CN102075388A - Behavior-based peer-to-peer (P2P) streaming media node identification method - Google Patents
Behavior-based peer-to-peer (P2P) streaming media node identification method Download PDFInfo
- Publication number
- CN102075388A CN102075388A CN2011100062447A CN201110006244A CN102075388A CN 102075388 A CN102075388 A CN 102075388A CN 2011100062447 A CN2011100062447 A CN 2011100062447A CN 201110006244 A CN201110006244 A CN 201110006244A CN 102075388 A CN102075388 A CN 102075388A
- Authority
- CN
- China
- Prior art keywords
- node
- breakpoint
- scheduling
- stream
- consistency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a behavior-based peer-to-peer (P2P) streaming media node identification method for use in a heavy-flow environment. In the method, the high consistency of breakpoint scheduling of data streams in a transmission process is taken as a behavior characteristic that differentiates P2P streaming media from other applications, a corresponding measurement model of the behavior characteristic is constructed, and a judgment threshold is given to realize the detection of the P2P streaming media node. The invention provides a behavior detection method which used to lack. The method has higher detection accuracy and a bright application prospect.
Description
Technical field
The invention belongs to Internet technical field, be specifically related to a kind of P2P stream media node detection method.
Background technology
Along with popularizing day by day and the fast development of the information transmission technology of Internet network, the transmission content on the Internet is transformed into the multi-medium data transmission that comprises text, audio frequency, video by simple literal transmission gradually.Such change not only makes Internet user can obtain more rich and varied information, is also representing the arriving in multi-media network epoch simultaneously.In the face of limited bandwidth and crowded Dial-up Network, video, the audio transmission of real-time implementation narrowband network, best solution adopts the transmission means of streaming video (streaming media) exactly.
Traditional distributed multimedia system mainly uses customer end/server mode, and server connects with mode and each client of clean culture.Because streaming media service has characteristics such as high bandwidth, longer duration, along with the quick increase of client's number, server resource such as bandwidth will soon be consumed totally, become system bottleneck.Scalability problem for resolution system, many researchs have all proposed series of solutions, the alleviation load technology and the content distributing network technology that propose as the IP multicasting technology, at server, yet owing to realize difficulty or can only limited alleviation server stress etc. reason, these technology all do not obtain significant effect.
The P2P pattern can be in large-scale network application the effective extensibility of raising system, so it just becomes certainty with combining of stream media technology.Streaming Media transmission research based on the P2P mode has also caused people's attention very soon, and correlation technique or prototype system constantly occur, and commercial operation also is gradually improved.Show that according to investigations the growth rate of total covering of moon number of users of P2P Streaming Media class software all was higher than the average growth rate of all softwares in every month in recent years.This explanation P2P stream media technology has become the basic trend of network flow-medium application development.
Continuous expansion, particularly file-sharing and the streaming media service used along with the P2P technology go from strength to strength, the safety problem that P2P network itself is potential and to resource, and particularly the abuse of network bandwidth resources has been subjected to the great attention of each Virtual network operator.
P2P is applied in and is experiencing fast-changing process closely in two years, and to complicated, to senior, its network configuration has also experienced the variation that is controlled to full distribution by the center by rudimentary by simply.From the angle of traffic management and monitoring, early stage P2P uses and adopts fixed port number, detects easily and is convenient to management.Developed into the port numbers that adopts dynamic random afterwards gradually, some traditional detection methods have lost effect.The novel P2P that emerges in large numbers in the recent period uses more and more has the consciousness of counterreconnaissance, adopts the gimmick of some encryptions, as the Http agreement that disguises oneself as, and transmits piecemeal and waits and escape identification and detect.At present, the identification that various P2P are used mainly is based on deep layer packet condition code coupling with detecting the technology that has formed matured product, and the extraction of condition code generally relies on the method for expert's manual analysis.Because it is very fast that P2P Streaming Media application software is upgraded, new application software emerges in an endless stream, and adopts the technology of characteristic matching such as various condition codes or port will be difficult to deal with.Therefore must carry out the recognition technology research based on P2P capaciated flow network behavioural characteristic pattern, to realize unknown P2P is used and encrypts the identification of back flow, this is the research with extremely important theory significance and practical value.
At present, the traffic characteristic identification that P2P file download type is used has obtained certain achievement, but belongs to mostly based on the simple stream statistical nature or based on specific program and the identification of the behavioural characteristic of non-intrinsically safe, thereby accuracy is not high.
Particularly the research at P2P stream media node behavioural characteristic does not almost have, and the renewal speed of P2P Streaming Media application software and popularity are rapid and much extensive more than the P2P software of file-sharing type.Because streaming media playing is instant, continuous, not needing to download complete file just can watch to local, any source of playing can utilize this mode to reach purpose instant and that propagate fast, and this makes original safety that has just occurred in the P2P file download application and problem of piracy become even more serious and be difficult to control.
Just under these circumstances, P2P stream media node recognition technology can solve these problems and play good booster action, for example: the network supervision personnel identify the P2P stream media node earlier from huge network traffics after, the broadcast of recombinating of the complete flow of gathering this node then can really realize the purpose of supervising.Therefore, seem particularly important and urgent at P2P stream media node network behavior Feature Recognition Study on Technology.
Summary of the invention
The object of the invention is to overcome the deficiency of existing detection method based on port and other packet condition code coupling, and a kind of P2P stream media node detection method based on behavior is provided, and fills up the current blank of still not having the behavioural characteristic detection method.
Based on the P2P stream media node recognition methods of behavior, be specially under a kind of large traffic environment: judge whether the breakpoint scheduling consistency of downloading node is positioned at the consistency threshold interval, if then this download node is the P2P stream media node;
The breakpoint scheduling consistency of described download node
N is for downloading the connection sum of node, downloads the L that is connected of node and i resource node
iBreakpoint scheduling consistency LS
i=connection L
iIn the continuous breakpoint number/connection L that passes of scheduling
iIn the breakpoint sum;
The continuous breakpoint that passes of described scheduling is defined as: at the connection L of node
iIn, it is breakpoint t that data packet transmission is interrupted constantly
1, the continuous biography of packet is the continuous some t that passes constantly
2, it is dispatching point t that control information contracts out now
3, if satisfy t
1≤ t
3≤ t
2, breakpoint t then
1Pass breakpoint for scheduling property is continuous.
Described consistency threshold interval is 0.8～1.
Technique effect of the present invention is embodied in: utilized the high characteristic of node resource temporality in the application of P2P Streaming Media, and proposed to describe this characteristic with the higher feature of breakpoint scheduling consistency value of node, judged the P2P stream media node with this.
1, the high analysis of causes of node resource temporality during the P2P Streaming Media is used:
(1) of short duration, user behavior pattern at random
Because continuity, sequential and real-time and the user of streaming media playing do not need the complete this locality that downloads to, the behavioral characteristics of P2P stream media system is remarkable unusually, is embodied in especially in the quick variation of node available resources.Because Streaming Media is the very strong application of user's property of participation, in the P2P file-sharing is used, can leave the no longer operation of intervention program behind the download command as long as the user has assigned, so application program can continuous service for a long time; And the user watch Streaming Media to the beginning to whole, and the time of watching by the length decision of a program or a film, be generally less than 2 hours, more important be since people's sense of fatigue determined to resemble other P2P uses can online several hrs, even several days.The user behavior essence that the P2P Streaming Media is used is " not patient ", they can switch in different channel and film, select oneself to like the content see, do not resemble the shared file system user and be ready to spend the time of several hours even several days to go pending file to download, this is to watch the instantaneity of Streaming Media to cause.And in the system of file type of download, only after having downloaded, the user just may estimate its quality to file, can't carry out the conversion of file in download behavior fast.Therefore P2P stream medium system node line duration and the mean value of connect hours all are less than other P2P system.In case and node switching channels or roll off the production line, then the resource that has before its promptly becomes unavailable.
(2) application is to the change of storage demand
Except the short characteristics of node line duration, the holding time of node resource also becomes shorter, because Streaming Media is online broadcast, after downloaded contents plays the user just no longer had value for preservation, so node can not stored downloaded resources on hard disk usually.When particularly live, node resource is only available before the downloading process neutralization is played, and will be covered by new resource soon after playing, and up duration is extremely of short duration.Even resemble the resource of forcing some VOD system client storage to be downloaded, but also be still in finite storage space, and along with the user continues to watch other film and can be override raw content at once.More susceptible condition is to watch midway owing to lose interest in and redirect, the resource that these behaviors directly cause this channel or film disappears or only preserves fragment from subscriber's main station, thereby the variation of node resource availability is also more faster than the variation of node availability in the Streaming Media application, and this is other less appearance of P2P system.
2, the breakpoint scheduling consistency value of P2P stream media node connection is higher is that the node resource temporality is realized technical embodiment in system.Resource is temporary to cause that the scheduling cutout analysis of causes of data flow transmission between node is as follows:
The data flow here is meant that certain P2P stream media node is from other node download stream media data the time, wall scroll is connected on inflow (download) direction by the stream of forming greater than 500 byte IP bags (not comprising the packets of information of transmitting communication information) (annotate: the identifying object among the present invention is primarily aimed at and carries out the P2P stream media node that Streaming Media is downloaded and play, and is not included in the node that line is not but downloaded behavior).In other document transmission system, generally all be disposable download schedule mechanism and the big piecemeal mechanism of data: because the probability of scheduling generation cutout is not high.(network does not have any shake and blocks) in the ideal case, the wall scroll data download stream of these system nodes does not have frequent cutout and occurs, the particularly bigger connection of data volume transmission, downloading process is accomplished without any letup especially.
And in the P2P stream media system, because it is very fast that node resource changes, in order in time to download to enough data, the mechanism that must adopt periodic scheduling to download, this disposable download schedule mechanism that is about to file system has applied in one section very little files in stream media (part second to tens of seconds required length scale of streaming media playing) download, and constantly repeat,, form the characteristics of minor cycle download, frequent scheduling up to the broadcast end of this files in stream media.Therefore, obvious and frequent cutout phenomenon can appear in the wall scroll data flow of P2P stream media node, this cutout phenomenon is different from the of short duration transmission that is caused by network jitter and stops, its appearance is very regular, and have characteristics that are different from other system: the continuous biography of each breakpoint all is by once downloading the caused (see figure 1) of scheduling download instruction that node sends, promptly so-called scheduling cutout.And the cutout that network jitter causes is to recover automatically, need not scheduling.
In sum, the scheduling of data flow cutout characteristic is inevitable and general for the P2P stream media system, then is to have the possibility that occurs for other document transmission system.But a connection in cautious all connect all frequent scheduling cutout that rule unanimity like this occurs can only be the P2P stream media system.Therefore, the present invention proposes a cover and node breakpoint scheduling consistency is measured is calculated and according to the method for threshold decision P2P stream media node.
In order to verify effect of the present invention, table 1 provides the contrast of the breakpoint scheduling coherence measurement value of the various types of P2P of some P2P stream media systems and other system.
Table 1 P2P flow media flux and other various P2P flow breakpoint scheduling coherence measurement contrast
From table 1, can find out, though stream media system under the strictest constraints (α=1), except that the Sopcast system, most measured values are all much larger than threshold value 0.9, under loose condition (α=5), then all measured values almost near 1.Be noted that the kugoo system belongs to the real-time VOD download of audio file, and audio file is generally very short and small, thus the breakpoint greater than 5 seconds sometimes may not can in the flow appears, but Just because of this, it is very good to the consistency of little breakpoint scheduling.
Corresponding with flow media flux is, the P2P flow that other non real-time is downloaded is owing to the phenomenon of " scheduling " only occurs with random chance, the IP bag that promptly meets the scheduling controlling package definition appears in the breakpoint by chance, form the situation of pseudo-scheduling, so no matter be under which kind of constraints, its consistency measured value is all much smaller than 0.9, even 0.8.Wherein, skype and qqgame flow can't not calculate breakpoint owing to relate to big transfer of data, and the transfer of data of emule system is then very coherent, and cutout is generally seldom arranged, so the situation of no breakpoint occurs yet.
According to above experimental result, with breakpoint length and effectively scheduling interval be 5 seconds condition and measure, and 0.9 to be threshold value, identification P2P stream media node effect is better.
Description of drawings
Fig. 1 has shown interior frequent characteristic that the scheduling cutout occurs of connection in the P2P stream media system;
Fig. 2 has described breakpoint scheduling The Characteristic Study object: data flow and complicated control information flow;
What Fig. 3 described is breakpoint and breakpoint scheduling;
Fig. 4 is an entire system frame diagram of the present invention;
Fig. 5 is for catching the bag schematic diagram;
Fig. 6 is stream Hash structural representation;
Fig. 7 is stream record StreamInfo structural representation;
Fig. 8 is the building process of stream record sheet;
Fig. 9 is for setting up stream record sheet copy schematic diagram;
Figure 10 is a nodes records list structure schematic diagram;
Figure 11 is a nodes records table building process.
Embodiment
The invention will be further described below in conjunction with the drawings and specific embodiments.
It is the behavioural characteristic that the P2P stream media node is different from other application that breakpoint scheduling height consistency appears in the present invention's proposition data flow in transmission course, and sets up corresponding measurement model, provides threshold value and carries out the detection of P2P stream media node.
1, sets up the node breakpoint and dispatch conforming measurement model
(1) notion and definition
Flows all on the both direction is gone into, gone out to research object of the present invention for downloading node (the P2P stream media node that hope is identified).Download and being connected between node and each resource node also to be divided into both direction, in different directions the different (see figure 2)s of research object.From resource node to downloading node direction, be on the direction of inbound traffics, the present invention only considers streaming media data transmission, research contents comprises that the IP greater than 500 bytes wraps (during mass data transfers, the IP bag all adopts maximum load to improve efficiency of transmission as far as possible, but the MTU of some network does not reach more than 1000 bytes, so the present invention advises adopting the restriction of 500 bytes).And download node to each resource node direction, promptly on the outflow direction, the present invention only studies the transmission of complicated control information.Simple control information is as the information of confirmation of receipt information, the obstruction/information that unblocks and keep-alive etc., it is very little that their common trait is to wrap load, the overwhelming majority is less than 15 bytes, because so little load can't be represented complex command, thereby can only play some simple control actions.Therefore, the complicated control information bag of mentioning among the present invention has area requirement to size.Provide to give a definition:
Define 2. resource nodes in the data flow of downloading node, if at moment t
1Occur having no progeny in the transmission, and work as data at moment t
2When beginning continuous the biography, t
1Be called breakpoint, t
2Be continuous point, the Δ t=t of passing
1-t
2It then is this breakpoint length (see figure 3).
Define 3. if a pair of breakpoint and continuous the biography between point, occur at least one scheduling controlling bag at the download node on the resource node direction, can think that then the continuous biography of this breakpoint is caused by scheduling, this breakpoint is called the scheduling breakpoint.Otherwise, if any scheduling controlling bag does not appear in a pair of breakpoint and continuous the biography between point, think that then this breakpoint does not have scheduling, this breakpoint does not belong to the scheduling breakpoint.First scheduling controlling contracts out existing moment t
3Be called dispatching point (t
1≤ t
3≤ t
2), dispatching point and the continuous interval delta T=t that passes point
2-t
3Be called the scheduling interval (see figure 3).
Definition 4: in P2P flow media flux pattern measurement model, be connected to measuring object with one in the single node flow.Connect and to be meant in the time range that this meshed network flow takes place, download a corresponding connection between node and a resource node for one, the network traffics summation that transmits between this download node and this resource node is exactly to belong to being connected of this resource node.
Definition 5: at the connection L of certain node
i(i=1,2 ..., n) in, n is the connection sum of this node, promptly has n resource node to exist with this node and is connected.The time series (is time of day with the second) that resource node occurs to the packet of downloading node direction is defined as the data download time sequence TU of this connection
i
Definition 6: at the connection L of certain node
i(i=1,2 ..., n) in, download node and contract out the scheduling time sequence TS that existing time series (is time of day with the second) is defined as this connection to the scheduling control information of resource node direction
i
Definition 7: at TU
iIn the time series, establish t1 and t2 and be a pair of adjacent time point (t1＜t2).Definition when t2-t1＞α, parameter alpha 〉=1 and be integer wherein, then t1 is TU
iA breakpoint in the time series, t2 is the continuous point that passes.
Definition 8: connect L at one
i(i=1,2 ..., (TU n)
i, TS
i) time series centering, if at TU
iIn have breakpoint t1, and have the continuous point that passes to be t2 behind the t1, if at TS
iIn free some t3, the t1≤t3 that satisfies condition≤t2 can judge that then to breakpoint t1 be the continuous breakpoint that passes of scheduling property.
Definition 9: define one and connect L
i(i=1,2 ..., breakpoint scheduling consistency LS n)
iContinuous breakpoint number/breakpoint the sum that passes of=scheduling.
Definition 10: define a breakpoint scheduling consistency of downloading node
Further, in above-described time series generative process, having the IP that meets definition to contract out existing time point will be recorded in the sequence, otherwise and, when occurring as if the packet that does not meet definition at certain time point, then this time point can not appear in the time series.
Further, described parameter alpha can be adjusted according to actual conditions in 1～10s.For example, when the Network Transmission poor quality, data flow transmission can more of short duration cutout occur owing to blocking, and this cutout does not belong to the scheduling cutout, and the too small meeting of α value at this moment causes than multibreak and judges by accident.And the α value excessive also can cause in short sense cycle, to seek less than abundant breakpoint or erroneous judgement be the problem of no breakpoint.
Carry out the network traffics collection by online one time, and by the above definition information Hash table that connects, breakpoint that occurs in every connection of real-time statistics and dispatching point sum, calculate the breakpoint scheduling consistency value of all P2P nodes then, whether surpass threshold value according to this value and judge whether this node is the P2P stream media node.Represent if the breakpoint that calculates by above method scheduling consistency surpasses threshold value each bar of node connect between and in single the connection each breakpoint all present the height consistency of the continuous biography behavior of scheduling.
(2) detection threshold chooses
Breakpoint scheduling in the P2P flow media flux is that normal play is essential, it is the behavior under a kind of program control, therefore the consistency that has height, connect in inner or many connections and can both embody at wall scroll no matter be at a node, and other system is not owing to need such control behavior, in wall scroll connects just by at random probability various possible coherence measurement values appear, therefore, the coherence measurement value mean value of many connections of a node also just is difficult to embody the very high measurement result of consistency.
The breakpoint scheduling consistency threshold value of judging stream media node should be decided to be 1 in theory, no matter be that wall scroll is connected, or a node.But because the identification of scheduling controlling packet is based on very coarse method, and the transmission quality under the actual big capaciated flow network environment be cannot say for sure to demonstrate,prove, the packet loss phenomenon happens occasionally, and therefore will allow to occur certain error, and the present invention is decided to be the consistency threshold value between 0.85 to 1.Can prove, when the breakpoint scheduling consistency of node
The time, wherein
And under the more situation of the connection of participating in calculating, if LS
iOnly with random chance value, mean value
To be difficult to surpass this value.
Specific implementation method of the present invention is as follows:
The overall structure of system as shown in Figure 4.See that on the whole it is three big modules that system is divided into: flow collection, packet reorganization and P2P stream media node identification module.The flow collection module adopts passive measurement mode collection network packet.In the packet reorganization, adopt hash algorithm that packet is reassembled as network flow, P2P Streaming Media identification module adopts each node is calculated breakpoint scheduling consistency and some behavioral trait values as P2P identification simultaneously, according to threshold value node is carried out the P2P Streaming Media then and judges.
Image feature by optical splitter or switch is mirrored to network traffics in the system, and system adopts the passive measurement mode to catch network packet, can not have any impact to network itself.As shown in Figure 5, system generally is deployed in the outlet of network, catch go out on the egress line go into the flow of (TX RX) both direction.
Step 2: the stream record sheet based on hash algorithm makes up
Stream is defined as has identical source IP address, purpose IP address, source port, the set of a series of packets of destination interface and agreement, corresponding above-mentioned connection in certain period.Here because and the actual initiator who is indifferent to connection whom is on earth, so so-called source and destination is only represented the direction of two equities, and the active initiator of disconnected reality and passive recipient.Just because of this, (as making Transmission Control Protocol stream establish the big direction of IP value is the source can to utilize the feature of source and destination IP, and UDP stream is opposite) come stored protocol information, and need not distribute extra memory with the minimizing EMS memory occupation to protocol information again, and can improve stream record search speed.
Adopt the method for Hash to convert packet to stream, write down the source IP address of every stream, purpose IP address, source port, destination interface, agreement (utilizing the size of source and destination IP to represent).
The source IP address that is input as packet of hash algorithm, purpose IP address, source port, destination interface and agreement (utilizing the size of source and destination IP to represent).Hash algorithm is output as 16 shaping variablees, can search size and be 65536 Hash table, determines stream information.Cryptographic Hash is the subscript of stream record array HashTable, and the element of array HashTable is a pointer that points to stream record StreamInfo.Stream record StreamInfo has write down this stream source IP address, purpose IP address, and source port, destination interface, agreement (utilizing the size of source and destination IP to represent), and calculate the essential information of connection breakpoint scheduling consistency value.The Hash table structure as shown in Figure 6.The structure of stream record StreamInfo is seen Fig. 7 (totally 64 bytes, grey block is represented pointer).
Flow the building process of record sheet, at first set up the stream record sheet of a sky, the size of record sheet is 65536, realizes the renewal of described stream record sheet again by the circulation of following steps, constructs described stream record sheet:
1) from network interface card, read an IP bag, extract the source IP address of IP bag, purpose IP address, source port, destination interface and agreement are carried out Hash operation;
2) in the stream record sheet, search the corresponding stream record of this IP bag according to the cryptographic Hash of calculating and whether exist, forward (3) to if exist, otherwise forward (4) to;
3) upgrade this stream recorded information.At first according to actual source, the purpose IP and the corresponding source of writing down, the relation of purpose IP of flowing of IP bag, if source IP is identical, purpose IP is also identical, then is defined as and uploads direction, (otherwise being IP bag source IP=stream record purpose IP, IP bag purpose IP=stream record source IP) then is defined as download directions.If secondly IP bag size is positioned at [65,300] byte, then be defined as the control information bag, be defined as packet greater than 500.If packet, check then whether the interval of time of advent (being the current latest data bag time of advent) of a packet on the same direction and this data packet arrival time reaches the requirement of breakpoint, if reach then breakpoint number adds one, and whether time of control information bag of checking up-to-date arrival in the other direction is among the time interval of these two packets, if scheduling times would add one, upgrade then to be packet time of advent on this direction this bag time of advent.If the control information bag, be this bag time of advent control information bag time of advent of then upgrading on this direction, changes (1);
4) a newly-built stream record and each variable of initialization comprise that direction (upload still and download) and the size (control information bag or packet) according to current data packet is provided with the up-to-date time of advent, and are inserted in the stream record sheet, change (1).
Based on the stream record sheet building process of hash algorithm as shown in Figure 8.
Step 3 is set up the copy of stream record sheet
The present invention is the cycle to carry out data analysis with five minutes (this time can be adjusted according to the real network situation).After obtaining 5 minutes IP bag in real time and utilizing hash algorithm to generate the stream record sheet, open new thread and set up stream record sheet copy, further analyze, meanwhile, main thread is still caught the stream record sheet that the IP bag is set up a new round in real time.Finished in five minutes with analysis as long as can satisfy the establishment of stream record sheet copy, even under the situation that continuous flow is caught, system still can move down in real time, as shown in Figure 9.
Step 4 ergodic flow record sheet
The present invention uses in the face of the P2P Streaming Media in node layer to detect.
Adopt hash algorithm convection current record sheet copy further to handle, make up nodes records table memory node information.Nodes records has comprised this IP addresses of nodes, and this node is initiated or all streams of reception.The structure of nodes records table is seen Figure 10 (16 bytes, grey block is represented pointer).
The building process of nodes records table is as follows:
At first set up the nodes records table of a sky, realize the renewal of described nodes records table again by the circulation of following steps, finally construct described nodes records table:
1) from the stream transcript, obtains a stream record;
2) extract source IP address in this stream record, Hash operation is carried out in this address;
3) judge according to the cryptographic Hash of calculating whether this source IP address exists in the nodes records table, change step 4), otherwise change step 5) if exist;
4) upgrade this nodal information in the nodes records table, being about to this stream record inserts in the stream chain of nodes records---and according to this node is source or purpose, the stream item is composed next_s (this node is the source) or the next_d (this node is a purpose) that writes down to stream, and will flow the address of writing down and compose to stream.Change step 6);
5) a newly-built nodal information writes down and is inserted in the nodes records table, changes 4);
6) extract purpose IP address in this stream record, Hash operation is carried out in this address;
7) judge according to the cryptographic Hash of calculating whether this source IP address exists in the nodes records table, change step 8), otherwise change step 9) if exist;
8) upgrade this nodal information in the nodes records table, being about to this stream record inserts in the stream chain of nodes records---and according to this node is source or purpose, the stream item is composed next_s (this node is the source) or the next_d (this node is a purpose) that writes down to stream, and will flow the address of writing down and compose to stream.Change step 10);
9) a newly-built nodal information writes down and is inserted in the nodes records table, changes 8);
10) whether judgement stream transcript travels through and finishes, if change step 11), otherwise change step 1);
11) described nodes records table makes up and finishes.
Nodes records table building process as shown in figure 11.
The breakpoint scheduling consistency of step 5 computing node
Calculate its breakpoint scheduling consistency for each node.Computational process is traversal nodes records table, and each node is traveled through stream under it.Initiatively initiate or each bar stream of passive reception at individual node, the scheduling number that calculates this stream is divided by breakpoint number, the i.e. breakpoint of stream scheduling consistency ", the breakpoint scheduling of these all streams of download node is average than doing sums, and the breakpoint that promptly obtains this download node is dispatched consistency.
The identification of step 6 P2P stream media node
If the breakpoint of node scheduling consistency calculated value promptly is judged to be the P2P stream media node, otherwise is non-P2P stream media node in interval [0.8,1].
Claims (2)
- Under the large traffic environment based on the P2P stream media node recognition methods of behavior, be specially: judge whether the breakpoint scheduling consistency of downloading node is positioned at the consistency threshold interval, if then this download node is the P2P stream media node;The breakpoint scheduling consistency of described download nodeN is for downloading the connection sum of node, downloads the L that is connected of node and i resource node iBreakpoint scheduling consistency LS i=connection L iIn the continuous breakpoint number/connection L that passes of scheduling iIn the breakpoint sum;
The continuous breakpoint that passes of described scheduling is defined as: at the connection L of node iIn, it is breakpoint t that data packet transmission is interrupted constantly 1, the continuous biography of packet is the continuous some t that passes constantly 2, it is dispatching point t that control information contracts out now 3, if satisfy t 1≤ t 3≤ t 2, breakpoint t then 1Pass breakpoint for scheduling property is continuous. - 2. P2P stream media node according to claim 1 recognition methods is characterized in that, described consistency threshold interval is 0.8～1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100062447A CN102075388A (en) | 2011-01-13 | 2011-01-13 | Behavior-based peer-to-peer (P2P) streaming media node identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100062447A CN102075388A (en) | 2011-01-13 | 2011-01-13 | Behavior-based peer-to-peer (P2P) streaming media node identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102075388A true CN102075388A (en) | 2011-05-25 |
Family
ID=44033741
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100062447A Pending CN102075388A (en) | 2011-01-13 | 2011-01-13 | Behavior-based peer-to-peer (P2P) streaming media node identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102075388A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106454888A (en) * | 2015-08-13 | 2017-02-22 | 国基电子（上海）有限公司 | Data transmission quality management and control system and method |
| CN110781576A (en) * | 2019-09-09 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Simulation node scheduling method, device and equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101795214A (en) * | 2010-01-22 | 2010-08-04 | 华中科技大学 | Behavior-based P2P detection method under large traffic environment |
| WO2010108535A1 (en) * | 2009-03-23 | 2010-09-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Event identification in peer to peer networks |
| CN101883030A (en) * | 2010-07-21 | 2010-11-10 | 华中科技大学 | Detection method of P2P nodes based on random measure of IP addresses |
-
2011
- 2011-01-13 CN CN2011100062447A patent/CN102075388A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010108535A1 (en) * | 2009-03-23 | 2010-09-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Event identification in peer to peer networks |
| CN101795214A (en) * | 2010-01-22 | 2010-08-04 | 华中科技大学 | Behavior-based P2P detection method under large traffic environment |
| CN101883030A (en) * | 2010-07-21 | 2010-11-10 | 华中科技大学 | Detection method of P2P nodes based on random measure of IP addresses |
Non-Patent Citations (1)
| Title |
|---|
| 《中国博士学位论文全文数据库 信息科技辑》 20090515 周丽娟 P2P流媒体识别方法的研究 中国学术期刊(光盘版)电子杂志社 , 第05期 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106454888A (en) * | 2015-08-13 | 2017-02-22 | 国基电子（上海）有限公司 | Data transmission quality management and control system and method |
| CN106454888B (en) * | 2015-08-13 | 2019-10-01 | 国基电子（上海）有限公司 | Data transmission quality managing and control system and method |
| CN110781576A (en) * | 2019-09-09 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Simulation node scheduling method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pianese et al. | PULSE: an adaptive, incentive-based, unstructured P2P live streaming system | |
| Wu et al. | Modeling dynamics of online video popularity | |
| EP2052503B1 (en) | Method for optimizing the transfer of information in a telecommunication network | |
| FR2906950A1 (en) | METHOD AND DEVICES FOR ADAPTING THE TRANSMISSION RATE OF A DATA STREAM IN THE PRESENCE OF INTERFERENCE. | |
| CN107404526A (en) | P2P remotely-sensed data network self-adapting dissemination systems and method based on SDN | |
| CN101795214B (en) | Behavior-based P2P detection method under large traffic environment | |
| Baldesi et al. | Improving P2P streaming in wireless community networks | |
| CN105357071A (en) | Identification method and identification system for network complex traffic | |
| García et al. | Statistical characterization of a real video on demand service: User behaviour and streaming-media workload analysis | |
| CN102075388A (en) | Behavior-based peer-to-peer (P2P) streaming media node identification method | |
| CN106210751B (en) | The method and system of internet video live broadcasting data hierarchy transmission | |
| CN103533048B (en) | Popular seed file acquisition methods in a kind of BT networks | |
| Yang et al. | The content pollution in peer-to-peer live streaming systems: Analysis and implications | |
| CN101090367B (en) | Data transmission method and device in reciprocal network | |
| Tian et al. | A novel caching mechanism for peer-to-peer based media-on-demand streaming | |
| CN101459546A (en) | Recognition method and apparatus for peer-to-peer node flow | |
| CN106034106B (en) | Network-building method, networking management method and device and system | |
| Gonzalez et al. | On the tweet arrival process at Twitter: Analysis and applications | |
| Horovitz et al. | Collabrium: Active traffic pattern prediction for boosting P2P collaboration | |
| Piatek et al. | Liveswarms: Adapting bittorrent for end host multicast | |
| Giacomazzi et al. | Performance Analysis of Peer-to-Peer Video Streaming Systems with Tree and Forest Topology | |
| FAROOQ | Static Peers for Peer-to-Peer Live Video Streaming | |
| So et al. | Adaptive neighbor management for cooperative P2P video-on-demand streaming | |
| Jia et al. | Preferential attachment topology formation of the P2P IPTV system | |
| Bracciale et al. | Simulation of Peer-to-peer streaming over large-scale networks using OPSS |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110525 |