CN103281256A - Network tomography-based end-to-end path packet loss rate detection method - Google Patents
Network tomography-based end-to-end path packet loss rate detection method Download PDFInfo
- Publication number
- CN103281256A CN103281256A CN2013101508435A CN201310150843A CN103281256A CN 103281256 A CN103281256 A CN 103281256A CN 2013101508435 A CN2013101508435 A CN 2013101508435A CN 201310150843 A CN201310150843 A CN 201310150843A CN 103281256 A CN103281256 A CN 103281256A
- Authority
- CN
- China
- Prior art keywords
- path
- packet loss
- inde
- matrix
- network
- 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
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a network tomography-based end-to-end path packet loss rate detection method. The method comprises the following steps of: S1, modeling the relationship between path packet loss rate and link packet loss rate in a network path to obtain Gx=b; S2, screening by a routing matrix G to obtain an initial path set Ginde, wherein the path packet loss rate of the Ginde is detected during the screening process; S3, judging whether the Ginde is a maximum linearly independent path set, if not, continuing to execute S4, and if so, setting the Ginde as a set G1 and turning to S5; S4, continuing to screen the Ginde by utilizing a path selection algorithm; and S5, acquiring the path packet loss rate in the extreme set G1 and calculating remaining path packet loss rate. By the network tomography-based end-to-end path packet loss rate detection method disclosed by the invention, the path packet loss rate and the link packet loss rate are modeled by acquiring network topology, and a network environment supports a user datagram protocol (UDP), the load of a network is low, and the number of the obtained detection paths is reduced obviously.
Description
Technical field
The present invention relates to technical field of the computer network, particularly a kind of end-to-end path packet loss detection method of chromatography Network Based.
Background technology
Along with Internet fast development, people improve constantly the degree of dependence of network service, and the performance of network is also had higher requirement.Network performance measurement is basis and the checking means of carrying out network traffics, topology, behavior modeling analysis.How to use lower cost, the operation conditions that measures whole network exactly becomes the focus of more and more researcher's researchs.
Network tomography is to survey by the probe to end-to-end transmission, obtains structure and the performance information of network internal.The network tomography technology can be under the situation that does not have the internal network node cooperation, by initiatively sending probe and inferring large scale network link level qos parameter and topological structure according to statistical method.The packet loss Detection Techniques are the awareness network behaviors, carry out network control, eliminate network bottleneck, improve basis and the important step of network performance, so have very important meaning as one of them branch.
Transmit in the process of information at IP network, congested being difficult to avoided, and the congested network that may cause can't normally provide service, thereby causes economic loss and potential safety hazard.Network end-to-end path packet loss and network link packet loss can reflect operation conditions and the Congestion Level SPCC of network, are the important indicators of network performance, therefore how to obtain the focus that path and link packet drop rate become numerous researcher's researchs rapidly and accurately.The detection problem of primary study end-to-end path packet loss herein.
The path monitoring method of network can be divided into initiatively measurement and passive measurement according to the mode of measuring.The passive measurement technology is only obtained the network performance situation by the method that catches the available data bag that tested network transmits, and its performance index that can analyze out can be subjected to the influence of monitoring range.Initiatively measuring is present emerging network measure mode, and it is measured the transmission situation that wraps in the network by the particular data packet, the record that initiatively send and obtains network performance data, have higher flexibility, but this method need take certain Internet resources.In order to save the Internet resources expense, existing network monitoring method based on initiatively measurement has mainly adopted following measure: 1) optimize and dispose watch-dog, to reduce hardware spending; 2) select the to try one's best quantity of few measuring route; 3) reduce overlapping measurement, to reduce link load as far as possible.
Existing network measuring system is divided into two classes: a class is general measuring system, and a class is only to measure the system of time delay.The former can measure a lot of performances of network, but when n terminal, needs O (n
2) the measurement complexity.And latter's one side measurement result is accurate inadequately; On the other hand, even this latency measurement system can measure the result, still can not provide the guess value of accurate percent of pass and mortality ratio.
In the network with n terminal, the definition path is the connection of terminal room, and link is the connection between the point-to-point, and the path is exactly a series of link set.Just exist O (n at n terminal room so
2) path of number.And the existing technical problem that will solve is: wish path subclass of trying one's best little of monitoring, thereby can infer packet loss and the time delay that all paths of residue.
The method of available technology adopting has: adopt multicast detection or uni-cast probe, in order to solve the limited problem of multicast detection application environment, can adopt uni-cast probe; Perhaps place monitor in initial logic topological tree inside, and collect all observed data of initial logic tree, set up the packet loss sequence of the leaf node of every subtree correspondence then, thereby infer the packet loss that its inner every link; Perhaps use topology and the packet loss of every detective path of unicast packet detection network, and carry out modeling according to the packet loss in the topological sum path of described network, utilize Non-Linear Programming to calculate link packet drop rate according to the model of building.
Yet prior art has the shortcoming that detectivity is low, complexity is high and cost on network communication is big.
Summary of the invention
(1) technical problem of Xie Jueing
The technical problem that the present invention solves is that the more excellent path of few distribution of how selecting to try one's best is surveyed, and obtains the calculated value of network packet loss rate accurately.
(2) technical scheme
The invention provides a kind of end-to-end path packet loss detection method of chromatography Network Based, described method comprises:
S1: the relation to path packet loss and link packet drop rate in the network is carried out modeling, obtains Gx=b, and wherein, G is route matrix, and x is the logarithm of every link percent of pass, and b is the logarithm of every paths percent of pass;
S2: G screens to the route matrix, obtains initial path set G
Inde, in screening process, detect described initial path set G
IndeThe path packet loss;
S3: judge described initial path set G
IndeWhether be very big linear independence set of paths, if not, execution in step S4 then continued; If then establish described initial path set G
IndeBe set G
1, and jump procedure S5;
S4: utilize routing algorithm that initial path is gathered G
IndeProceed screening;
S5: in step S2, survey the described initial path set G that obtains
IndeThe path packet loss in obtain described set G
1In the path packet loss, and calculate the residual paths packet loss; Described residual paths is that described route matrix G filters out described set G
1Remaining path, back.
Preferably, among the step S1 percent of pass be 1 with the difference of packet loss.
Preferably, among the step S2 described route matrix G screened specifically and comprises:
S21: described route matrix G is divided into groups according to the path node j that sets out, and the row of the path of each grouping in described route matrix G number is recorded in S set
jIn;
S22: extract S set
jIn first data m, the path that obtains the capable correspondence of matrix m is vectorial u, judge described vectorial u whether with G
IndeIn SYSTEM OF LINEAR VECTOR irrelevant, if linear independence, jump procedure S23 then is if linear correlation then deposits the number of going m in set T
j
S23: survey the path percent of pass of described vectorial u correspondence, and judge whether the path percent of pass of described vectorial u correspondence is 1, if 1, then with S set
jIn data m from S set
jMiddle deletion, and upgrade described route matrix G, and then upgrade G
IndeIf be not 1, then the path with described vectorial u correspondence joins G
IndeIn.
Preferably, judge the described vectorial u first time and G
IndeIn vector whether during linear independence, establish G
IndeBe null matrix.
Preferably, the described route matrix G of described renewal specifically comprises: be 1 column vector deletion with the value of the capable correspondence of m among the described route matrix G.
Preferably, the G that finally obtains among the determining step S23
IndeOrder whether equal to upgrade the way of escape by the order of matrix G, if unequal, then will gather T
jAssignment is to S set
j, jump procedure S22; If equate, then continue step S3.
Preferably, routing algorithm is specially described in the step S4:
S41: get initial path set G
IndeFirst row be made as v, calculate R
1=orth (G
1 Τ)
Τv
ΤAnd R
2=‖ v ‖
2-‖ R
1‖
2, wherein, T represents transposition, || || expression delivery, G
1Be very big linear independence set of paths;
S42: judge R
2Whether be 0, if be not 0, then v, continue step S43 if satisfying linear independence; If 0, jump procedure S44 then;
S43: v is joined G
1In, and judge whether v is G
1Last column, if, then with G
1Return, otherwise execution in step S44;
S44: initial path is gathered G
IndeThe next line assignment give v, execution in step S41.
Preferably, G among the step S41
1Initial value be null matrix.
Preferably, calculating the residual paths packet loss among the step S5 specifically comprises:
S51: establishing the residual paths matrix is G
Rem, with G
RemWith described very big linear independence set of paths G
1Recombinate, obtain the matrix of recombinating
S52: to reorganization matrix G
FinalCarry out elementary rank transformation, then residual paths percent of pass logarithm is reorganization matrix G
FinalCarry out matrix and described very big linear independence set of paths G that elementary rank transformation obtains
1In path percent of pass logarithm long-pending.
Three) beneficial effect
The present invention proposes a kind of end-to-end path packet loss detection method of chromatography Network Based, by obtaining network topology path packet loss and link packet drop rate are carried out modeling, and support the network environment of UDP, less to the load that network causes, and the detective path quantity that obtains obviously reduces.
Description of drawings
Fig. 1 is method flow diagram provided by the invention;
Fig. 2 is network topology structure figure in the embodiment of the invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described.
Embodiment 1:
The invention provides a kind of end-to-end path packet loss detection method of chromatography Network Based, this method flow diagram as shown in Figure 1, described method comprises:
S1: the relation to path packet loss and link packet drop rate in the network is carried out modeling, obtains Gx=b, and wherein, G is route matrix, and x is the logarithm of every link percent of pass, and b is the logarithm of every paths percent of pass;
By surveying the packet loss of topological and every the detective path that obtains network, set up algebraic equation Gx=b.G is route matrix, a line display one paths among the G, and a link, G are shown in a tabulation
IkBe the element of the capable k row of the i of matrix, value is 0 or 1, works as G
IkRepresented that path i did not comprise link k at=0 o'clock, work as G
Ik=1 expression path i comprises link k.
Among the step S1 percent of pass be 1 with the difference of packet loss.
S2: G screens to the route matrix, obtains initial path set G
Inde, in screening process, detect described initial path set G
IndeThe path packet loss;
Among the step S2 described route matrix G screened specifically and comprises:
S21: described route matrix G is divided into groups according to the path node j that sets out, and the row of the path of each grouping in described route matrix G number is recorded in S set
jIn;
S22: extract S set
jIn data m, the path that obtains the capable correspondence of matrix m is vectorial u, judge described vectorial u whether with G
IndeIn SYSTEM OF LINEAR VECTOR irrelevant, if linear independence, if jump procedure S23 then is linear correlation, then with the capable set T that deposits in of m
j
S23: survey the path percent of pass of described vectorial u correspondence, and judge whether the path percent of pass of described vectorial u correspondence is 1, if 1, then with S set
jIn m capable of S set
jMiddle deletion, and upgrade described route matrix G, and then upgrade G
IndeIf be not 1, then the path with described vectorial u correspondence joins G
IndeIn.
Judge the described vectorial u first time and G
IndeIn vector whether during linear independence, establish G
IndeBe null matrix.Form by a series of 0,1 among the route matrix G, then upgrade described route matrix G and specifically comprise: be 1 column vector deletion with the value of the capable correspondence of m among the described route matrix G.The G that finally obtains among the determining step S23
IndeOrder whether equal to upgrade the way of escape by the order of matrix G, if unequal, then will gather T
jAssignment is to S set
j, jump procedure S22; If equate, then continue step S3.
S3: judge described initial path set G
IndeWhether be very big linear independence set of paths if not, then continue execution in step S4; If then establish described initial path set G
IndeBe set G
1, and jump procedure S5;
S4: utilize routing algorithm that initial path is gathered G
IndeProceed screening;
Routing algorithm is specially following steps described in the step S4:
S41: get initial path set G
IndeFirst row be made as v, calculate R
1=orth (G
1 Τ)
Τv
ΤAnd R
2=‖ v ‖
2-‖ R
1‖
2, wherein, T represents transposition, || || expression delivery, G
1Be very big linear independence set of paths;
S42: judge R
2Whether be 0, if be not 0, then v, continue step S33 if satisfying linear independence; If 0, jump procedure S44 then;
S43: v is joined G
1In, and judge whether v is G
1Last column, if, then with G
1Return, otherwise execution in step S44;
S44: initial path is gathered G
IndeThe next line assignment give v, execution in step S41.
G among the step S41
1Initial value be null matrix.
S5: in step S2, survey the described initial path set G that obtains
IndeThe path packet loss in obtain described set G
1In the path packet loss, and calculate the residual paths packet loss; Described residual paths is that described route matrix G filters out described set G
1Remaining path, back.
Calculating the residual paths packet loss among the step S5 specifically comprises:
S51: establishing the residual paths matrix is G
Rem, with G
RemWith described very big linear independence set of paths G
1Recombinate, obtain the matrix of recombinating
S52: to reorganization matrix G
FinalCarry out elementary rank transformation, then the residual paths packet loss is reorganization matrix G
FinalCarry out matrix and described very big linear independence set of paths G that elementary rank transformation obtains
1In the path packet loss long-pending.
Embodiment 2:
Part topological structure with a practical IP network is that example is showed path of the present invention packet loss detection method below.Exist service end to end in this topological structure between any two main frames.e
1-e
6Represent and have a data harvester in the link network in the network, the data message that exists in real-time monitoring and the collection network.This topological structure as shown in Figure 2.
The end-to-end path packet loss detection method implementation step that the present invention proposes a kind of chromatography Network Based is as follows:
101) read network topology and detection information, concrete steps comprise:
Be that end of probe is carried out the traceroute(Internet mutually with 3 main frames in scheming earlier) obtain network topology, it is as follows to obtain route topological:
Each line display one paths is respectively A → B, A → C, B → A, B → C, C → A, C → B; A link is shown in each tabulation, respectively the e in the corresponding diagram
1To e
6And their opposite direction (from left to right).
102) route matrix G is handled, survey while screening, obtain initial path set G
Inde:
103) judge G
IndeBe very big linear independence set of paths, then G
1=G
Inde, obtain initial path set G
IndeThe logarithm of path percent of pass be
104) according to the packet loss of the very big linear independence set of paths reasoning residual paths that obtains, according to G
1Obtain G
FinalAs follows:
To G
FinalCarry out elementary rank transformation and obtain its row simplest formula G
sFor:
Then
The value that obtains b is:
The value that obtains the path percent of pass after the rearrangement is 1.0000,1.0000,0.9310,0.9800,0.9500,1.0000, and is identical with actual value.
Because the order of route matrix G is 5, all are not when using our algorithm, and detective path bar number is 5, and use after our algorithm herein, by the route matrix deleted that the detective path bar number that obtains is reduced to 3, has saved communication overhead to a great extent.
Above execution mode only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; under the situation that does not break away from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (9)
1. the end-to-end path packet loss detection method of chromatography Network Based is characterized in that described method comprises:
S1: the relation to path packet loss and link packet drop rate in the network is carried out modeling, obtains Gx=b, and wherein, G is route matrix, and x is the logarithm of every link percent of pass, and b is the logarithm of every paths percent of pass;
S2: G screens to the route matrix, obtains initial path set G
Inde, in screening process, detect described initial path set G
IndeThe path packet loss;
S3: judge described initial path set G
IndeWhether be very big linear independence set of paths, if not, execution in step S4 then continued; If then establish described initial path set G
IndeBe set G
1, and jump procedure S5;
S4: utilize routing algorithm that initial path is gathered G
IndeProceed screening;
S5: in step S2, survey the described initial path set G that obtains
IndeThe path packet loss in obtain described set G
1In the path packet loss, and calculate the residual paths packet loss; Described residual paths is that described route matrix G filters out described set G
1Remaining path, back.
2. method according to claim 1 is characterized in that, among the step S1 percent of pass be 1 with the difference of packet loss.
3. method according to claim 1 is characterized in that, among the step S2 described route matrix G is screened specifically to comprise:
S21: described route matrix G is divided into groups according to the path node j that sets out, and the row of the path of each grouping in described route matrix G number is recorded in S set
jIn;
S22: extract S set
jIn first data m, the path that obtains the capable correspondence of matrix m is vectorial u, judge described vectorial u whether with G
IndeIn SYSTEM OF LINEAR VECTOR irrelevant, if linear independence, jump procedure S23 then is if linear correlation then deposits the number of going m in set T
j
S23: survey the path percent of pass of described vectorial u correspondence, and judge whether the path percent of pass of described vectorial u correspondence is 1, if 1, then with S set
jIn data m from S set
jMiddle deletion, and upgrade described route matrix G, and then upgrade G
IndeIf be not 1, then the path with described vectorial u correspondence joins G
IndeIn.
4. method according to claim 3 is characterized in that, judges described vectorial u for the first time and G
IndeIn vector whether during linear independence, establish G
IndeBe null matrix.
5. method according to claim 3 is characterized in that, the described route matrix G of described renewal specifically comprises: be 1 column vector deletion with the value of the capable correspondence of m among the described route matrix G.
6. according to claim 1 or 3 each described methods, it is characterized in that the G that finally obtains among the determining step S23
IndeOrder whether equal to upgrade the way of escape by the order of matrix G, if unequal, then will gather T
jAssignment is to S set
j, jump procedure S22; If equate, then continue step S3.
7. method according to claim 1 is characterized in that, routing algorithm is specially described in the step S4:
S41: get initial path set G
IndeFirst row be made as v, calculate R
1=orth (G
1 Τ)
Τv
TAnd R
2=‖ v ‖
2-‖ R
1‖
2, wherein, T represents transposition, || || expression delivery, G
1Be very big linear independence set of paths;
S42: judge R
2Whether be 0, if be not 0, then v, continue step S43 if satisfying linear independence; If 0, jump procedure S44 then;
S43: v is joined G
1In, and judge whether v is G
1Last column, if, then with G
1Return, otherwise execution in step S44;
S44: initial path is gathered G
IndeThe next line assignment give v, execution in step S41.
8. method according to claim 7 is characterized in that, G among the step S41
1Initial value be null matrix.
9. method according to claim 1 is characterized in that, calculates the residual paths packet loss among the step S5 and specifically comprises:
S51: establishing the residual paths matrix is G
Rem, with G
RemWith described very big linear independence set of paths G
1Recombinate, obtain the matrix of recombinating
S52: to reorganization matrix G
FinalCarry out elementary rank transformation, then residual paths percent of pass logarithm is reorganization matrix G
FinalCarry out matrix and described very big linear independence set of paths G that elementary rank transformation obtains
1In path percent of pass logarithm long-pending.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310150843.5A CN103281256B (en) | 2013-04-26 | 2013-04-26 | The end-to-end path packet loss detection method of chromatography Network Based |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310150843.5A CN103281256B (en) | 2013-04-26 | 2013-04-26 | The end-to-end path packet loss detection method of chromatography Network Based |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103281256A true CN103281256A (en) | 2013-09-04 |
CN103281256B CN103281256B (en) | 2016-05-25 |
Family
ID=49063709
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310150843.5A Expired - Fee Related CN103281256B (en) | 2013-04-26 | 2013-04-26 | The end-to-end path packet loss detection method of chromatography Network Based |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103281256B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106685686A (en) * | 2016-09-12 | 2017-05-17 | 电子科技大学 | Network topology estimation method based on simulated annealing |
CN106713035A (en) * | 2016-12-23 | 2017-05-24 | 西安电子科技大学 | Method for positioning congestion link based on group testing |
CN106936656A (en) * | 2015-12-30 | 2017-07-07 | 华为技术有限公司 | A kind of methods, devices and systems for realizing packet loss detection |
CN113206764A (en) * | 2021-05-26 | 2021-08-03 | 广东电网有限责任公司 | Link packet loss rate reasoning method based on network characteristics |
CN113285835A (en) * | 2021-05-26 | 2021-08-20 | 广东电网有限责任公司 | Carrier network link packet loss rate inference method based on service characteristics |
CN113315679A (en) * | 2021-05-26 | 2021-08-27 | 广东电网有限责任公司 | Link packet loss rate inference method and system based on network resource characteristics |
US11290338B1 (en) | 2019-06-07 | 2022-03-29 | Bae Systems Information And Electronic Systems Integration Inc. | System and method for inferring a network topology from the edge |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102684825A (en) * | 2011-03-18 | 2012-09-19 | 北京市天元网络技术股份有限公司 | Detection method for calculating network packet loss rate |
-
2013
- 2013-04-26 CN CN201310150843.5A patent/CN103281256B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102684825A (en) * | 2011-03-18 | 2012-09-19 | 北京市天元网络技术股份有限公司 | Detection method for calculating network packet loss rate |
Non-Patent Citations (1)
Title |
---|
顾然 等: "基于非线性规划的链路丢包率推理算法", 《电子与信息学报》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106936656A (en) * | 2015-12-30 | 2017-07-07 | 华为技术有限公司 | A kind of methods, devices and systems for realizing packet loss detection |
CN106685686A (en) * | 2016-09-12 | 2017-05-17 | 电子科技大学 | Network topology estimation method based on simulated annealing |
CN106713035A (en) * | 2016-12-23 | 2017-05-24 | 西安电子科技大学 | Method for positioning congestion link based on group testing |
US11290338B1 (en) | 2019-06-07 | 2022-03-29 | Bae Systems Information And Electronic Systems Integration Inc. | System and method for inferring a network topology from the edge |
CN113206764A (en) * | 2021-05-26 | 2021-08-03 | 广东电网有限责任公司 | Link packet loss rate reasoning method based on network characteristics |
CN113285835A (en) * | 2021-05-26 | 2021-08-20 | 广东电网有限责任公司 | Carrier network link packet loss rate inference method based on service characteristics |
CN113315679A (en) * | 2021-05-26 | 2021-08-27 | 广东电网有限责任公司 | Link packet loss rate inference method and system based on network resource characteristics |
Also Published As
Publication number | Publication date |
---|---|
CN103281256B (en) | 2016-05-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103281256A (en) | Network tomography-based end-to-end path packet loss rate detection method | |
CN102449957B (en) | Ip network fault locating method, apparatus, and system | |
CN104753733B (en) | The detection method and device of exception of network traffic data | |
US8355341B2 (en) | Delay time measurement apparatus, storage medium storing delay time measurement program and network system | |
US20130163407A1 (en) | System and method for determining network element criticality | |
CN105490834A (en) | Probe deployment method based on vertex cover and weak vertex cover | |
CN102684902B (en) | Based on the network failure locating method of probe prediction | |
CN102739527B (en) | Network packet loss rate detection method | |
CN105812210A (en) | Distributed network performance measuring system | |
JP2008283621A (en) | Apparatus and method for monitoring network congestion state, and program | |
CN114374626B (en) | Router performance detection method under 5G network condition | |
CN104640125A (en) | Method and device for assessing quality of WLAN (wireless local area network) business | |
CN104518899A (en) | Network routing traffic simulation method and device | |
CN103139014B (en) | Network quality evaluating method based on bypass and device | |
Jabbar et al. | An approach to quantifying resilience in mobile ad hoc networks | |
CN102883359A (en) | Method, device and system for measuring key nodes of wireless sensor network | |
Chen et al. | An efficient solution to locate sparsely congested links by network tomography | |
CN107644474A (en) | A kind of Power Line Inspection System for intelligent grid | |
CN102684825B (en) | Detection method for calculating network packet loss rate | |
CN109104307A (en) | A kind of key node cognitive method of dynamic data chain network | |
CN102868572A (en) | Path delay measuring method based on end-to-end detection | |
CN102075383A (en) | Neural network-based low amplitude network flow anomaly detection method | |
CN105391580A (en) | Network model description method applicable to SpW/SpF | |
CN109756379A (en) | A kind of network performance abnormality detection and localization method based on the decomposition of matrix difference | |
Zhang et al. | Lossy links diagnosis for wireless sensor networks by utilising the existing traffic information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160525 |
|
CF01 | Termination of patent right due to non-payment of annual fee |