including the steps of:
an INT detection packet is initiated to acquire telemetry information through a fusion terminal of an electric power communication network, and the telemetry information is uploaded to a master station platform;
Analyzing and counting the telemetry information through the master station platform to obtain a full-network queue depth matrix and a link packet loss statistical result;
carrying out packet loss link data recovery according to the link packet loss statistical result, the full network queue depth matrix and a pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix;
performing recovery effect evaluation on each packet loss link according to the network state recovery matrix to generate a link set to be recovered by the probe;
and initiating an active detection packet to detect each packet loss link in the link set to be detected through the fusion terminal according to a source routing technology, acquiring corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result.
Further, the link packet loss statistical result comprises packet loss type and packet loss positioning result of each packet loss link; the packet loss type comprises black hole packet loss, congestion packet loss and random packet loss; the packet loss positioning result comprises packet loss time and packet loss position;
analyzing and counting the telemetry information through the master station platform, and acquiring a link packet loss statistical result comprises the following steps:
When link packet loss exists, acquiring an INT transmission flow count and an INT reception flow count, and judging whether each link packet loss is a black hole packet loss according to the INT transmission flow count and the INT reception flow count;
if the packet loss is not the black hole packet loss, acquiring corresponding continuous packet loss duration and continuous packet loss length, and judging whether the packet loss is congestion packet loss according to the continuous packet loss duration and the continuous packet loss length;
and if the congestion packet loss is judged, judging the corresponding link packet loss as random packet loss.
Further, the step of determining whether each link packet loss is a black hole packet loss according to the INT transmit traffic count and the INT receive traffic count includes:
according to the INT sending flow count and the INT receiving flow count, respectively obtaining a corresponding sending flow increment and a corresponding receiving flow increment;
and calculating an increment difference value of the sending flow increment and the receiving flow increment, judging whether the increment difference value is larger than or equal to a preset difference value threshold, and if so, judging the corresponding link packet loss as black hole packet loss.
Further, the step of determining whether the packet loss is congestion according to the duration of continuous packet loss and the length of continuous packet loss includes:
Judging whether the continuous packet loss duration and the continuous packet loss length meet preset congestion packet loss conditions or not; the preset congestion packet loss condition is expressed as:
wherein,representing continuous packet loss duration; />Representing the continuous packet loss length; />And->Respectively representing a duration threshold and a packet loss length threshold;
if so, judging the corresponding link packet loss as congestion packet loss.
Further, the step of recovering the lost packet link data according to the link packet loss statistical result, the full network queue depth matrix and the pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix includes:
obtaining a corresponding congestion packet loss marking matrix, a black hole packet loss marking matrix and a network state mask matrix according to the link packet loss statistical result;
acquiring a network topology matrix of the power communication network, and inputting the network topology matrix, the congestion packet loss marking matrix, the network state mask matrix and the whole network queue depth matrix into a generator model in the generating type countermeasure network to recover packet loss data, so as to obtain a corresponding first network state recovery matrix;
obtaining a corresponding second network state recovery matrix according to the first network state recovery matrix, the full-network queue depth matrix and the network state mask matrix;
And obtaining the network state recovery matrix according to the second network state recovery matrix, the full network queue depth matrix and the black hole packet loss marking matrix.
Further, the network state recovery matrix is expressed as:
in the method, in the process of the invention,
wherein,representing a network state recovery matrix; />Representing a second network state recovery matrix; />Representing a first network state recovery matrix; />Representing a full-network queue depth matrix; />Representing a black hole packet loss marking matrix;NaNrepresenting a black hole packet loss filling value; />Represents a network topology matrix, and->When the link isiAndjwhen passing through the same switch, the switch is compromised>Otherwise；/>Marking matrix for indicating congestion packet loss>Is>A column; />Expressed in the process->Is>When in column, only the link with congestion packet loss needs to be considered, and other links passing through the same switch with the link; />Representing a network state mask matrix; />Representing the multiplication of the corresponding elements;Vrepresenting randomly generated noise subject to gaussian distribution.
Further, the step of evaluating the recovery effect of each packet loss link according to the network state recovery matrix and generating the link set to be recovered by the probe includes:
inputting the network state recovery matrix and the network state mask matrix into a discriminator model in the generated countermeasure network to perform network state mask matrix estimation to obtain a corresponding network state mask estimation matrix;
Obtaining recovery effect evaluation values of each packet loss link according to the network state mask matrix and the network state mask estimation matrix; the recovery effect evaluation value is expressed as:
wherein,indicate->An evaluation value of recovery effect of each packet loss link; />And->Respectively represent network state masksModulo matrix->And a network status mask estimation matrix->Is>A row;
and judging whether the recovery effect evaluation value of each packet loss link is larger than a preset effect threshold value, if so, adding the corresponding packet loss link serial number into the to-be-probed recovery link set.
In a second aspect, an embodiment of the present invention provides a power communication network performance evaluation system, the system including:
the remote sensing acquisition module is used for initiating an INT detection packet to acquire remote sensing information through a fusion terminal of the power communication network and uploading the remote sensing information to the master station platform;
the information analysis module is used for carrying out analysis statistics on the telemetry information through the master station platform to obtain a full-network queue depth matrix and a link packet loss statistical result;
the data recovery module is used for carrying out packet loss link data recovery according to the link packet loss statistical result, the full-network queue depth matrix and a pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix;
The effect evaluation module is used for evaluating the recovery effect of each packet loss link according to the network state recovery matrix and generating a link set to be recovered by the probe;
and the state evaluation module is used for initiating an active detection packet to detect each packet loss link in the link set to be detected through the fusion terminal according to a source routing technology, collecting corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result.
In a third aspect, embodiments of the present invention further provide a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when executing the computer program.
In a fourth aspect, embodiments of the present invention also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the above method.
The method comprises the steps of initiating INT detection packets to collect telemetry information through a fusion terminal of an electric power communication network, uploading the telemetry information to a master station platform, analyzing and counting the telemetry information through the master station platform, recovering packet loss link data according to a link packet loss statistical result, the whole network queue depth matrix and a pre-established generation type anti-network to obtain a corresponding network state recovery matrix, evaluating recovery effects of all packet loss links according to the network state recovery matrix to generate a to-be-detected recovery link set, initiating active detection packets to detect all packet loss links in the to-be-detected recovery link set through the fusion terminal according to a source routing technology, collecting corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result. Compared with the prior art, the power communication network performance evaluation method can acquire the depth condition of the whole network queue in real time, can effectively recover lost data by adopting different data recovery strategies according to different packet loss types, and actively transmits packets to detect links with poor recovery data effect under the condition that the in-band network telemetry data recovery is not reliable enough, so that real-time reliable feedback of the whole network state is realized, and guarantee is provided for stable and reliable operation of a communication network.
Drawings
Fig. 1 is a schematic diagram of an application scenario of a power communication network performance evaluation method in an embodiment of the present invention;
FIG. 2 is a flow chart of a method for evaluating performance of a power communication network according to an embodiment of the present invention;
fig. 3 is an application schematic diagram of packet loss positioning by adopting a multi-bit cyclic marking method in an alternate dyeing method in the embodiment of the invention;
fig. 4 is a schematic diagram showing the comparison of application performances of the full network data recovery method (considering GAN of surrounding links) and GAN algorithm and KNN algorithm without considering surrounding links according to the present invention based on fixed random packet loss rate and preset effect threshold in the embodiment of the present invention under different congestion levels;
fig. 5 is a schematic diagram showing the comparison of the application performance of the full network data recovery and active detection combination algorithm (GAN-SR method) of the present invention and the GAN algorithm considering surrounding links, and the INT method not using data recovery, based on the fixed random packet loss rate and the preset effect threshold, under different congestion link duty ratios in the embodiment of the present invention;
FIG. 6 is a schematic diagram of a power communication network performance evaluation system according to an embodiment of the present invention;
fig. 7 is an internal structural view of a computer device in an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantageous effects of the present application more apparent, the present invention will be further described in detail with reference to the accompanying drawings and examples, and it should be understood that the examples described below are only illustrative of the present invention and are not intended to limit the scope of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The method for evaluating the performance of the power communication network can be understood as a full-network state real-time reliable feedback tracking method for carrying out link detection by adopting an active probe method under the condition that the in-band network telemetry data recovery is not reliable enough, and can be applied to a terminal or a server as shown in figure 1. The terminal may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers and portable wearable devices, and the server may be implemented by a separate server or a server cluster formed by a plurality of servers. The server can carry out efficient and reliable network performance evaluation by adopting the power communication network performance evaluation method provided by the invention according to actual application requirements, and the obtained network performance evaluation is used for subsequent research of the server or is transmitted to the terminal for the terminal user to check and analyze; the following examples will explain the power communication network performance evaluation method of the present invention in detail.
In one embodiment, as shown in fig. 2, there is provided a power communication network performance evaluation method, including the steps of:
s11, an INT detection packet is initiated to acquire telemetry information through a fusion terminal of an electric power communication network, and the telemetry information is uploaded to a master station platform; the fusion terminal can be understood as network terminal equipment for generating and sending service data packets in the power communication network, and can be a tablet personal computer, a PAD or a smart phone, and the corresponding master station platform can be understood as a terminal or a server for carrying out centralized management analysis on service data transferred by each fusion terminal through a local network and an operator network; the method for the fusion terminal to initiate the INT detection packet to collect the telemetry information in the embodiment can be understood as providing support of in-band network telemetry INT technology for the fusion terminal of the power communication network, initiating the INT detection information from the fusion terminal along with the service data packet, and uploading the telemetry information collected in the transmission process to the master station platform in the last hop of the network; it should be noted that, the specific type of transmitting the service data packet and the sending of the INT detection packet may be implemented in combination with the prior art according to practical application requirements, and telemetry information collected in practical application may include information such as an ingress and egress timestamp, an ingress and egress port, identification information of the switch, a queue number, a sending byte, and a queue depth of the service data packet in the switch, which are not described herein again.
S12, analyzing and counting the telemetry information through the master station platform to obtain a full-network queue depth matrix and a link packet loss statistical result; the full-network queue depth matrix can be understood as each link queue depth matrix generated by tracking the queue length of each service data packet forwarding path based on the INT technology, and the specific acquisition process is realized by referring to the prior art and is not described herein; the acquisition of the full-network queue depth matrix using the INT technique can be expressed as:
wherein,representing link->In the transmission of +.>Queue depth when the service data packets are received; />Representing the number of links involved in the whole network; />A telemetry packet number representing a single link; due to the packet loss phenomenon, the method comprises the following steps of->May be missing;
after receiving service data packets with telemetry information, the master station platform counts and positions the loss condition of INT detection packets of each link, and sequentially evaluates the loss type of each service data packet of each link according to the investigation sequence of black hole packet loss, congestion packet loss and random packet loss; correspondingly, the link packet loss statistical result comprises packet loss types and packet loss positioning results of each packet loss link; the packet loss type comprises black hole packet loss, congestion packet loss and random packet loss; the packet loss positioning result comprises packet loss time and packet loss position; specifically, the step of analyzing and counting the telemetry information through the master station platform and obtaining the link packet loss statistical result comprises the following steps:
When link packet loss exists, acquiring an INT transmission flow count and an INT reception flow count, and judging whether each link packet loss is a black hole packet loss according to the INT transmission flow count and the INT reception flow count; the INT transmit traffic count and the INT receive traffic count may be respectively obtained by corresponding traffic match counters, which are not described in detail herein; specifically, the step of determining whether each link packet loss is a black hole packet loss according to the INT transmit traffic count and the INT receive traffic count includes:
according to the INT sending flow count and the INT receiving flow count, respectively obtaining a corresponding sending flow increment and a corresponding receiving flow increment; the sending flow increment and the receiving flow increment can be obtained by respectively making differences between the current flow matching counter value and the flow matching counter value corresponding to the previous period;
calculating an increment difference value of the sending flow increment and the receiving flow increment, judging whether the increment difference value is larger than or equal to a preset difference value threshold value, and if so, judging the corresponding link packet loss as black hole packet loss; the delta value is understood as the lost flow value in the transmission flow transfer process, and when the lost flow exceeds a certain threshold, the lost flow value is considered to belong to the black hole packet loss, namely, when the transmission flow delta (INT source) is obtained And receive traffic delta (INT sink)>The method comprises the following steps:
marking the packet loss type as black hole packet loss; wherein,representing a corresponding preset difference threshold value,can be selected according to the actual application requirements, and is not particularly limited herein;
after the packet loss link with the packet loss type of black hole packet loss is determined by the method, the depth matrix of the full network queue can be obtainedXCorresponding black hole packet loss marking matrixThe method comprises the following steps:
wherein,representing link->In the transmission of +.>The loss type of each service data packet is a mark of whether the black hole is lost, if yes, the data packet is +.>Lost and is when the black hole is lost, +.>1, the rest are 0;
if the packet loss is not the black hole packet loss, acquiring corresponding continuous packet loss duration and continuous packet loss length, and judging whether the packet loss is congestion packet loss according to the continuous packet loss duration and the continuous packet loss length; the duration of continuous packet loss can be understood as the time difference between the beginning and the end of the continuous packet loss event; the corresponding continuous packet loss length can be understood as the sequence number difference of the beginning and the end of the continuous packet loss event; specifically, the step of judging whether the congestion packet loss exists according to the continuous packet loss duration and the continuous packet loss length includes:
Judging whether the continuous packet loss duration and the continuous packet loss length meet preset congestion packet loss conditions or not; the preset congestion packet loss condition is expressed as:
wherein,representing continuous packet loss duration; />Representing the continuous packet loss length; />And->The duration threshold and the packet loss length threshold can be set according to actual application requirements, and are not particularly limited herein;
if yes, judging the corresponding link packet loss as congestion packet loss;
after determining the packet loss link with the packet loss type being congestion packet loss by the method, the depth matrix of the queue of the whole network can be obtainedXCorresponding congestion packet loss marking matrixThe method comprises the following steps:
wherein,representing link->In the transmission of +.>The loss type of each service data packet is a congestion loss mark, if the data packet is + ->Lost and congestion lost ++>1, the rest are 0;
if the congestion packet loss is judged, judging the corresponding link packet loss as random packet loss; wherein, random packet loss can be understood as scattered discontinuous packet loss, and after confirming that the packet loss does not belong to the black hole packet loss and congestion packet loss types, the random packet loss can be regarded as random packet loss of the network;
in addition, the embodiment can also perform packet loss positioning on each packet loss data, and the specific process is as follows:
For each data stream, each telemetry node on the stream has a special counter for marking and storing the order of arrival of the traffic packets; the Multi-bit cyclic marking method (Multi-bit Cycle Marking, MCM) in the alternative dyeing method shown in fig. 3 is adopted, two bits of data are used for marking the arrived data packet, namely, each 11, 10, 01, 00 sequence is a marking period, if the packet loss condition exists, the marking sequence is disturbed, and further, the packet loss moment and the packet loss position can be diagnosed; as shown in fig. 3, when packet loss occurs, the tag sequences are scrambled into 11, 10, 01, 00, 11, 10, 01, 11, 10, 01, 00 and …, and the packet loss time (as indicated by the arrow in fig. 3) can be positioned according to the lost tag, and meanwhile, the packet loss position can be accurately positioned by adopting the switching strategy for the incoming data packet based on each switch.
S13, carrying out packet loss link data recovery according to the link packet loss statistical result, the whole network queue depth matrix and a pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix; the generated countermeasure network can be understood as a GAN network model which is built on the master station platform based on the collected historical operation data of the power communication network in advance, and comprises a generator model and a discriminator model, wherein the structures of the generator model and the discriminator model are provided with two hidden layers, a ReLU activation function is used, an output layer uses a Sigmoid activation function to carry out MinMax normalization, and details are omitted here;
The generation type countermeasure network is mainly used for realizing lost data recovery by executing the following different data recovery strategies on links corresponding to different packet loss types:
1) For random packet loss type data, only the time correlation of the telemetry information of the link is considered during training, and the telemetry information at the packet loss moment is estimated according to the telemetry information of other times of the link;
2) For congestion packet loss type data, other links which pass through the same switch with the link are considered besides telemetry information of other times of the link during training;
3) For the black hole packet loss type data, data recovery is not carried out, and the corresponding position in the network state matrix is filled with NaN;
in order to enable the trained generative countermeasure network to meet the requirements of efficient and reliable lost data recovery, the present embodiment preferably performs optimization training on the generator model and the discriminator model in the network according to the above 3 data recovery strategies, according to the following method:
in the historical operation data (historical network queue depth matrix) collected during each training, according to the preset training batch size, extracting a part of training data, and setting corresponding training parameters, for example, the training parameters comprise batch_size set to 128, hit_rate set to 0.95, alpha set to 100, and item set to 10000:
In order to save the calculation cost and improve the calculation efficiency, the input data of the GAN is preprocessed and post-processed in a distinguishing way according to the three packet loss types in the training process, but the same GAN model is used, specifically, the training data is usedCorresponding network topology matrix->Congestion packet loss marking matrix>Mask matrix->Randomly generated gaussian-distributed noise V as input, a generator model is trained as shown in the following formula:
wherein,when the link isiAnd linkjWhen passing through the same switch, the switch is compromised>Otherwise->The method comprises the steps of carrying out a first treatment on the surface of the Here will->Pretreatment is carried out>Marking matrix for indicating congestion packet loss>Is>Column (S)/(S)>Expressed in the process->Is>When in column, only the link with congestion packet loss needs to be considered, and other links passing through the same switch with the link; />，/>In the time-course of which the first and second contact surfaces,otherwise, 1; />Representing the multiplication of the corresponding elements;
in each training round, only the part of lost packets adopts the recovered data, the rest data is kept unchanged, and the recovered network state matrix generated by the generator after one training round is expressed as:
for the data of the black hole packet loss type, the data recovery is not carried out, and the corresponding position in the network state matrix is filled with NaN, so that the final form of the network state matrix after the data recovery is obtained ：
It should be noted that, the loss function used in the model training of the generator in this embodiment is preferably divided into two parts: when the data is not lost, the loss function is the difference between the result generated by the generator and the original data; when the data is lost, the loss function is the probability that the discriminator judges that the generated false data is true, and is specifically expressed as follows:
wherein,is a superparameter and +.>Expressed as:
correspondingly, the input of the discriminator model of the GAN generates a complete network state matrix with false data output by the generator modelMask matrix->Output as estimated network state mask matrix +.>：
The loss function of the discriminator model described above is as follows:
the objective function of performing the countermeasure generation training on the generator model and the discriminator model to perform network state recovery is as follows:
wherein,DandGrespectively referred to as an identifier model and a generator model,finger->Transpose of->Refers to pair->Averaging all elements in the list; wherein (1)>Refer to +.>. The same references are given in the description of the training process below;
applying cross entropy loss functions in classification problems：
Order theThe objective function may be written as:
the physical meaning of the objective function is to the estimated network state mask matrix And real mask matrix->The cross entropy loss function of the network state evaluation analysis system is optimized, and a generated network model which can directly carry out network state evaluation analysis based on the link packet loss statistical result acquired in real time and the full network queue depth matrix can be obtained through the multi-batch iterative training;
specifically, the step of recovering the lost packet link data according to the link packet loss statistical result, the full network queue depth matrix and the pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix includes:
obtaining a corresponding congestion packet loss marking matrix, a black hole packet loss marking matrix and a network state mask matrix according to the link packet loss statistical result; the method for obtaining the congestion packet loss marking matrix, the black hole packet loss marking matrix and the network state mask matrix can be referred to the related description, and will not be repeated here;
acquiring a network topology matrix of the power communication network, and inputting the network topology matrix, the congestion packet loss marking matrix, the network state mask matrix and the whole network queue depth matrix into a generator model in the generating type countermeasure network to recover packet loss data, so as to obtain a corresponding first network state recovery matrix; wherein the first network state recovery matrix may be understood as a data recovery matrix initially generated by the generator model;
Obtaining a corresponding second network state recovery matrix according to the first network state recovery matrix, the full-network queue depth matrix and the network state mask matrix; the second network state recovery matrix can be understood as a data recovery matrix obtained by recovering only partial data of lost packets based on the first network state recovery matrix;
obtaining the network state recovery matrix according to the second network state recovery matrix, the full network queue depth matrix and the black hole packet loss marking matrix; the network state recovery matrix can be understood as a final form of the network state matrix after the data recovery is obtained by filling the corresponding position in the second network state recovery matrix with NaN without recovering the black hole packet loss type data; specifically, the network state recovery matrix is expressed as:
in the method, in the process of the invention,
wherein,representing a network state recovery matrix; />Representing a second network state recovery matrix; />Representing a first network state recovery matrix; />Representing a full-network queue depth matrix; />Representing a black hole packet loss marking matrix;NaNrepresenting a black hole packet loss filling value; />Represents a network topology matrix, and->When the link is iAndjwhen passing through the same switch, the switch is compromised>Otherwise；/>Marking matrix for indicating congestion packet loss>Is>A column; />Expressed in the process->Is>When in column, only the link with congestion packet loss needs to be considered, and other links passing through the same switch with the link; />Representing a network state mask matrix; />Representing the multiplication of the corresponding elements;Vrepresenting randomly generated noise subject to gaussian distribution.
S14, evaluating the recovery effect of each packet loss link according to the network state recovery matrix, and generating a link set to be recovered by the probe; the link set to be recovered by the probe can be understood as a set of packet loss links with poor recovery effect through the network state; specifically, the step of evaluating the recovery effect of each packet loss link according to the network state recovery matrix and generating the link set to be recovered by the probe includes:
inputting the network state recovery matrix and the network state mask matrix into a discriminator model in the generated countermeasure network to perform network state mask matrix estimation to obtain a corresponding network state mask estimation matrix; wherein the network state mask estimation matrix is expressed as:
wherein,representing a network state recovery matrix; / >Representing a network state mask matrix; />Representing a discriminator model; />Representing a network state mask estimation matrix;
obtaining recovery effect evaluation values of each packet loss link according to the network state mask matrix and the network state mask estimation matrix; the recovery effect evaluation value is expressed as:
wherein,represent the firstiAn evaluation value of recovery effect of each packet loss link; />And->Respectively represent the network status mask matrix->And a network status mask estimation matrix->Is>A row;
judging whether the recovery effect evaluation value of each packet loss link is larger than a preset effect threshold value, if so, adding the corresponding packet loss link serial number into the to-be-detected probe recovery link set; the preset effect threshold can be selected according to actual application requirements, and is not particularly limited herein; when the recovery effect evaluation value of the packet loss link meets the following formula, the recovery effect is considered to be poor, and the link detection is required to be performed by adopting an active probe method:
wherein,represent the firstiAn evaluation value of recovery effect of each packet loss link; />Representing a preset effect threshold.
S15, initiating an active detection packet to detect each packet loss link in the link set to be detected through the fusion terminal according to a source routing technology, acquiring corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result; the source Routing technology can be understood as that a probe packet is actively initiated at the fusion terminal, a transmission path of the active probe packet is designated through a Source Routing (SR), and the active probe packet is sent by using an SR-INT telemetry architecture and a probe data packet format combining the SR technology and the INT technology and carrying an SR-INT payload based on a UDP data packet; in practical application, the INT generator generates an SR-INT detection data packet at the first hop of the monitoring path, rewrites the data packet head to allocate an SR label stack, and adds local INT information before forwarding the data packet; the link data measured by the active probe method can replace the link data with poor recovery effect in the network state recovery matrix obtained by carrying out packet loss link data recovery based on the generation type countermeasure network, so as to form complete and reliable real-time all-network link queue depth data, and a final network state evaluation result is obtained.
According to the scheme, after the network state recovery matrix is obtained by combining the network state recovery matrix in a band with the network telemetry technology and the active probe technology, the state of the power communication network is comprehensively perceived in real time, the deficiency of coverage of the traditional method is overcome, meanwhile, the lost data recovery of different types of lost links is realized by utilizing the generated type anti-network (GAN), the recovery effect of each lost link is evaluated according to the network state recovery matrix, a to-be-probed recovery link set is generated, and according to the source routing technology, an active probe packet is initiated to detect each lost link in the to-be-probed recovery link set through the fusion terminal, corresponding link probe data are acquired, and the network state recovery matrix is updated according to the link probe data, so that the network state evaluation result is obtained.
In order to effectively verify the application effect of the method, the embodiment further sets that under the condition of fixed random packet loss rate and preset effect threshold, the method performs performance comparison verification on the basis of different congestion degrees and the data recovery of the generated anti-network algorithm and the KNN (K-Nearest Neighbor) algorithm without considering surrounding links, and performs performance comparison verification on the basis of different congestion link duty ratios and the INT method with the data recovery of the generated anti-network without performing data recovery based on considering surrounding links:
assuming that the fixed random packet loss rate is 0.05, the network bandwidth, the data rate and the like are fixed values; based on the fixed random packet loss rate and the preset effect threshold, under different congestion degrees (the time of the link congestion state is the percentage of the total time), comparing the performance of the whole network data recovery algorithm (considering the GAN of the surrounding links) with the performance of other two algorithms (not considering the GAN and KNN algorithm of the surrounding links), and evaluating the performance by adopting the mean square error MSE of the recovered data relative to the original data to obtain a result shown in figure 4; as can be easily seen from fig. 4, the method provided by the invention considers that more data resources are used for recovery when the link is congested, and has lower MSE and better recovery performance under the condition of higher congestion degree;
Based on the fixed random packet loss rate and the preset effect threshold, comparing the performance of the full network data recovery and active detection combination algorithm (GAN-SR method) with the performance of the GAN algorithm considering surrounding links and the INT method not using the data recovery algorithm under different congestion link duty ratios, and obtaining the result shown in figure 5; as is apparent from fig. 5, compared with the conventional INT method, the INT method for data recovery by only using the GAN algorithm, the GAN-SR method combined with active detection provided by the present invention has a smaller full-network MSE when the congestion link occupies a larger area, and can more accurately and comprehensively reflect the network state when the full-network congestion link is more.
Although the steps in the flowcharts described above are shown in order as indicated by arrows, these steps are not necessarily executed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders.
In one embodiment, as shown in fig. 6, there is provided a power communication network performance evaluation system, the system comprising:
the remote sensing acquisition module 1 is used for initiating an INT detection packet to acquire remote sensing information through a fusion terminal of the power communication network and uploading the remote sensing information to the master station platform;
The information analysis module 2 is used for carrying out analysis statistics on the telemetry information through the master station platform to obtain a full-network queue depth matrix and a link packet loss statistical result;
the data recovery module 3 is used for carrying out packet loss link data recovery according to the link packet loss statistical result, the full-network queue depth matrix and a pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix;
the effect evaluation module 4 is used for evaluating the recovery effect of each packet loss link according to the network state recovery matrix, and generating a link set to be recovered by the probe;
and the state evaluation module 5 is used for initiating an active detection packet to detect each packet loss link in the link set to be detected through the fusion terminal according to a source routing technology, collecting corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result.
For specific limitations on the power communication network performance evaluation system, reference may be made to the above limitation on the power communication network performance evaluation method, and corresponding technical effects may be equally obtained, which is not described herein. The various modules in the power communication network performance evaluation system described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
Fig. 7 shows an internal structural diagram of a computer device, which may be a terminal or a server in particular, in one embodiment. As shown in fig. 7, the computer device includes a processor, a memory, a network interface, a display, a camera, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by a processor implements a power communication network performance assessment method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those of ordinary skill in the art that the architecture shown in fig. 7 is merely a block diagram of some of the architecture relevant to the present application and is not intended to limit the computer device on which the present application may be implemented, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have the same arrangement of components.
In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when the computer program is executed.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, implements the steps of the above method.
In summary, the method and system for evaluating the performance of the power communication network provided by the embodiment of the invention realize that the fusion terminal of the power communication network initiates INT detection packets to collect telemetry information, the telemetry information is uploaded to the master station platform, analysis statistics is carried out on the telemetry information through the master station platform, after the full network queue depth matrix and link packet loss statistics result are obtained, the corresponding network state recovery matrix is obtained by recovering packet loss link data according to the link packet loss statistics result, the full network queue depth matrix and a pre-established generation type anti-network, the recovery effect evaluation is carried out on each packet loss link according to the network state recovery matrix to generate a to-be-detected link set, and each packet loss link in the to-be-detected link set is detected through the fusion terminal according to the source routing technology, corresponding link detection data are acquired, and the network state recovery matrix is updated according to the link detection data.
In this specification, each embodiment is described in a progressive manner, and all the embodiments are directly the same or similar parts referring to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments. It should be noted that, any combination of the technical features of the foregoing embodiments may be used, and for brevity, all of the possible combinations of the technical features of the foregoing embodiments are not described, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples represent only a few preferred embodiments of the present application, which are described in more detail and are not thereby to be construed as limiting the scope of the invention. It should be noted that modifications and substitutions can be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and substitutions should also be considered to be within the scope of the present application. Therefore, the protection scope of the patent application is subject to the protection scope of the claims.
Claims (10)
1. A method for evaluating the performance of an electrical power communication network, the method comprising the steps of:
an INT detection packet is initiated to acquire telemetry information through a fusion terminal of an electric power communication network, and the telemetry information is uploaded to a master station platform;
analyzing and counting the telemetry information through the master station platform to obtain a full-network queue depth matrix and a link packet loss statistical result;
carrying out packet loss link data recovery according to the link packet loss statistical result, the full network queue depth matrix and a pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix;
performing recovery effect evaluation on each packet loss link according to the network state recovery matrix to generate a link set to be recovered by the probe;
and initiating an active detection packet to detect each packet loss link in the link set to be detected through the fusion terminal according to a source routing technology, acquiring corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result.
2. The power communication network performance evaluation method according to claim 1, wherein the link packet loss statistics include packet loss type and packet loss positioning result of each packet loss link; the packet loss type comprises black hole packet loss, congestion packet loss and random packet loss; the packet loss positioning result comprises packet loss time and packet loss position;
Analyzing and counting the telemetry information through the master station platform, and acquiring a link packet loss statistical result comprises the following steps:
when link packet loss exists, acquiring an INT transmission flow count and an INT reception flow count, and judging whether each link packet loss is a black hole packet loss according to the INT transmission flow count and the INT reception flow count;
if the packet loss is not the black hole packet loss, acquiring corresponding continuous packet loss duration and continuous packet loss length, and judging whether the packet loss is congestion packet loss according to the continuous packet loss duration and the continuous packet loss length;
and if the congestion packet loss is judged, judging the corresponding link packet loss as random packet loss.
3. The power communication network performance evaluation method according to claim 2, wherein the step of judging whether each link packet loss is a black hole packet loss based on the INT transmission traffic count and the INT reception traffic count comprises:
according to the INT sending flow count and the INT receiving flow count, respectively obtaining a corresponding sending flow increment and a corresponding receiving flow increment;
and calculating an increment difference value of the sending flow increment and the receiving flow increment, judging whether the increment difference value is larger than or equal to a preset difference value threshold, and if so, judging the corresponding link packet loss as black hole packet loss.
4. The power communication network performance evaluation method according to claim 2, wherein the step of determining whether or not to be a congestion packet loss based on the continuous packet loss duration and the continuous packet loss length includes:
judging whether the continuous packet loss duration and the continuous packet loss length meet preset congestion packet loss conditions or not; the preset congestion packet loss condition is expressed as:
wherein,representing continuous packet loss duration; />Representing the continuous packet loss length; />And->Respectively representing a duration threshold and a packet loss length threshold;
if so, judging the corresponding link packet loss as congestion packet loss.
5. The method for evaluating the performance of an electric power communication network according to claim 1, wherein the step of obtaining the corresponding network state recovery matrix by recovering the lost packet link data according to the link lost packet statistics, the full network queue depth matrix and a pre-constructed generation type countermeasure network comprises:
obtaining a corresponding congestion packet loss marking matrix, a black hole packet loss marking matrix and a network state mask matrix according to the link packet loss statistical result;
acquiring a network topology matrix of the power communication network, and inputting the network topology matrix, the congestion packet loss marking matrix, the network state mask matrix and the whole network queue depth matrix into a generator model in the generating type countermeasure network to recover packet loss data, so as to obtain a corresponding first network state recovery matrix;
Obtaining a corresponding second network state recovery matrix according to the first network state recovery matrix, the full-network queue depth matrix and the network state mask matrix;
and obtaining the network state recovery matrix according to the second network state recovery matrix, the full network queue depth matrix and the black hole packet loss marking matrix.
6. The power communication network performance evaluation method of claim 5, wherein the network state recovery matrix is expressed as:
in the method, in the process of the invention,
wherein,representing a network state recovery matrix; />Representing a second network state recovery matrix; />Representing a first network state recovery matrix; />Representing a full-network queue depth matrix; />Representing a black hole packet loss marking matrix;NaNrepresenting a black hole packet loss filling value; />Represents a network topology matrix, and->When the link isiAndjwhen passing through the same switch, the switch is compromised>Otherwise->；/>Marking matrix for indicating congestion packet loss>Is>A column; />Expressed in the process->Is>When in column, only the link with congestion packet loss needs to be considered, and other links passing through the same switch with the link; />Representing a network state mask matrix; />Representing the multiplication of the corresponding elements;Vrepresenting randomly generated noise subject to gaussian distribution.
7. The power communication network performance evaluation method according to claim 6, wherein the step of evaluating recovery effects of each packet loss link according to the network state recovery matrix, and generating a set of links to be probed for recovery includes:
inputting the network state recovery matrix and the network state mask matrix into a discriminator model in the generated countermeasure network to perform network state mask matrix estimation to obtain a corresponding network state mask estimation matrix;
obtaining recovery effect evaluation values of each packet loss link according to the network state mask matrix and the network state mask estimation matrix; the recovery effect evaluation value is expressed as:
wherein,represent the firstiAn evaluation value of recovery effect of each packet loss link; />And->Respectively represent the network status mask matrix->And a network status mask estimation matrix->Is>A row;
and judging whether the recovery effect evaluation value of each packet loss link is larger than a preset effect threshold value, if so, adding the corresponding packet loss link serial number into the to-be-probed recovery link set.
8. A power communication network performance assessment system, the system comprising:
the remote sensing acquisition module is used for initiating an INT detection packet to acquire remote sensing information through a fusion terminal of the power communication network and uploading the remote sensing information to the master station platform;
The information analysis module is used for carrying out analysis statistics on the telemetry information through the master station platform to obtain a full-network queue depth matrix and a link packet loss statistical result;
the data recovery module is used for carrying out packet loss link data recovery according to the link packet loss statistical result, the full-network queue depth matrix and a pre-constructed generation type countermeasure network to obtain a corresponding network state recovery matrix;
the effect evaluation module is used for evaluating the recovery effect of each packet loss link according to the network state recovery matrix and generating a link set to be recovered by the probe;
and the state evaluation module is used for initiating an active detection packet to detect each packet loss link in the link set to be detected through the fusion terminal according to a source routing technology, collecting corresponding link detection data, and updating the network state recovery matrix according to the link detection data to obtain a network state evaluation result.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311705506.8A CN117411806B (en) | 2023-12-13 | 2023-12-13 | Power communication network performance evaluation method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311705506.8A CN117411806B (en) | 2023-12-13 | 2023-12-13 | Power communication network performance evaluation method, system, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117411806A true CN117411806A (en) | 2024-01-16 |
| CN117411806B CN117411806B (en) | 2024-03-08 |
Family
ID=89496522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311705506.8A Active CN117411806B (en) | 2023-12-13 | 2023-12-13 | Power communication network performance evaluation method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117411806B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117880055A (en) * | 2024-03-12 | 2024-04-12 | 灵长智能科技(杭州)有限公司 | Network fault diagnosis method, device, equipment and medium based on transmission layer index |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150067819A1 (en) * | 2013-08-28 | 2015-03-05 | Hola Networks Ltd. | System and Method for Improving Internet Communication by Using Intermediate Nodes |
| CN112529148A (en) * | 2020-11-20 | 2021-03-19 | 郑州轻工业大学 | Intelligent QoS inference method based on graph neural network |
| CN113595830A (en) * | 2021-07-30 | 2021-11-02 | 百果园技术(新加坡)有限公司 | Method, device, equipment and storage medium for detecting network packet loss state |
| CN113938407A (en) * | 2021-09-02 | 2022-01-14 | 北京邮电大学 | Data center network fault detection method and device based on in-band network telemetry system |
| CN114567575A (en) * | 2022-04-29 | 2022-05-31 | 南京邮电大学 | INT-based packet loss detection method |
| CN116506354A (en) * | 2023-06-30 | 2023-07-28 | 天地信息网络研究院(安徽)有限公司 | NDN router and active congestion packet loss recovery method thereof |
-
2023
- 2023-12-13 CN CN202311705506.8A patent/CN117411806B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150067819A1 (en) * | 2013-08-28 | 2015-03-05 | Hola Networks Ltd. | System and Method for Improving Internet Communication by Using Intermediate Nodes |
| CN112529148A (en) * | 2020-11-20 | 2021-03-19 | 郑州轻工业大学 | Intelligent QoS inference method based on graph neural network |
| CN113595830A (en) * | 2021-07-30 | 2021-11-02 | 百果园技术(新加坡)有限公司 | Method, device, equipment and storage medium for detecting network packet loss state |
| CN113938407A (en) * | 2021-09-02 | 2022-01-14 | 北京邮电大学 | Data center network fault detection method and device based on in-band network telemetry system |
| CN114567575A (en) * | 2022-04-29 | 2022-05-31 | 南京邮电大学 | INT-based packet loss detection method |
| CN116506354A (en) * | 2023-06-30 | 2023-07-28 | 天地信息网络研究院(安徽)有限公司 | NDN router and active congestion packet loss recovery method thereof |
Non-Patent Citations (2)
| Title |
|---|
| "软件定义数据中心网络数据平面资源管控机制研", 中国优秀博士学位论文全文库, 15 July 2023 (2023-07-15) * |
| LIZHUANG TAN等: "A Packet Loss Monitoring System for In-Band Network Telemetry: Detection, Localization, Diagnosis and Recovery", IEEE, 3 November 2021 (2021-11-03) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117880055A (en) * | 2024-03-12 | 2024-04-12 | 灵长智能科技(杭州)有限公司 | Network fault diagnosis method, device, equipment and medium based on transmission layer index |
| CN117880055B (en) * | 2024-03-12 | 2024-05-31 | 灵长智能科技(杭州)有限公司 | Network fault diagnosis method, device, equipment and medium based on transmission layer index |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117411806B (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117411806B (en) | Power communication network performance evaluation method, system, equipment and storage medium | |
| JP6823501B2 (en) | Anomaly detection device, anomaly detection method and program | |
| Liu et al. | IEEE P2668-compliant multi-layer IoT-DDoS defense system using deep reinforcement learning | |
| Janabi et al. | Convolutional neural network based algorithm for early warning proactive system security in software defined networks | |
| Tan et al. | A packet loss monitoring system for in-band network telemetry: Detection, localization, diagnosis and recovery | |
| CN112769869B (en) | SDN network security prediction method based on Bayesian attack graph and corresponding system | |
| CN110149207A (en) | New type of continuous variable quantum key delivering method based on machine learning | |
| CN111181923A (en) | Flow detection method and device, electronic equipment and storage medium | |
| Liu et al. | Cooperative topology sensing of wireless networks with distributed sensors | |
| Malboubi et al. | A learning-based measurement framework for traffic matrix inference in software defined networks | |
| Cassales et al. | IDSA-IoT: an intrusion detection system architecture for IoT networks | |
| Jin et al. | Mobile network traffic pattern classification with incomplete a priori information | |
| CN102714652B (en) | Monitoring Data network comprises the conversational communication of multiple data flow | |
| Fenil et al. | Towards a secure software defined network with adaptive mitigation of dDoS attacks by machine learning approaches | |
| Di Pietro et al. | Introducing epidemic models for data survivability in unattended wireless sensor networks | |
| Taleb et al. | Hybrid machine learning classification and inference of stalling events in mobile videos | |
| Guo et al. | FullSight: A feasible intelligent and collaborative framework for service function chains failure detection | |
| Wang et al. | Network traffic classification based on federated semi-supervised learning | |
| Bouchoucha et al. | Topology inference of unknown networks based on robust virtual coordinate systems | |
| Desai et al. | Packet forwarding misbehavior detection in next-generation networks | |
| Belavagi et al. | Improved Intrusion Detection System using Quantal Response Equilibrium-based Game Model and Rule-based Classification | |
| Devi et al. | Classification of network traffic on isp link and analysis of network bandwidth during covid-19 | |
| Reddy et al. | Prediction of Packet Delivery Ratio Using Lasso Regression in Comparison with Linear Regression Algorithm for Multi Input Multi Output Network | |
| Rossi et al. | Energy efficient data collection via supervised in-network classification of sensor data | |
| Shyu et al. | A multiagent-based intrusion detection system with the support of multi-class supervised classification |
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 |