CN101631030B - Prediction method of self-adapting digital home network flow media transmission band width - Google Patents

Prediction method of self-adapting digital home network flow media transmission band width Download PDF

Info

Publication number
CN101631030B
CN101631030B CN2009100416868A CN200910041686A CN101631030B CN 101631030 B CN101631030 B CN 101631030B CN 2009100416868 A CN2009100416868 A CN 2009100416868A CN 200910041686 A CN200910041686 A CN 200910041686A CN 101631030 B CN101631030 B CN 101631030B
Authority
CN
China
Prior art keywords
transmission bandwidth
server
packet
client
band width
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.)
Expired - Fee Related
Application number
CN2009100416868A
Other languages
Chinese (zh)
Other versions
CN101631030A (en
Inventor
余荣
高如超
谢胜利
黄明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN2009100416868A priority Critical patent/CN101631030B/en
Publication of CN101631030A publication Critical patent/CN101631030A/en
Application granted granted Critical
Publication of CN101631030B publication Critical patent/CN101631030B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention provides a prediction method of self-adapting digital home network flow media transmission band width. A server is used for transmitting a package train to a client for measuring to obtain an initial transmission band width value; the bound of the value is determined, and the coding scheme and the coding ratio used for the value band width are determined; then, the server predicts practical transmission band width according to the initial transmission band width value, the coding ratio and a decoding ratio fed back by the client, and the band width is compared with the bound of the initial band width value to determine whether the next practical transmission band width value is predicted or not. The invention can predict the initial transmission band width which can be used in a digital home network and can predict the practical transmission band width in the digital home network flow media in a self-adapting mode so as to ensure that video images can code in a self-adapting mode and transmit according to the network band width condition, which better utilizes limited network band width.

Description

Adaptive digital home network flow media transmission band width Forecasting Methodology
Technical field
The present invention relates to the digital home network technology, particularly a kind of adaptive digital home network flow media transmission band width Forecasting Methodology.
Background technology
Digital home is meant various families and individual digital product, integrate functions such as audiovisual entertainment, information service and household control according to what the Modern Family life requirement was formed, and realize the comprehensive intelligent system of information interaction and socialization home services by cable TV, broadband connections, radio communication etc. and the external world.Digital home is the concrete product that IT, information household appliances, communication highly merge.Digital home's notion was risen in the North America from the later stage nineties, reached a relative climax to calendar year 2001.Early stage digital home is that the minority techie can have, use that luxurious ornaments---intelligent network is emphasized in system design, by automated manners such as wireless, telephone remote, internet controls, realize control for systems such as illumination, household electrical appliances, security alarm, temperature and illumination detections.
The essence of digital home is based on the home network and the loaded service thereof of IP technology.Professional varied, varied in the home network, such as: family's communication, home entertaining, household safe, tele-medicine, household electronic government affairs, home library, household electric business or the like.Video image information becomes topmost information flow in digital home, but because the restriction of hardware and the bottleneck of Network Transmission bandwidth, present digital home gateway is unsatisfactory to the processing of media information, present home network device can't effectively be handled the high speed media information flow, address this problem, can start with from the following aspects:
At first, need provide coding/decoding system efficiently.For example the video flowing as hundreds of Mbps will transmit, and may have only hundreds of kbps to several Mbps at the network bandwidth that shared video end can provide, and just there is very big bandwidth bottleneck in this, and code decode algorithm is realized efficiently.In some cases, also may have a plurality of video flowings simultaneously, all need to upload high-speed video stream to home gateway such as camera in the family and video telephone, like this, code decode algorithm just seems particularly important efficiently.
Secondly, need coding/decoding system more flexibly, make digital home network need coding/decoding system can network bandwidth adaptive, user terminal and business demand, this mainly shows the following aspects: 1. the resolution of each user terminal is different, and this just needs code decode algorithm can adapt to the user terminal of different resolution; 2. because different network environments is inconsistent to the network bandwidth that user terminal provides, and network often also fluctuates for the bandwidth that current business provides, and this just needs coding/decoding system can make full use of the limited network bandwidth provides the most effective most interested top-quality video information to the user; 3. at different business and user's request, also be different to the processing and the demand of media information, this just needs video coding and decoding system can satisfy business and user's request flexibly.And existing home network device all is to adopt fixing coding and decoding scheme but not telescopic, therefore can't realize this demand.
At last, the intelligent network mode that needs support high speed media information to interconnect.3C (being computer, communication and consumption electronic product) merges and has become the development trend of present digital home network, and how to realize that interconnecting of these three kinds of networks and relevant device seems particularly important in the development of digital home network.Simultaneously in order to support the coding/decoding system of high speed media information flow flexibly and efficiently, need digital home network can carry out real-time network test and monitoring and carry out device description accurately.
Therefore in digital home network, in order to utilize limited bandwidth resources, more reasonable, more effectively realize high speed media information interconnecting between heterogeneous network, different terminals, need a cover effectively, flexibly, can the adaptive terminal demand and the audio/video encoding/decoding technology of network environment.Moreover, in order to cooperate the enforcement of this encoding and decoding technique in digital home network, also need to develop corresponding network bandwidth monitoring technique, and need to improve relevant terminal equipment description document.
Network bandwidth measuring technique has following classification:
(1) by whether injecting packet to network internal, the bandwidth measurement technology can be divided into passive measurement (Passive Measurement) and initiatively measure (Active Measurement);
(2) whether need the cooperation of node router by measuring process, the link bandwidth that the bandwidth measurement technology can be divided into hop-by-hop measure and end to end path bandwidth measure;
(3) by different estimating, the bandwidth measurement technology can be divided into the measuring technique of link bandwidth (Link Capacity), path bandwidth (Path Capacity), available bandwidth (Available Bandwidth) and batch data transmittability (BTC-Bulk Transfer Capacity).
To the monitoring of digital home network situation mainly is monitoring to current network available bandwidth situation, what therefore mainly use is the Measurement Algorithm of available bandwidth, and common available bandwidth measurement technology can be divided into direct detection according to detection mode and iteration is surveyed two big classes.PGM (Probe Gap Model) model is a kind of of direct detection (Direct Probing), and it estimates available bandwidth by investigating packet to interval variation, and source host is with speed R iSend the packet string, the speed that the packet string arrives destination host is R 0, available bandwidth A can be calculated by following formula,
A = C TL - R i ( C TL R 0 - 1 ) , Work as R i＞A;
The prerequisite of using the PGM model is compact link bandwidth C TLBe given value.The basic ideas that iteration is surveyed are to make path congestion artificially with the high-speed inspection flow, obtain available bandwidth then; This method is R to the network charge velocity iBurst packets to string, when bag to speed during greater than available bandwidth, flash crowd takes place in the path, the sequential relationship between packet changes, and analyzes the time delay feature of packet and can estimate the outbound path available bandwidth; Because available bandwidth the unknown, measuring process are actually the iterative process that bag constantly changes speed, so the iteration detection also claims PRM (Probe Rate Model) model.Different with the PGM model, compact link bandwidth C is surveyed and do not required to iteration TLBe given value, bag is to speed R iCan be linear change or change by certain function rule.The PRM measuring process adopts following decision condition, when recording R 0＜R iThe time, think R i＞A; When recording R 0=R iThe time, think R i≤ A; R can not appear 0＞R iChange R iCarry out iteration and survey, find R 0=R iCritical point finally obtains path available bandwidth A.PathChirp belongs to the PRM model.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, a kind of adaptive digital home network flow media transmission band width Forecasting Methodology is provided, this method can realize initial transmission bandwidth available in the digital home network is detected, and adaptively the actual transmission bandwidth of digital home network flow media is predicted, make video image carry out adaptive coding and transmission, utilized the limited network bandwidth better according to the bandwidth situation of network.
The present invention is achieved through the following technical solutions: a kind of adaptive digital home network flow media transmission band width Forecasting Methodology may further comprise the steps:
(1) behind the system initialization, server needs to carry out the client transmission bag string that transmission bandwidth is measured in network, carries out transmission bandwidth and measures, and obtains the initial transmission bandwidth numerical value B (t) between server and the client;
(2) server is determined the numerical upper limits B of this transmission bandwidth according to the initial transmission bandwidth numerical value B (t) that obtains Th_HWith numerical lower limits B Th_L, and the encoding scheme and the encoding rate f of transmission bandwidth when being identified for this numerical value s(t);
(3) every time t ', client sends current decoding rate f to server in real time rAnd standby (t), by server record; Server is according to fixed encoding scheme and encoding rate f simultaneously s(t) send Video stream information to client;
(4) every time t 0, server is according to the initial transmission bandwidth numerical value B (t), the encoding rate f that obtain s(t) and decoding rate f r(t), to t 0After actual transmission bandwidth numerical value predict, obtain t 0After actual transmission bandwidth numerical value B (t+t 0);
(5) server is with the actual transmission bandwidth numerical value B (t+t that obtains 0) respectively with the numerical upper limits B in initial transmission broadband Th_HAnd numerical lower limits B Th_LCompare; If comparative result is B Th_L≤ B (t+t 0)≤B Th_H, then jump to step (3), carry out the prediction of actual transmission bandwidth numerical value next time; If comparative result is B (t+t 0)＞B Th_HOr B (t+t 0)＜B Th_L, then jump to step (1), carry out the transmission bandwidth prediction of server and client again.
Server sends the bag string to client and adopts the pathchirp method in the described step (1), specifically may further comprise the steps:
(1-1) server carries out the client transmission bag string that transmission bandwidth is measured to needs, and the interval in the bag string between packet and the packet is exponential increasing trend;
After (1-2) client receives bag string, with the queuing delay of each packet in the bag string of receiving with whether measure successful feedback information and return server;
After (1-3) server receives feedback information, respectively the transmission bandwidth of each packet correspondence is estimated earlier, obtained the estimated value E of each packet correspondence k (m), pass through formula at last
D ( m ) = Σ k = 1 N - 1 E k ( m ) Δ k Σ k = 1 N - 1 Δ k
Estimated value E to each packet correspondence k (m)Get weighted average D (m), as the corresponding transmission bandwidth estimated value of this bag string; Wherein, k represents k packet, and m represents the m time measurement, Δ kRepresent the time interval between k packet and (k+1) individual packet.
(1-4) serve estimated value D at last to a plurality of bag string correspondences that record (m)Average, obtain the initial transmission bandwidth numerical value B (t) between server and the client.
Wherein the pathchirp method mainly is that server transmission bag string (interval in the bag string between packet and the packet is exponential increasing trend) to client, carries out the prediction of its initial transmission bandwidth numerical value according to each the packet queuing delay in the client feedback information and other information then.Owing to measure the incipient stage we have no to understand to the available bandwidth situation of network initial, and do not know the residing scope of available bandwidth, therefore this moment we need send a lot of packets to network can with transmission bandwidth measure, the packet rate scope that sends may be between 1-100Mbps, even may be bigger, this depends on network type.We can obtain Preliminary study to the network bandwidth measurement by this at the beginning the time, for example our speed range of the packet that sends in network is 1-100Mbps at the beginning, and client receives the queuing delay that can obtain wrapping each packet in the string after this a string detection flows; Client feeds back to server with these data then, and the survey tool of server is calculated the estimated value of the transmission bandwidth of measurement this time according to the pathchirp algorithm, has so just obtained the Preliminary study to the network bandwidth.
When adopting the pathchirp method to calculate, in the ideal case, the packet queuing delay that client receives is a monotonically increasing, but owing to the background traffic that happens suddenly can occur, the queuing delay of each packet can not be monotonically increasing usually in a bag string.At this moment, pathchirp need utilize the signal shape of these queuing delays that the transmission bandwidth of each packet correspondence is an estimated value E k (m), then to a plurality of E k (m)Get weighted average, be used as the estimated value D of the transmission bandwidth of this bag string (m):
D ( m ) = Σ k = 1 N - 1 E k ( m ) Δ k Σ k = 1 N - 1 Δ k
At last to the estimated value D of a plurality of bag string correspondences of recording (m)Average, obtain the initial transmission bandwidth B (t) between server and the client.
In order to calculate E exactly k (m), pathchirp belongs to the zone of skew to the queuing delay signal segmentation of each packet one-tenth and does not belong to the zone of skew, and this will use pathchirp skew partitioning algorithm, if intuitively go up the q of continuous several bags k (m)(q k (m)Be the queuing delay of k packet in m the bag string) greater than 0 and increase, so these to be surrounded by may be the part of some busy periods of congested formation on the path.Concrete details is as follows, our target be to identify skew begin wrap sequence number i and end packet sequence number j, so each q occurs k (m)＜q K+1 (m)Bag i all might be the skew starting point; We define the end point j of skew for working as
q ( j ) - q ( i ) < max i &le; k &le; j [ q ( k ) - q ( i ) ] F
The time first bag, F is attenuation coefficient (decrease factor) in the formula; The queuing delay q (j) of ordering at j has been fallen by coefficient F increases to the maximum of j from i queuing delay.If j-i＞L (L is a constant, generally gets 5), then we i to the packet between the j as skew.
Estimated value E for each packet k (m), the arbitrary packet k in each bag string can be in a kind of in following three kinds of situations:
(a) if k belongs to skew and the q that a meeting stops k (m)＜q K+1 (m)The time, then
E k (m)＝R k
(b) if k belongs to the skew that can not stop, then
E k (m)＝R 1， &ForAll; k > l
L is the starting point of skew.
(c) for the packet k that does not belong to above two kinds of situations, then establish E k (m)=R 1This has comprised that those packets that do not belong to skew and those belong to skew but queuing delay is the packet that successively decreases, because last skew does not stop, therefore selects l=N-1.Here R kBe the speed of k packet, R 1Be the speed of the 1st packet, N is the number of contained packet in the bag string.
The numerical upper limits B of initial transmission bandwidth described in the step (2) Th_HSpan be B Th_H=B (t)+m, the numerical lower limits B of initial transmission bandwidth Th_LSpan be B Th_L=B (t)-m, wherein m is 5%～10% of B (t).
Encoding scheme described in the step (2) adopts the scalable video technology, and described encoding rate is tabled look-up according to the numerical value of current bandwidth and obtained.
T ' described in the step (3) is the feedback time interval of client, t 0Be the predicted time interval that server carries out the actual transmission bandwidth numerical prediction, t '＜t 0, t &prime; = [ 1 3 , 1 2 ] t 0 .
(the t+t of actual transmission bandwidth numerical value B described in the step (4) 0) calculate by following formula:
B ( t + t 0 ) = B ( t ) + &eta;B ( t ) + &PartialD; f r ( t ) &PartialD; t ;
In the formula, f s(t) be the t encoding rate of server constantly; f r(t) be the t decoding rate of client constantly; &eta; = f s ( t ) f r ( t ) Be self study speed, the forecast updating amplitude of promptly current actual transmission bandwidth.
Server is with predicted time interval t in the step (5) 0After actual transmission bandwidth numerical value B (t+t 0) compare with initial transmission bandwidth numerical value B (t), judge whether to carry out once the prediction of the actual transmission bandwidth numerical value of adaptive digital home network again.If B (t+t 0) numerical value between the numerical value bound of B (t), then carry out the prediction of actual transmission bandwidth numerical value next time; If B (t+t 0) numerical value greater than the numerical upper limits B of B (t) Th_H, illustrate that current actual transmission bandwidth numerical value may be greater than the initial transmission bandwidth upper limit B of this bag string Th_H, server need be adjusted encoding scheme and improve encoding rate f s(t), thus improve bandwidth utilization; If B (t+t 0) numerical value less than the numerical lower limits B of B (t) Th_L, illustrate that current actual transmission bandwidth numerical value may be less than the initial transmission lower band B of this bag string Th_L, server need be adjusted encoding scheme and reduce encoding rate f s(t), thus guarantee that client can the smooth playing video.
Compared with prior art, the present invention has following beneficial effect:
The present invention is by the measurement to transmission bandwidth in the network, allow the server operating position of transmission bandwidth in the monitor network in real time, make video image carry out adaptive coding and transmission according to the situation of transmission bandwidth in the network, on the basis that initial transmission bandwidth numerical value is determined, carry out the actual transmission bandwidth prediction, can save offered load, be reduced to the greatest extent the burden of learning the Network Transmission bandwidth usage and network being caused to greatest extent, make server can utilize limited bandwidth resources, more reasonable, realize that more effectively high speed media information is at heterogeneous network, interconnecting between the different clients.
Description of drawings
Fig. 1 is the interval schematic diagram between each packet in the bag string of the present invention.
Fig. 2 is the queuing delay schematic diagram of each packet in the bag string of the present invention.
Fig. 3 is a method flow diagram of the present invention.
Embodiment
Below in conjunction with embodiment and accompanying drawing, the present invention is described in further detail, but embodiments of the present invention are not limited thereto.
Embodiment
A kind of adaptive digital home network flow media transmission band width Forecasting Methodology of present embodiment as shown in Figure 3, may further comprise the steps:
(1) behind the system initialization, server needs to carry out the client transmission bag string that transmission bandwidth is measured in network, carries out transmission bandwidth and measures, and obtains the initial transmission bandwidth numerical value B (t) between server and the client;
(2) server is determined the numerical upper limits B of this transmission bandwidth according to the initial transmission bandwidth numerical value B (t) that obtains Th_HWith numerical lower limits B Th_L, and the encoding scheme and the encoding rate f of transmission bandwidth when being identified for this numerical value s(t);
(3) every time t ', client sends current decoding rate f to server in real time r(t), standby by server record; Server is according to fixed encoding scheme and encoding rate f simultaneously s(t) send Video stream information to client;
(4) every time t 0, server is according to the initial transmission bandwidth numerical value B (t), the encoding rate f that obtain s(t) and decoding rate f r(t), to t 0After actual transmission bandwidth numerical value predict, obtain t 0After actual transmission bandwidth numerical value B (t+t 0);
(5) server is with the actual transmission bandwidth numerical value B (t+t that obtains 0) respectively with the numerical upper limits B in initial transmission broadband Th_HAnd numerical lower limits B Th_LCompare; If comparative result is B Th_L≤ B (t+t 0)≤B Th_H, then jump to step (3), carry out the prediction of actual transmission bandwidth numerical value next time; If comparative result is B (t+t 0)＞B Th_HOr B (t+t 0)＜B Th_L, then jump to step (1), carry out the transmission bandwidth prediction of server and client again.
Server sends the bag string to client and adopts the pathchirp method in the described step (1), specifically may further comprise the steps:
(1-1) server carries out the client transmission bag string that transmission bandwidth is measured to needs, and as shown in Figure 1, the interval in the bag string between packet and the packet is exponential increasing trend;
After (1-2) client receives bag string, with the queuing delay of each packet in the bag string of receiving with whether measure successful feedback information and return server;
After (1-3) server receives feedback information, respectively the transmission bandwidth of each packet correspondence is estimated earlier, obtained the estimated value E of each packet correspondence k (m), pass through formula at last
D ( m ) = &Sigma; k = 1 N - 1 E k ( m ) &Delta; k &Sigma; k = 1 N - 1 &Delta; k
Estimated value E to each packet correspondence k (m)Get weighted average D (m), as the corresponding transmission bandwidth estimated value of this bag string; Wherein, k represents k packet, and m represents the m time measurement, Δ kRepresent the time interval between k packet and (k+1) individual packet.
(1-4) serve estimated value D at last to a plurality of bag string correspondences that record (m)Average, obtain the initial transmission bandwidth B (t) between server and the client.
Wherein the pathchirp method mainly is that server transmission bag string (interval in the bag string between packet and the packet is exponential increasing trend) to client, carries out the prediction of its initial transmission bandwidth numerical value according to each the packet queuing delay in the client feedback information and other information then.Owing to measure the incipient stage we have no to understand to the available bandwidth situation of network initial, and do not know the residing scope of available bandwidth, therefore this moment we need send a lot of packets to network can with transmission bandwidth measure, the packet rate scope that sends may be between 1-100Mbps, even may be bigger, this depends on network type.We can obtain Preliminary study to the network bandwidth measurement by this at the beginning the time, for example our speed range of the packet that sends in network is 1-100Mbps at the beginning, and client receives the queuing delay that can obtain wrapping each packet in the string after this a string detection flows; Client feeds back to server with these data then, and the survey tool of server is calculated the available bandwidth estimated value of this time measuring according to the pathchirp algorithm, has so just obtained the Preliminary study to the network bandwidth.
When adopting the pathchirp method to calculate, in the ideal case, the packet queuing delay that client receives is a monotonically increasing, but owing to the background traffic that happens suddenly can occur, the queuing delay of each packet can not be monotonically increasing usually in a bag string, as shown in Figure 2.At this moment, pathchirp need utilize the signal shape of these queuing delays that the transmission bandwidth of each packet correspondence is an estimated value E k (m), then to a plurality of E k (m)Get weighted average, be used as the estimated value D of the transmission bandwidth of this bag string (m):
D ( m ) = &Sigma; k = 1 N - 1 E k ( m ) &Delta; k &Sigma; k = 1 N - 1 &Delta; k
At last to the estimated value D of a plurality of bag string correspondences of recording (m)Average, obtain the initial transmission bandwidth B (t) between server and the client.
In order to calculate E exactly k (m), pathchirp belongs to the zone of skew to the queuing delay signal segmentation of each packet one-tenth and does not belong to the zone of skew, and this will use pathchirp skew partitioning algorithm, if intuitively go up the q of continuous several bags k (m)(q k (m)The queuing delay of representing k packet in m the bag string) greater than 0 and increase, so these to be surrounded by may be the part of some busy periods of congested formation on the path.Concrete details is as follows, our target be to identify skew begin wrap sequence number i and end packet sequence number j, so each q occurs k (m)＜q K+1 (m)Bag i all might be the skew starting point; We define the end point j of skew for working as
q ( j ) - q ( i ) < max i &le; k &le; j [ q ( k ) - q ( i ) ] F
The time first bag, F is attenuation coefficient (decrease factor) in the formula; The queuing delay q (j) of ordering at j has been fallen by coefficient F increases to the maximum of j from i queuing delay.If j-i＞L (L is a constant, generally gets 5), then we i to the packet between the j as skew.
Estimated value E for each packet k (m), the arbitrary packet k in each bag string can be in a kind of in following three kinds of situations:
(a) if k belongs to skew and the q that a meeting stops k (m)＜q K+1 (m)The time, then
E k (m)＝R k
(b) if k belongs to the skew that can not stop, then
E k (m)＝R 1， &ForAll; k > l
L is the starting point of skew.
(c) for the packet k that does not belong to above two kinds of situations, then establish E k (m)=R 1This has comprised that those packets that do not belong to skew and those belong to skew but queuing delay is the packet that successively decreases, because last skew does not stop, therefore selects l=N-1.Here R kBe the speed of k packet, R 1Be the speed of the 1st packet, N is the number of contained packet in the bag string.
The numerical upper limits B of initial transmission bandwidth in the step (2) Th_HSpan be B Th_H=B (t)+m, the numerical lower limits B of initial transmission bandwidth Th_LSpan be B Th_L=B (t)-m, wherein m is 5%～10% of B (t).
Encoding scheme adopts the scalable video technology in the step (2), and encoding rate is tabled look-up according to the numerical value of current bandwidth and obtained.
T ' is the feedback time interval of client in the step (3), t 0Be the predicted time interval that server carries out the actual transmission bandwidth numerical prediction, t '＜t 0, t &prime; = [ 1 3 , 1 2 ] t 0 .
(the t+t of actual transmission bandwidth numerical value B described in the step (4) 0) calculate by following formula:
B ( t + t 0 ) = B ( t ) + &eta;B ( t ) &PartialD; f r ( t ) &PartialD; t ;
In the formula, f s(t) be the t encoding rate of server constantly; f r(t) be the t decoding rate of client constantly; &eta; = f s ( t ) f r ( t ) Be self study speed, the forecast updating amplitude of promptly current actual transmission bandwidth.
Server is with predicted time interval t in the step (5) 0After actual transmission bandwidth numerical value B (t+t 0) compare with initial transmission bandwidth numerical value B (t), judge whether to carry out once the prediction of the actual transmission bandwidth numerical value of adaptive digital home network again.If B (t+t 0) numerical value between the numerical value bound of B (t), then carry out the prediction of actual transmission bandwidth numerical value next time; If B (t+t 0) numerical value greater than the numerical upper limits B of B (t) Th_H, illustrate that current actual transmission bandwidth numerical value may be greater than the initial transmission bandwidth upper limit B of this bag string Th_H, server need be adjusted encoding scheme and improve encoding rate f s(t), thus improve bandwidth utilization; If B (t+t 0) numerical value less than the numerical lower limits B of B (t) Th_L, illustrate that current actual transmission bandwidth numerical value may be less than the initial transmission lower band B of this bag string Th_L, server need be adjusted encoding scheme and reduce encoding rate f s(t), thus guarantee that client can the smooth playing video.
As mentioned above, just can realize the present invention preferably, the foregoing description is preferred embodiment of the present invention only, is not to be used for limiting practical range of the present invention; Be that all equalizations of doing according to content of the present invention change and modification, all contained by claim of the present invention scope required for protection.

Claims (3)

1. adaptive digital home network flow media transmission band width Forecasting Methodology is characterized in that, may further comprise the steps:
(1) behind the system initialization, server needs to carry out the client transmission bag string that transmission bandwidth is measured in network, carries out transmission bandwidth and measures, and obtains the initial transmission bandwidth numerical value B (t) between server and the client;
(2) server is determined the numerical upper limits B of this transmission bandwidth according to the initial transmission bandwidth numerical value B (t) that obtains Th_HWith numerical lower limits B Th_L, and the encoding scheme and the encoding rate f of transmission bandwidth when being identified for this numerical value s(t);
(3) every time t ', client sends current decoding rate f to server in real time r(t), standby by server record; Server is according to fixed encoding scheme and encoding rate f simultaneously s(t) send Video stream information to client;
(4) every time t 0, server is according to the initial transmission bandwidth numerical value B (t), the encoding rate f that obtain s(t) and decoding rate f r(t), to t 0After actual transmission bandwidth numerical value predict, obtain t 0After actual transmission bandwidth numerical value B (t+t 0);
(5) server is with the actual transmission bandwidth numerical value B (t+t that obtains 0) respectively with the numerical upper limits B in initial transmission broadband Th_HAnd numerical lower limits B Th_LCompare; If comparative result is B Th_L≤ B (t+t 0)≤B Th_H, then jump to step (3), carry out the prediction of actual transmission bandwidth numerical value next time; If comparative result is B (t+t 0)＞B Th_HOr B (t+t 0)＜B Th_L, then jump to step (1), carry out the transmission bandwidth prediction of server and client again;
Wherein, server sends the bag string to client and adopts the pathchirp method in the described step (1), specifically may further comprise the steps:
(1-1) server carries out the client transmission bag string that transmission bandwidth is measured to needs, and the interval in the bag string between packet and the packet is exponential increasing trend;
After (1-2) client receives bag string, with the queuing delay of each packet in the bag string of receiving with whether measure successful feedback information and return server;
After (1-3) server receives feedback information, respectively the transmission bandwidth of each packet correspondence is estimated earlier, obtained the estimated value E of each packet correspondence k (m), pass through formula at last
D ( m ) = &Sigma; k = 1 N - 1 E k ( m ) &Delta; k &Sigma; k = 1 N - 1 &Delta; k
Estimated value E to each packet correspondence k (m)Get weighted average D (m), as the corresponding transmission bandwidth estimated value of this bag string; Wherein, k represents k packet, and m represents the m time measurement, Δ kRepresent the time interval between k packet and (k+1) individual packet, N is the number of contained packet in the bag string;
(1-4) last server is to the estimated value D of a plurality of bag string correspondences of recording (m)Average, obtain the initial transmission bandwidth numerical value B (t) between server and the client;
T ' described in the step (3) is the feedback time interval of client, t 0Be the predicted time interval that server carries out the actual transmission bandwidth numerical prediction, t '＜t 0,
Figure FSB00000594466800021
(the t+t of actual transmission bandwidth numerical value B described in the step (4) 0) calculate by following formula:
B ( t + t 0 ) = B ( t ) + &eta;B ( t ) &PartialD; f r ( t ) &PartialD; t ;
In the formula, f s(t) be the t encoding rate of server constantly; f r(t) be the t decoding rate of client constantly;
Figure FSB00000594466800023
Be self study speed, the forecast updating amplitude of promptly current actual transmission bandwidth.
2. according to the described adaptive digital home network flow media transmission band width Forecasting Methodology of claim 1, it is characterized in that the numerical upper limits B of initial transmission bandwidth described in the step (2) Th_HSpan be B Th_H=B (t)+m, the numerical lower limits B of initial transmission bandwidth Th_LSpan be B Th_L=B (t)-m, wherein m is 5%～10% of B (t).
3. according to the described adaptive digital home network flow media transmission band width Forecasting Methodology of claim 1, it is characterized in that encoding scheme described in the step (2) adopts the scalable video technology, described encoding rate is tabled look-up according to the numerical value of current bandwidth and is obtained.
CN2009100416868A 2009-08-05 2009-08-05 Prediction method of self-adapting digital home network flow media transmission band width Expired - Fee Related CN101631030B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009100416868A CN101631030B (en) 2009-08-05 2009-08-05 Prediction method of self-adapting digital home network flow media transmission band width

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009100416868A CN101631030B (en) 2009-08-05 2009-08-05 Prediction method of self-adapting digital home network flow media transmission band width

Publications (2)

Publication Number Publication Date
CN101631030A CN101631030A (en) 2010-01-20
CN101631030B true CN101631030B (en) 2011-12-07

Family

ID=41575989

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009100416868A Expired - Fee Related CN101631030B (en) 2009-08-05 2009-08-05 Prediction method of self-adapting digital home network flow media transmission band width

Country Status (1)

Country Link
CN (1) CN101631030B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101883281B (en) * 2010-06-13 2013-12-25 北京北大众志微系统科技有限责任公司 Static image coding method and system for remote display system
GB2519391B (en) * 2014-04-02 2015-10-21 Imagination Tech Ltd Enhanced media quality management
CN106331577B (en) * 2015-06-25 2020-08-14 阿里巴巴集团控股有限公司 Method and device for adjusting codes
CN105898403A (en) * 2015-11-16 2016-08-24 乐视云计算有限公司 Online media service code stream self-adaptive method and system
CN111277870B (en) * 2020-03-05 2022-09-30 广州市百果园信息技术有限公司 Bandwidth prediction method, device, server and storage medium
CN111447108B (en) * 2020-03-23 2022-12-06 中电福富信息科技有限公司 Progressive time delay bandwidth speed measurement analysis method and system based on deep learning
CN113163233B (en) * 2021-02-04 2022-11-22 福州大学 Bandwidth detection method based on real-time video streaming transmission
CN116614393B (en) * 2023-07-21 2023-11-28 中融云数(北京)技术有限公司 Weak network scene bandwidth estimation method and system based on RTC communication

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1848810A (en) * 2006-05-15 2006-10-18 武汉虹旭信息技术有限责任公司 Stream media transmitting rate controlling method
CN101297298A (en) * 2005-10-25 2008-10-29 皇家飞利浦电子股份有限公司 Method and apparatus for traffic flow control of streaming video data
CN101466124A (en) * 2007-12-19 2009-06-24 中国移动通信集团公司 Control method and system for network bandwidth

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101297298A (en) * 2005-10-25 2008-10-29 皇家飞利浦电子股份有限公司 Method and apparatus for traffic flow control of streaming video data
CN1848810A (en) * 2006-05-15 2006-10-18 武汉虹旭信息技术有限责任公司 Stream media transmitting rate controlling method
CN101466124A (en) * 2007-12-19 2009-06-24 中国移动通信集团公司 Control method and system for network bandwidth

Also Published As

Publication number Publication date
CN101631030A (en) 2010-01-20

Similar Documents

Publication Publication Date Title
CN101631030B (en) Prediction method of self-adapting digital home network flow media transmission band width
EP2288086B1 (en) Network monitoring device, bus system monitoring device, method and program
CN108293067B (en) Managing communication congestion for internet of things devices
US11350150B2 (en) Method for estimation of quality of experience (QoE) metrics for video streaming using passive measurements
CN101414957B (en) Buffering queue management method for wireless LAN
US20190386927A1 (en) Scheduled Transmission of Data
CN101138200A (en) Method and apparatus for evaluation of service quality of a real time application operating over a packet-based network
US20140169192A1 (en) System and Method for Estimating an Effective Bandwidth
CN100463416C (en) Method for testing maximum available bandwidth of network
CN102045222A (en) Real-time overall test method of network system
CN101414938B (en) Self-adapting real-time detection method for digital household network available bandwidth
WO2011156011A1 (en) Method and apparatus for congestion control
CN109167734A (en) The method and apparatus for identifying transmission control protocol state
Go et al. An SDN-based framework for improving the performance of underprovisioned IP Video Surveillance networks
Becchi From Poisson processes to self-similarity: a survey of network traffic models
US11190430B2 (en) Determining the bandwidth of a communication link
Zhao et al. A method for estimating the proportion of nonresponsive traffic at a router
Sheikhi et al. Congestion detection for video traffic in wireless sensor networks
GB2540568A (en) An adaptive real-time detecting method of available bandwidth in digital home networks
Kozlovskiy et al. DEVELOPMENT OF A MODIFIED METHOD OF NETWORK TRAFFIC FORMING
Ashaari et al. Performance Analysis on Multiple Device Connections of Small Office Home Office Network
WO2022244254A1 (en) Periodic flow detection device, delay inference device, periodic flow detection method, and program
Huang et al. End-to-end internet delay dynamics
CN104539486A (en) End-to-end available bandwidth estimating method based on TCP acknowledgement packet pair
Lin et al. Video streaming over in-home power line networks

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20111207

Termination date: 20170805