WO2016084334A1 - 可用帯域推定システム、可用帯域推定方法、受信装置及び受信装置のプログラムの記憶媒体 - Google Patents
可用帯域推定システム、可用帯域推定方法、受信装置及び受信装置のプログラムの記憶媒体 Download PDFInfo
- Publication number
- WO2016084334A1 WO2016084334A1 PCT/JP2015/005715 JP2015005715W WO2016084334A1 WO 2016084334 A1 WO2016084334 A1 WO 2016084334A1 JP 2015005715 W JP2015005715 W JP 2015005715W WO 2016084334 A1 WO2016084334 A1 WO 2016084334A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- packet
- delay
- calculated
- serialization
- train
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0882—Utilisation of link capacity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/06—Receivers
- H04B1/16—Circuits
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0852—Delays
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/56—Queue scheduling implementing delay-aware scheduling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
Definitions
- the present invention relates to an available bandwidth estimation system, an available bandwidth estimation method, a reception device, and a storage medium for a program of the reception device.
- the usable bandwidth (also called usable bandwidth) of a communication line is a free bandwidth obtained by subtracting other traffic (hereinafter referred to as cross traffic) flowing through the network from the physical bandwidth of the bottleneck link of the communication line.
- cross traffic other traffic
- Estimating the available bandwidth is important for video chat, videophone, video conference, etc., where two-way video is transmitted between terminals for communication.
- the reason is that by suppressing the video transmission rate below the estimated available bandwidth, packet loss caused by the sum of the video transmission rate and cross traffic exceeding the physical bandwidth of the bottleneck link of the communication line is suppressed. This is because video quality can be ensured.
- the transmission side apparatus sets a plurality of packets whose packet sizes gradually increase as a group (hereinafter referred to as a single packet) with the transmission interval of each packet for estimating the available bandwidth equal. , Referred to as a packet train). Further, the receiving side apparatus estimates the available bandwidth by detecting a change in the reception interval of each packet in the received packet train.
- the packet train 10 shown in FIG. 1 passes through the network and the packet transmission rate exceeds the usable bandwidth of the network, the packets in the packet train 10 are temporarily stored in devices such as routers and switches on the network. Queued (hereinafter, the queuing delay is called queuing delay).
- the packet reception interval at the reception side device becomes larger than the transmission interval at the transmission side device.
- the reception interval between the packet 3 and the packet 4 in the packet train 11 shown in FIG. 1 is larger than the transmission interval.
- the invention of Patent Document 1 uses this property to identify a location where the packet reception interval at the receiving side device begins to become larger than the transmission interval at the transmitting side device, and to determine the packet size corresponding to that location. Is divided by the transmission interval to calculate the available bandwidth.
- the delay that occurs in packet communication is as follows: 1. propagation delay; 2. processing delay; 3. queuing delay; There are four types of serialization delays.
- the serialization delay may be referred to as a transmission delay.
- Propagation delay is the time required for a signal to propagate through a communication line at a speed close to the speed of light.
- the processing delay is a time required for a device such as a router or a switch to determine which communication line to send a packet to by referring to a route table in the device when the packet is received.
- the queuing delay is as described above.
- the serialization delay is the time required to send all bits of a packet to the communication line, and can be calculated by packet size ⁇ physical band of the communication line. That is, the serialization delay increases as the packet size increases.
- Patent Document 1 is intended to prevent deterioration in estimation accuracy of available bandwidth in a network.
- the cross traffic fluctuates during transmission of the packet train
- the radio wave strength fluctuates in the wireless network
- Noise is generated.
- Patent Document 1 analyzes a general tendency of queuing delay of a plurality of packets from a macro viewpoint. To do. Accordingly, the invention of Patent Document 1 can reduce the influence of noise when specifying the rising position of the queuing delay.
- queuing delay ⁇ tr (i) ⁇ tr (i ⁇ 1) ⁇ ⁇ ⁇ ts (i) ⁇ ts (i ⁇ 1) ⁇ (equation 1) is calculated.
- an ideal queuing delay transition assumed when no noise is generated is a combination of a horizontal straight line and a straight line having an inclination.
- the invention of Patent Document 1 calculates the sum of square errors with the queuing delay calculated by Equation 1 for each of a plurality of ideal curves obtained by combining a horizontal straight line and a straight line having an inclination.
- the invention of Patent Document 1 specifies an ideal curve that minimizes the total sum of the calculated square errors.
- the invention of Patent Document 1 specifies a joining position of a horizontal straight line and an inclined straight line of the ideal curve that best fits the queuing delay as a rising position of the queuing delay from a macro viewpoint.
- the transition of the queuing delay Q (i) assumed when no noise is generated is a combination of a horizontal straight line and a parabola. The reason is as follows. First, since the queuing delay does not occur when the transmission rate for each packet is lower than the true value of the available bandwidth, the ideal transition of the queuing delay Q (i) becomes a horizontal straight line.
- FIG. 3 illustrates a method for specifying the rising position of the queuing delay Q (i) described above (hereinafter referred to as a second specifying method).
- the second specifying method plots the queuing delay Q (i) 12 on a plane having the packet number i on the horizontal axis and the queuing delay Q (i) on the vertical axis.
- the second specifying method uses an ideal curve 13 obtained by combining a horizontal straight line and a parabola as a transition of an ideal queuing delay Q (i) assumed when no noise is generated. Plot minutes. Note that there are (N ⁇ 1) ideal curves 13 when the number of packets in the packet train is N, but only five are displayed in FIG.
- the 2nd specific method calculates the sum total of the square error with queuing delay Q (i) 12 for every ideal curve 13.
- FIG. The second specifying method specifies an ideal curve 13-4 that minimizes the sum of the square errors as shown in FIG.
- the second identification method identifies the coupling point between the horizontal straight line of the ideal curve 13-4 and the parabola that best fits the observed cueing delay 12 as the rising position of the cueing delay from a macro viewpoint. To do.
- Patent Document 1 and the second specifying method do not consider the existence of a serialization delay even though the packet size increases and the serialization delay increases as the packet number advances. For this reason, the invention of Patent Document 1 and the second specifying method compare a queuing delay including a serialization delay with an ideal curve. That is, the invention of Patent Document 1 and the second specifying method may erroneously recognize the rising position of the queuing delay and output an incorrect estimated value.
- FIG. 4 shows a specific example of the second specifying method for specifying the rising position of the queuing delay Q (i).
- FIG. 4 shows the queuing delay 12 observed when a packet train consisting of about 100 packets is transmitted / received in a real network, but actually a serialization delay is mixed. The reason that the queuing delay 12 tends to rise linearly is that serialization delay is mixed. The reason why the slope of the queuing delay 12 is steeper near the end of the packet train (near the 85th packet) is that it is affected by the queuing delay.
- the ideal curve 13 is selected as the ideal curve that best fits. The packet around the 60th is erroneously recognized as the rising position of the queuing delay. And the said 2nd specific method outputs the estimated value used as an underestimation based on the packet around the 60th misrecognized as the rising position of a queuing delay. That is, the second specifying method cannot secure the estimation accuracy.
- the object of the present invention is to solve the above-mentioned problems, and even if a serialization delay is mixed in a predetermined delay to be observed, it is possible to specify an accurate rising position of a single queuing delay and ensure an estimation accuracy.
- An object of the present invention is to provide an available bandwidth estimation system, an available bandwidth estimation method, a reception device, and a storage medium for a program of the reception device.
- the available bandwidth estimation system of the present invention includes a transmission device and a reception device, and is an available bandwidth estimation system for estimating an available bandwidth of a network connecting the transmission device and the reception device, wherein the transmission device includes a packet A packet train configured by arranging packet pairs in which the size is increased at a constant rate and two packets of the same size are arranged at a transmission interval of zero at equal intervals; A serialization delay based on a reception interval between two packets constituting a packet pair included in the received packet train, and calculating an observation delay of a packet included in the received packet train. And the calculated serialization delay from the calculated observation delay It calculates the queuing delay by removing, based on the queuing delay the calculated estimates the available bandwidth, and wherein the.
- the available bandwidth estimation method of the present invention comprises a transmission device and a reception device, and in the available bandwidth estimation method for estimating an available bandwidth of a network connecting the transmission device and the reception device, A packet size is increased at a constant rate, a packet train configured by arranging packet pairs in which two packets of the same size are arranged at a transmission interval of zero is arranged at equal intervals, the packet train is received, Calculate an observation delay of a packet included in the received packet train, calculate a serialization delay based on a reception interval of two packets constituting a packet pair included in the received packet train, and calculate from the calculated observation delay A queuing delay is calculated by removing the calculated serialization delay, and an available bandwidth is estimated based on the calculated queuing delay.
- the receiving apparatus of the present invention receives a packet train configured by arranging packet pairs in which a packet size is increased at a constant rate and two packets of the same size are arranged at a transmission interval of zero at equal intervals, The observation delay of two packets constituting the packet pair included in the received packet train is calculated, and the serialization delay is calculated based on the reception interval of the two packets configuring the packet pair included in the received packet train.
- the storage medium of the program of the present invention is not temporary, and is configured by arranging packet pairs in which the packet size increases at a constant rate and two packets of the same size are arranged at a transmission interval of zero at equal intervals.
- the packet train is received, the observation delay of the packet included in the received packet train is calculated, and the serialization delay is calculated based on the reception interval of two packets constituting the packet pair included in the received packet train
- Calculation processing and receiving device Recording a control program of the receiving apparatus to be executed by a computer, characterized in that.
- the usable bandwidth estimation system, the usable bandwidth estimation method, the receiving device, and the storage medium of the program of the receiving device according to the present invention can accurately determine the rising position of a single queuing delay even if a serialization delay is mixed into the observed predetermined delay. This is advantageous in that the estimation accuracy can be ensured.
- the packet train transmitted / received by invention of patent document 1 is shown. Indicates the type of delay that occurs in packet communication.
- An example of a second specifying method for specifying the observed rising position of the queuing delay is shown.
- a specific example of the second specifying method for specifying the observed rising position of the queuing delay is shown.
- the structural example of the available band estimation system 50 of the 1st Embodiment of this invention is shown.
- the block block diagram of the receiver 30 of the 1st Embodiment of this invention is shown.
- movement flow of the available band estimation system 50 of the 1st Embodiment of this invention is shown.
- the structural example of the available band estimation system 50B of the 2nd Embodiment of this invention is shown.
- the block block diagram of the transmitter 20B of the 2nd Embodiment of this invention is shown.
- the information which the data storage means 202 of the 2nd Embodiment of this invention stores is shown.
- the block block diagram of the receiver 30B of the 2nd Embodiment of this invention is shown.
- the observation delay 80 in the 2nd Embodiment of this invention is shown.
- the reception interval of the even-numbered packet in the same packet pair and the odd-numbered packet in the 2nd Embodiment of this invention is shown.
- the observation delay and queuing delay whose packet number is odd in the 2nd Embodiment of this invention are shown.
- the information which the data storage means 303 of the 2nd Embodiment of this invention stores is shown.
- movement flow of the available band estimation system 50B of the 2nd Embodiment of this invention is shown.
- An example of the packet train transmitted / received by the available bandwidth estimation system 50B of the second exemplary embodiment of the present invention is shown.
- FIG. 5 shows a configuration example of the available bandwidth estimation system 50 of the first embodiment.
- the available bandwidth estimation system 50 includes a transmission device 20 and a reception device 30.
- the transmission device 20 and the reception device 30 are connected via a network 40 configured by the Internet, for example.
- the transmission device 20 transmits a packet train configured by arranging a predetermined number of packet pairs at regular intervals, the packet size increasing at a constant rate.
- the packet pair is a packet in which two packets having the same size are arranged with a transmission interval of zero.
- the receiving device 30 receives the packet train transmitted from the transmitting device 20, and calculates the observation delay of the packet included in the received packet train. In addition, the reception device 30 calculates a serialization delay based on the reception interval between two packets that form a packet pair included in the received packet train. The receiving device 30 calculates a queuing delay by removing the calculated serialization delay from the calculated observation delay. Then, the receiving device 30 estimates the available bandwidth based on the calculated queuing delay.
- FIG. 6 is a block diagram showing details of the receiving device 30 of the first embodiment.
- the receiving device 30 includes a serialization delay calculating unit 300, a serialization delay removing unit 301, and an estimated value calculating unit 302.
- the serialization delay calculation means 300 receives a packet train configured by arranging packet pairs in which the packet size is increased at a constant rate and two packets of the same size are arranged at a transmission interval of zero at equal intervals. Further, the serialization delay calculation unit 300 calculates the observation delay of the packet included in the received packet train. Further, the serialization delay calculation unit 300 calculates the serialization delay based on the reception interval between the two packets constituting the packet pair included in the received packet train.
- the serialization delay removing unit 301 calculates a queuing delay by removing the serialization delay calculated by the serialization delay calculating unit 300 from the observation delay calculated by the serialization delay calculating unit 300.
- the estimated value calculation means 302 estimates the available bandwidth based on the queuing delay calculated by the serialization delay removal means 301.
- Step 10 (hereinafter, step is expressed as S)
- the transmission apparatus 20 transmits a packet train configured by arranging a predetermined number of packet pairs whose packet sizes increase at a constant rate at equal intervals.
- the packet pair is a packet in which two packets having the same size are arranged with a transmission interval of zero.
- the receiving device 30 receives the packet train transmitted from the transmitting device 20, and calculates the observation delay of the packet included in the received packet train. In addition, the reception device 30 calculates a serialization delay based on the reception interval between two packets that form a packet pair included in the received packet train.
- the receiving device 30 calculates the queuing delay by removing the calculated serialization delay from the calculated observation delay. Then, the receiving device 30 estimates the available bandwidth based on the calculated queuing delay.
- the exact rising position of the single queuing delay is specified and the estimation accuracy is ensured. Can do.
- FIG. 8 shows a configuration example of the available bandwidth estimation system 50B of the second embodiment.
- the available bandwidth estimation system 50B includes a transmission device 20B and a reception device 30B.
- the transmission device 20B and the reception device 30B are connected via the network 40.
- the transmitting device 20B transmits predetermined information.
- the receiving device 30B receives predetermined information transmitted from the transmitting device 20B.
- the transmitting device 20B and the receiving device 30B are a personal computer (Personal Computer, PC), a portable computer (Personal Digital Assistant, PDA), a mobile phone, a smartphone, a tablet terminal, a fixed phone, a street multimedia terminal, an in-vehicle terminal, a network connection It may be a TV with a function, a set-top box with a network connection function, a game machine, a printer with a network connection function, a scanner with a network connection function, or other similar devices having a function of exchanging predetermined information with the outside.
- PC Personal Computer
- PDA Personal Digital Assistant
- the network 40 is a communication line that connects the transmission device 20B and the reception device 30B.
- the network 40 may be configured by the Internet.
- a device (not shown) other than the transmitting device 20B and the receiving device 30B may be connected to the network 40, and cross traffic may flow between these devices (not shown).
- FIG. 9 is a block diagram showing details of the transmission apparatus 20B of the second embodiment.
- the transmitting device 20B includes a packet pair generation unit 200, a packet train generation unit 201, and a data storage unit 202.
- the packet pair generation unit 200 generates a plurality of packet pairs based on the leading packet size 61, the number of packet pairs 62, and the packet size increment 63 stored in the data storage unit 202 described later.
- the packet pair is a packet in which two packets having the same size are brought into close contact with each other with a transmission interval of zero. Note that the packet size of each generated packet pair is different.
- the reason for providing a difference in the packet size of each packet pair as described above is to generate a packet train in which the packet size of the subsequent packet pair increases monotonously with respect to the packet size of the preceding packet pair.
- the packet train generation unit 201 collects a series of packet pairs generated by the packet pair generation unit 200 into one, and generates a packet train. In the packet train, the transmission interval of each packet pair is equal. Further, the increment of the packet size between each packet pair is a constant amount.
- the packet train generation unit 201 generates a packet train based on a packet pair transmission interval 60, a leading packet size 61, a packet pair number 62, and a packet size increment 63 stored in a data storage unit 202 described later.
- the packet train generation unit 201 transmits the generated packet train to the reception device 30B. At this time, the packet train generation means 201 transmits the packet in the packet train to be transmitted after embedding the time of the transmission device 20B immediately before transmission.
- the packet train generating means 201 may transmit the time after the time of the transmitting apparatus 20B immediately before transmission is embedded only in the first packet in the packet train.
- information on the transmission interval of the packet pair is stored in advance in the receiving device 30B, and the receiving device 30B calculates the transmission time of each packet based on the transmission time of the leading packet and the transmission interval of the packet pair. To do.
- the data storage means 202 stores predetermined information. More specifically, as shown in FIG. 10, the data storage unit 202 stores a packet pair transmission interval 60, a head packet size 61, a packet pair number 62, and a packet size increment 63.
- the packet pair transmission interval 60 is a transmission interval between packet pairs in the packet train to be transmitted (hereinafter referred to as T).
- the leading packet size 61 is the packet size of the leading packet pair in the packet train to be transmitted (hereinafter referred to as P1).
- the packet pair number 62 is the number of packet pairs in the packet train to be transmitted (hereinafter referred to as N).
- FIG. 11 is a block diagram showing details of the receiving device 30B of the second embodiment.
- the receiving device 30B includes a serialization delay calculation unit 300B, a serialization delay removal unit 301B, an estimated value calculation unit 302B, and a data storage unit 303.
- the serialization delay calculation unit 300B receives the packet train transmitted from the transmission device 20B. Then, the serialization delay calculation unit 300B calculates the observation delay 80 based on the received packet train.
- the observation delay 80 is a delay observed for a packet in the received packet train, and includes a predetermined delay related to packet communication such as a serialization delay and a queuing delay.
- Tr (j) represents the reception time of the j-th packet in the reception device 30B. As shown in FIG.
- the observation delay 80 is obtained by transmitting the first packet from the transmission device 20B from the elapsed time from the reception of the first packet by the reception device 30B until the reception of the jth packet. Is the value obtained by subtracting the elapsed time from when the j-th packet is transmitted.
- the observation delay 80 takes the form of a function with the packet number as an argument.
- serialization delay calculation unit 300B calculates a predetermined parameter for calculating the serialization delay based on the received packet train. Then, the serialization delay calculation unit 300B calculates a serialization delay based on the calculated parameters.
- the serialization delay calculation unit 300B calculates a reception interval between an odd-numbered packet that precedes and an even-numbered packet that follows in a packet pair included in the received packet train.
- the odd-numbered packet that precedes and the even-numbered packet that follows from within each packet pair are transmitted with a transmission interval of zero.
- the calculated reception interval is the value of only the serialization delay generated between the odd-numbered packet that precedes and the even-numbered packet that follows in each packet pair, or the sum of the serialization delay and the queuing delay.
- the serialization delay calculation means 300B is a graph in which the horizontal axis is an even packet number (2, 4,..., 2N) and the vertical axis is the calculated reception interval. Generate.
- the serialization delay calculation means 300B is a straight line having the gentlest positive slope passing through two points among the set of points of the reception interval 70 plotted in the graph, and above the straight line.
- a straight line 71 that satisfies the condition that all the points are in the region is calculated.
- the above two points are points representing the reception interval when only the serialization delay occurs.
- the calculated straight line 71 is a straight line indicating the relationship between the packet size and the value of the serialization delay generated in the packet having the packet size.
- the serialization delay calculation unit 300B calculates the slope and intercept of the calculated straight line 71 and stores them in the data storage unit 303.
- the serialization delay calculation unit 300 ⁇ / b> B calculates the serialization delay generated in the odd-numbered packet using the calculated slope and intercept of the straight line 71 and stores it in the data storage unit 303.
- Equation 3 “a” represents the slope of the straight line 71, “b” represents the intercept of the straight line 71, “x” represents the packet number, and “y” represents the reception interval (serialization delay).
- the serialization delay calculation means 300B substitutes 3 for “x” in Equation 3 to calculate the serialization delay generated in the third packet with the packet size of P1 + 2 ⁇ ⁇ P.
- the serialization delay calculation unit 300B performs the same calculation for all odd-numbered packets and calculates the serialization delay generated in each odd-numbered packet.
- the serialization delay removing unit 301B calculates a pure queuing delay 82 by removing the influence of the serialization delay from the observation delay 80 calculated by the serialization delay calculating unit 300B. More specifically, as shown in FIG. 14, the serialization delay removing unit 301B first has an abscissa indicating an odd packet number (3, 5,..., 2N ⁇ 1) and an ordinate indicating an observation delay 80 Generate a graph that is At this time, the observation delay 80 includes predetermined delays related to packet communication such as queuing delay and serialization delay. The serialization delay removing unit 301B subtracts the serialization delay value calculated by the serialization delay calculating unit 300B from the value of the observation delay 80 of the odd-numbered packet.
- the serialization delay removing unit 301B can calculate a pure queuing delay (hereinafter referred to as q (i)) 82 which is not affected by the serialization delay.
- the serialization delay removing unit 301B stores the calculated q (i) in the data storage unit 303.
- the serialization delay removing unit 301B subtracts the value of the serialization delay y related to the third packet from the value of the observation delay d (3) related to the third packet, so that the pure queuing delay q without the influence of the serialization delay is obtained. (3) is calculated.
- the serialization delay calculation means 300B performs the same calculation for all the values of the observation delay 80 of the odd-numbered packet, and calculates the queuing delay 82 of the odd-numbered packet.
- the estimated value calculating unit 302B analyzes the queuing delay 82 calculated by the serialization delay removing unit 301B, calculates an available band estimated value 83, and stores it in the data storage unit 303. More specifically, as shown in FIG. 14, the estimated value calculation means 302B indicates that the abscissa indicates an odd packet number (3, 5,..., 2N-1), and the ordinate indicates the queuing calculated above. A graph with delay 82 is generated. Then, the estimated value calculation means 302B has (N ⁇ 1) ideal curves obtained by combining a horizontal straight line and a parabola as an ideal queuing delay transition assumed when no noise or the like occurs. Is plotted on the graph created above.
- the parabola of each ideal curve can be expressed in an integrated form of a straight line ( ⁇ P ⁇ Ts / Pk) ⁇ m + (P1 ⁇ P) ⁇ Ts / Pk having an inclination described in Patent Document 1.
- Pk represents the kth packet size.
- K represents an odd packet number when the packet transmission rate exceeds the true value of the available bandwidth for the first time.
- the estimated value calculation means 302B calculates the sum of the square error with the calculated queuing delay 82 for each of the plotted ideal curves, and as shown in FIG. Is identified.
- the estimated value calculation means 302B calculates the packet number of the connection point between the horizontal straight line and the parabola in the identified ideal curve 84.
- the estimated value calculation unit 302B divides the packet size at the calculated packet number by the transmission interval T to calculate an estimated value of the usable bandwidth.
- the data storage means 303 stores predetermined information. More specifically, the data storage unit 303 stores an observation delay 80, a serialization delay 81, a queuing delay 82, and an estimated value 83, as shown in FIG.
- the observation delay 80 is a delay generated in the received packet in the packet train, and is calculated by the above-described serialization delay calculation unit 300B.
- the observation delay 80 includes a predetermined delay related to packet communication such as a serialization delay and a queuing delay.
- the serialization delay 81 is a value of the serialization delay generated in the odd-numbered packet, calculated based on the slope and intercept of the straight line 71 by the serialization delay calculation unit 300B.
- the queuing delay 82 is a pure queuing delay obtained by removing the influence of the serialization delay from the observation delay 80. As described above, the queuing delay 82 is calculated by the serialization delay removing unit 301B.
- the estimated value 83 is an estimated value of the available bandwidth calculated by the estimated value calculating unit 302B.
- the packet pair generation unit 200 of the transmission device 20B generates N packet pairs that are composed of two packet pairs that have the same packet size and are closely attached with a transmission interval of zero. Note that the packet size of each packet pair is different.
- the packet train generation unit 201 of the transmission apparatus 20B combines the generated N packet pairs into one to generate a packet train.
- the transmission interval of each packet pair is equal. Further, the increment of the packet size between each packet pair is a constant amount.
- the transmission device 20B transmits the generated packet train to the reception device 30B.
- FIG. 17 shows an example of a packet train 90 transmitted by the packet train generation means 201.
- the packet train 90 shown in FIG. 17 is composed of four packet pairs. Two packets in the same packet pair have the same size. For example, the packet sizes of the first packet pair, packet 1 and packet 2, are both P1. Further, the packets in each packet pair are transmitted with a transmission interval of zero. For example, after transmitting packet 1 in the first packet pair, transmitting apparatus 20B transmits packet 2 with a transmission interval of zero. Further, the transmission interval of each packet pair is T at a constant interval. For example, the transmission device 20B transmits the packet 2 and transmits the packet 3 after T has elapsed. Further, the amount of increase in packet size between packet pairs is a fixed amount. For example, the amount of increase in the packet size in the first packet pair and the second packet pair is ⁇ P.
- the serialization delay calculation unit 300B of the reception device 30B receives the packet train transmitted from the transmission device 20B, and calculates the observation delay 80 of the packet in the received packet train.
- FIG. 17 shows an example of the packet train 91 received by the serialization delay calculation unit 300B.
- the received packet train 91 shows a case where a queuing delay occurs between the packet 4 and the packet 5.
- serialization delays having different values are generated between the packet 1 and the packet 2 and between the packet 3 and the packet 4.
- a serialization delay also occurs between the packet 2 and the packet 3. Therefore, the reception interval of packet 2 and packet 3 is the sum of transmission interval T and serialization delay.
- both a queuing delay and a serialization delay are generated between the packet 4 and the packet 5. Therefore, the reception interval of the packets 4 and 5 is the sum of the transmission interval T of the packet pair, the queuing delay, and the serialization delay.
- both a queuing delay and a serialization delay occur between the packets after the packet 5.
- the serialization delay calculation means 300B of the receiving device 30B has an even packet number (2, 4,..., 2N) on the horizontal axis and an odd packet that precedes and an even packet that follows in the packet pair. Generate a graph with the reception interval between. For example, in the case of the received packet train 91 shown in FIG. 17, the serialization delay calculation unit 300B calculates the reception interval between the packet 1 and the packet 2 based on the reception time of the packet 1 and the reception time of the packet 2. In this case, the calculated reception interval is a serialization delay that occurs between packet 1 and packet 2.
- the reception interval calculated based on the reception time of the packet 5 and the reception time of the packet 6 is a total value of the serialization delay and the queuing delay that occur between the packet 5 and the packet 6. That is, the graph generated by the serialization delay calculation unit 300B is a graph in which the serialization delay value generated between the packets in the packet pair or the total value of the serialization delay and the queuing delay is plotted.
- the serialization delay calculation unit 300B is a straight line having the most gradual positive slope passing through two points among the set of reception interval points plotted on the generated graph, and more than the straight line.
- a straight line 71 that satisfies the condition that all the points are in the upper region is calculated.
- the serialization delay calculation unit 300B calculates the slope and intercept of the straight line 71.
- the serialization delay calculation unit 300B stores the calculated slope and intercept of the straight line 71 in the data storage unit 303.
- the calculated slope and intercept of the straight line 71 are used when calculating the serialization delay generated in the received packet.
- serialization delay occurs between the packets 1 and 2 and between the packets 3 and 4. Further, serialization delay and queuing delay occur between the packets 5 and 6 and between the packets 7 and 8. That is, a straight line connecting the reception intervals of the packet 1 and the packet 2 in which no queuing delay is mixed and the reception intervals of the packet 3 and the packet 4 is a straight line that satisfies the above condition.
- the serialization delay calculation means 300B calculates the serialization delay generated in the odd-numbered packet using the calculated slope and intercept of the straight line 71.
- the serialization delay removing unit 301B of the receiving device 30B subtracts the serialization delay value calculated by the serialization delay calculating unit 300B from the value of the observation delay 80 of the odd-numbered packet.
- the serialization delay removing unit 301B calculates a pure queuing delay 82 that is not affected by the serialization delay.
- the serialization delay removing unit 301B stores the calculated pure queuing delay 82 in the data storage unit 303.
- the serialization delay removal unit 301B subtracts the value of the serialization delay generated in the third packet calculated by the serialization delay calculation unit 300B from the value of the observation delay d (3), and does not affect the serialization delay. Ing delay q (3) is calculated.
- the estimated value calculation means 302B of the receiving device 30B plots the queuing delay 82, obtains an ideal curve that best fits the queuing delay 82, and specifies the rising position of the queuing delay.
- the estimated value calculation unit 302B divides the packet size corresponding to the specified position by the transmission interval T of the packet pair, and calculates an estimated value of the available bandwidth. Then, the estimated value calculation unit 302B stores the calculated estimated value of the available bandwidth in the data storage unit 303.
- the transmission device 20B generates a packet train in which a plurality of packets are combined into one, and transmits this to the reception device 30B.
- the transmission interval of each packet pair is equal.
- the increment of the packet size between each packet pair is a constant amount.
- the receiving device 30B calculates the observation delay observed in the packets in the packet train based on the packet train received from the transmitting device 20B.
- the receiving device 30B calculates a serialization delay based on the reception interval between the two packets that form the packet pair.
- the receiving device 30B calculates a pure queuing delay that is not affected by the serialization delay by subtracting the serialization delay from the observation delay, and estimates the available bandwidth based on the calculated queuing delay.
- the second embodiment of the present invention even when a serialization delay is mixed in a predetermined delay to be observed, it is possible to identify an accurate rising position of a single queuing delay and to ensure estimation accuracy.
- the transmission rate of the video is estimated as an available bandwidth. The following can be suppressed. That is, according to the second embodiment of the present invention, the packet loss caused by the sum of the video transmission rate and the cross traffic exceeding the physical bandwidth of the bottleneck link of the communication line is suppressed, and the video quality is ensured. can do.
- the serialization delay calculation unit 300B calculates a straight line 71 that satisfies a predetermined condition.
- the predetermined condition is a straight line having the slowest positive slope passing through two points in the set of reception interval points plotted on the generated graph, and all the regions above the straight line It is a condition that there is a point.
- the calculation method of the straight line 71 is not limited to the above method.
- the serialization delay calculation unit 300B may perform linear regression on all plotted points and calculate the straight line 71 and the slope and intercept of the straight line 71.
- the receiving device 30B may include a control unit (not shown), and the control unit may control each unit of the receiving device 30B.
- the control means may be constituted by, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. Further, for example, the CPU of the control means may control each means based on a program read from the ROM. Further, the CPU of the control means may read the program from a storage medium storing a computer-readable program using a reading device (not shown) and execute control of each means.
- a control unit may control each unit of the receiving device 30B.
- the control means may be constituted by, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. Further, for example, the CPU of the control means may control each means based on a program read from the ROM. Further, the CPU of the control means may read the program from a storage medium storing a computer
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
パケットサイズが一定の割合で増加し、同一サイズの2つのパケットが送信間隔ゼロで配置されたパケットペアを等間隔で配置することによって構成されたパケットトレインを送信し、前記パケットトレインを受信し、前記受信したパケットトレインに含まれるパケットの観測遅延を算出し、前記受信したパケットトレインに含まれるパケットペアを構成する2つのパケットの受信間隔に基づき、シリアライゼーション遅延を算出し、前記算出した観測遅延から前記算出したシリアライゼーション遅延を除去することでキューイング遅延を算出し、前記算出したキューイング遅延に基づき、可用帯域を推定することを特徴とする。
本願発明の第1の実施形態について説明する。
送信装置20は、パケットサイズが一定の割合で増加する、所定の数のパケットペアを等間隔で配置することによって構成されたパケットトレインを送信する。ここで、パケットペアは、同一サイズの2つのパケットが送信間隔ゼロで配置されたものである。
受信装置30は、送信装置20から送信されるパケットトレインを受信し、受信したパケットトレインに含まれるパケットの観測遅延を算出する。また、受信装置30は、受信したパケットトレインに含まれる、パケットペアを構成する2つのパケットの受信間隔に基づき、シリアライゼーション遅延を算出する。
受信装置30は、算出した観測遅延から、算出したシリアライゼーション遅延を除去することでキューイング遅延を算出する。そして、受信装置30は、算出したキューイング遅延に基づき、可用帯域を推定する。
本願発明の第2の実施形態について、図面を参照して詳細に説明する。
送信装置20Bのパケットペア生成手段200は、パケットサイズが同一であり、送信間隔ゼロで密着する2つのパケットのペアで構成されるパケットペアをN個生成する。なお、各パケットペアのパケットサイズはそれぞれ異なるものとする。
受信装置30Bのシリアライゼーション遅延算出手段300Bは、送信装置20Bから送信されるパケットトレインを受信し、受信したパケットトレイン内のパケットの観測遅延80を算出する。図17は、シリアライゼーション遅延算出手段300Bが受信したパケットトレイン91の一例を示している。受信したパケットトレイン91は、パケット4とパケット5との間においてキューイング遅延が発生した場合を示している。ここで、パケット1及びパケット2の間、パケット3及びパケット4の間には、それぞれ異なる値のシリアライゼーション遅延が発生している。また、パケット2及びパケット3の間にも、シリアライゼーション遅延が発生している。そのため、パケット2及びパケット3の受信間隔は、送信間隔Tと、シリアライゼーション遅延とを合算したものとなる。また、パケット4及びパケット5の間には、キューイング遅延及びシリアライゼーション遅延の両方が発生している。そのため、パケット4及びパケット5の受信間隔は、パケットペアの送信間隔Tと、キューイング遅延と、シリアライゼーション遅延とを合算したものとなる。また、パケット5以降の各パケット間には、キューイング遅延及びシリアライゼーション遅延の両方が生じている。
受信装置30Bのシリアライゼーション遅延算出手段300Bは、横軸を偶数のパケット番号(2、4、...、2N)、縦軸を各パケットペア内で先発となる奇数パケットと後発となる偶数パケットとの間の受信間隔とするグラフを生成する。例えば、図17に示す受信したパケットトレイン91の場合、シリアライゼーション遅延算出手段300Bは、パケット1の受信時刻及びパケット2の受信時刻に基づき、パケット1とパケット2との間の受信間隔を算出する。この場合、算出した受信間隔は、パケット1及びパケット2の間に生じたシリアライゼーション遅延となる。また、パケット5の受信時刻及びパケット6の受信時刻に基づき算出される受信間隔は、パケット5とパケット6との間に生じたシリアライゼーション遅延とキューイング遅延との合算値となる。すなわち、シリアライゼーション遅延算出手段300Bが生成するグラフは、パケットペア内のパケット間で発生した、シリアライゼーション遅延の値、または、シリアライゼーション遅延及びキューイング遅延の合算値をプロットしたグラフとなる。
受信装置30Bのシリアライゼーション遅延除去手段301Bは、奇数番号のパケットの観測遅延80の値から、シリアライゼーション遅延算出手段300Bが算出したシリアライゼーション遅延の値を減算する。これにより、シリアライゼーション遅延除去手段301Bは、シリアライゼーション遅延の影響のない、純粋なキューイング遅延82を算出する。そして、シリアライゼーション遅延除去手段301Bは、算出した純粋なキューイング遅延82をデータ記憶手段303に格納する。例えば、シリアライゼーション遅延除去手段301Bは、観測遅延d(3)の値から、シリアライゼーション遅延算出手段300Bが算出した3番目のパケットに発生したシリアライゼーション遅延の値を減算し、シリアライゼーション遅延の影響のない、キューイング遅延q(3)を算出する。
受信装置30Bの推定値算出手段302Bは、キューイング遅延82をプロットし、キューイング遅延82に対して最も良くフィットする理想曲線を求め、キューイング遅延の立ち上がり位置を特定する。また、推定値算出手段302Bは、特定した位置に対応するパケットサイズをパケットペアの送信間隔Tで除算し、可用帯域の推定値を算出する。そして、推定値算出手段302Bは、算出した可用帯域の推定値をデータ記憶手段303に格納する。
11、91 受信したパケットトレイン
12、82 キューイング遅延
13、84 理想曲線
20、20B 送信装置
200 パケットペア生成手段
201 パケットトレイン生成手段
202、303 データ記憶手段
30、30B 受信装置
300、300B シリアライゼーション遅延算出手段
301、301B シリアライゼーション遅延除去手段
302、302B 推定値算出手段
40 ネットワーク
50、50B 可用帯域推定システム
60 パケットペア送信間隔
61 先頭パケットサイズ
62 パケットペア個数
63 パケットサイズ増分
70 受信間隔
71 直線
80 観測遅延
81 シリアライゼーション遅延
82 キューイング遅延
83 推定値
Claims (9)
- 送信装置と受信装置とを備え、前記送信装置と前記受信装置とを接続するネットワークの可用帯域を推定する可用帯域推定システムであって、
前記送信装置は、パケットサイズが一定の割合で増加し、同一サイズの2つのパケットが送信間隔ゼロで配置されたパケットペアを等間隔で配置することによって構成されたパケットトレインを送信し、
前記受信装置は、前記送信装置から送信されるパケットトレインを受信し、前記受信したパケットトレインに含まれるパケットの観測遅延を算出し、前記受信したパケットトレインに含まれるパケットペアを構成する2つのパケットの受信間隔に基づき、シリアライゼーション遅延を算出し、前記算出した観測遅延から前記算出したシリアライゼーション遅延を除去することでキューイング遅延を算出し、前記算出したキューイング遅延に基づき、可用帯域を推定する、
ことを特徴とする可用帯域推定システム。 - 前記受信装置は、パケットペア順に並べた、前記パケットペアを構成する2つのパケットの受信間隔の点の集合のうち、任意の2つの点を通る、最も緩やかな正の傾きを持つ直線であり、かつ、前記直線よりも上の領域に全ての点があるという条件を満たす直線を算出し、前記算出した直線の傾き及び切片に基づき、前記シリアライゼーション遅延を算出することを特徴とする請求項1に記載の可用帯域推定システム。
- 前記受信装置は、前記算出した観測遅延から、前記直線の傾き及び切片に基づき算出される前記シリアライゼーション遅延を減算し、前記キューイング遅延を算出することを特徴とする請求項2に記載の可用帯域推定システム。
- 前記受信装置は、前記受信したパケットトレインに含まれる同一パケットペア内の、先発となる奇数パケットの受信時刻と、後発となる偶数パケットの受信時刻との差分に基づき、前記パケットの受信間隔を算出することを特徴とする請求項1乃至3のいずれか1項に記載の可用帯域推定システム。
- 前記受信装置は、前記キューイング遅延に影響を与えるノイズが存在しない理想的な状況におけるキューイング遅延を表現する理想曲線を算出し、
前記算出したキューイング遅延と前記理想曲線の数式との間の二乗誤差を算出し、
前記算出した二乗誤差が最小となる理想曲線に基づき、前記可用帯域の推定を行うことを特徴とする請求項1乃至4のいずれか1項に記載の可用帯域推定システム。 - 前記受信装置は、パケットの送信レートが可用帯域の真値を超えない範囲では、傾きがゼロの第1の直線とし、パケットの送信レートが可用帯域の真値を超える範囲では、傾きがパケットサイズの増分と送信間隔との積を送信レートが初めて可用帯域の真値を超えるパケットのパケットサイズで除算した値である第2の直線を積分した放物線とし、前記第1の直線と前記放物線とを結合した前記理想曲線を算出することを特徴とする請求項5に記載の可用帯域推定システム。
- 送信装置と受信装置とを備え、前記送信装置と前記受信装置とを接続するネットワークの可用帯域を推定する可用帯域推定方法において、
パケットサイズが一定の割合で増加し、同一サイズの2つのパケットが送信間隔ゼロで配置されたパケットペアを等間隔で配置することによって構成されたパケットトレインを送信し、
前記パケットトレインを受信し、前記受信したパケットトレインに含まれるパケットの観測遅延を算出し、前記受信したパケットトレインに含まれるパケットペアを構成する2つのパケットの受信間隔に基づき、シリアライゼーション遅延を算出し、前記算出した観測遅延から前記算出したシリアライゼーション遅延を除去することでキューイング遅延を算出し、前記算出したキューイング遅延に基づき、可用帯域を推定する、
ことを特徴とする可用帯域推定方法。 - パケットサイズが一定の割合で増加し、同一サイズの2つのパケットが送信間隔ゼロで配置されたパケットペアを等間隔で配置することによって構成されたパケットトレインを受信し、前記受信したパケットトレインに含まれるパケットの観測遅延を算出し、前記受信したパケットトレインに含まれるパケットペアを構成する2つのパケットの受信間隔に基づき、シリアライゼーション遅延を算出するシリアライゼーション遅延算出手段と、
前記算出した観測遅延から前記算出したシリアライゼーション遅延を除去することでキューイング遅延を算出するシリアライゼーション遅延除去手段と、
前記算出したキューイング遅延に基づき、可用帯域を推定する推定値算出手段と、
を備えることを特徴とする受信装置。 - パケットサイズが一定の割合で増加し、同一サイズの2つのパケットが送信間隔ゼロで配置されたパケットペアを等間隔で配置することによって構成されたパケットトレインを受信し、前記受信したパケットトレインに含まれるパケットの観測遅延を算出し、前記受信したパケットトレインに含まれるパケットペアを構成する2つのパケットの受信間隔に基づき、シリアライゼーション遅延を算出するシリアライゼーション遅延算出処理と、
前記算出した観測遅延から前記算出したシリアライゼーション遅延を除去することでキューイング遅延を算出するシリアライゼーション遅延除去処理と、
前記算出したキューイング遅延に基づき、可用帯域を推定する推定値算出処理と、
を受信装置のコンピュータに実行させるための受信装置の制御プログラムを記録した、一時的でない、プログラムの記憶媒体。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016561232A JP6575529B2 (ja) | 2014-11-27 | 2015-11-17 | 可用帯域推定システム、可用帯域推定方法、受信装置及び受信装置の制御プログラム |
US15/527,805 US10243825B2 (en) | 2014-11-27 | 2015-11-17 | Available bandwidth estimation system, available bandwidth estimation method, reception apparatus, and reception apparatus program storage medium |
EP15862836.2A EP3226489A4 (en) | 2014-11-27 | 2015-11-17 | Available-band estimation system, available-band estimation method, reception apparatus, and reception apparatus program storage medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014239480 | 2014-11-27 | ||
JP2014-239480 | 2014-11-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016084334A1 true WO2016084334A1 (ja) | 2016-06-02 |
Family
ID=56073929
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/005715 WO2016084334A1 (ja) | 2014-11-27 | 2015-11-17 | 可用帯域推定システム、可用帯域推定方法、受信装置及び受信装置のプログラムの記憶媒体 |
Country Status (4)
Country | Link |
---|---|
US (1) | US10243825B2 (ja) |
EP (1) | EP3226489A4 (ja) |
JP (1) | JP6575529B2 (ja) |
WO (1) | WO2016084334A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018037782A (ja) * | 2016-08-30 | 2018-03-08 | ソフトバンク株式会社 | 通信網の可用帯域を推定するシステム及びその方法 |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPWO2019003955A1 (ja) * | 2017-06-29 | 2020-04-30 | ソニー株式会社 | 通信システムおよび制御装置 |
WO2020091798A1 (en) * | 2018-11-01 | 2020-05-07 | Hewlett Packard Enterprise Development Lp | Available network bandwidth estimation using a one-way-delay noise filter with bump detection |
CN111756595B (zh) | 2019-03-29 | 2022-03-25 | 华为技术有限公司 | 可用带宽的测量方法和通信装置 |
CN111224820B (zh) * | 2019-12-30 | 2023-10-13 | 视联动力信息技术股份有限公司 | 一种带宽计算的方法和装置 |
US11153192B2 (en) * | 2020-02-29 | 2021-10-19 | Hewlett Packard Enterprise Development Lp | Techniques and architectures for available bandwidth estimation with packet pairs selected based on one-way delay threshold values |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009118602A2 (en) * | 2008-03-26 | 2009-10-01 | Telefonaktiebolaget L M Ericsson (Publ) | Available bandwidth estimation in a packet-switched communication network |
WO2014017140A1 (ja) * | 2012-07-27 | 2014-01-30 | 日本電気株式会社 | 可用帯域推定システム、方法及びプログラム |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5685913B2 (ja) * | 2009-12-11 | 2015-03-18 | 日本電気株式会社 | ネットワーク帯域計測システム、ネットワーク帯域計測方法およびプログラム |
JP6048102B2 (ja) | 2012-12-10 | 2016-12-21 | 日本電気株式会社 | 情報処理システム |
-
2015
- 2015-11-17 US US15/527,805 patent/US10243825B2/en active Active
- 2015-11-17 JP JP2016561232A patent/JP6575529B2/ja active Active
- 2015-11-17 WO PCT/JP2015/005715 patent/WO2016084334A1/ja active Application Filing
- 2015-11-17 EP EP15862836.2A patent/EP3226489A4/en not_active Withdrawn
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009118602A2 (en) * | 2008-03-26 | 2009-10-01 | Telefonaktiebolaget L M Ericsson (Publ) | Available bandwidth estimation in a packet-switched communication network |
WO2014017140A1 (ja) * | 2012-07-27 | 2014-01-30 | 日本電気株式会社 | 可用帯域推定システム、方法及びプログラム |
Non-Patent Citations (1)
Title |
---|
See also references of EP3226489A4 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018037782A (ja) * | 2016-08-30 | 2018-03-08 | ソフトバンク株式会社 | 通信網の可用帯域を推定するシステム及びその方法 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2016084334A1 (ja) | 2017-09-07 |
US10243825B2 (en) | 2019-03-26 |
JP6575529B2 (ja) | 2019-09-18 |
US20170324635A1 (en) | 2017-11-09 |
EP3226489A4 (en) | 2018-06-06 |
EP3226489A1 (en) | 2017-10-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6575529B2 (ja) | 可用帯域推定システム、可用帯域推定方法、受信装置及び受信装置の制御プログラム | |
CN112217686B (zh) | 一种评估往返时延的系统、方法及装置 | |
CN102484602B (zh) | 一种用于测量网络传输容量的方法、设备和系统 | |
JP6135670B2 (ja) | 可用帯域推定システム、方法及びプログラム | |
US10805196B2 (en) | Packet loss and bandwidth coordination | |
US8582453B2 (en) | System for measuring the transmission bandwidth for multimedia streaming and method for same | |
WO2011040006A1 (ja) | 送信装置、受信装置、送受信システム、及び送受信方法 | |
US8873590B2 (en) | Apparatus and method for correcting jitter | |
WO2012170738A1 (en) | Multipath rate adaptation | |
Daldoul et al. | Performance and scalability evaluation of IEEE 802.11 v/aa multicast transport | |
JP6468193B2 (ja) | 通信装置及び通信装置におけるネットワーク利用可能帯域推定方法、並びにネットワーク利用可能帯域推定プログラム | |
US20120191878A1 (en) | Synchronization system and method | |
CN110169134A (zh) | 无线质量支持视频体验的检测方法及装置 | |
US8243607B2 (en) | Packet delay characteristic measuring apparatus and method | |
JPWO2018212085A1 (ja) | 送信装置、可用帯域推定装置、可用帯域推定システム、方法 | |
JP4800250B2 (ja) | 必要十分な受信バッファサイズを決定するパケット受信装置、方法及びプログラム | |
WO2016117319A1 (ja) | 可用帯域計測システム、可用帯域計測方法、送信側装置、受信側装置、送信側装置制御プログラムが記憶された記憶媒体および受信側装置制御プログラムが記憶された記憶媒体 | |
WO2016103674A1 (ja) | ストリーム受信装置、通信システム、ストリーム送信のタイミングを推定する方法および記録媒体 | |
Nihei et al. | A QoE indicator and a transmission control method for VoIP on mobile networks considering delay spikes | |
JP2006211233A (ja) | Ip通信装置 | |
KR101107325B1 (ko) | 실시간 멀티미디어 서비스에 대한 네트워크 구간별 품질측정 방법 및 시스템 | |
JP4541381B2 (ja) | 映像送受信システム及び映像切替方法 | |
JP6717209B2 (ja) | パケット列送・受信装置、ネットワーク状態推定システムおよび制御方法 | |
CN104243238A (zh) | 测试控制平面限速值的方法、测试设备及系统 | |
JP2009171453A (ja) | 送信ノード、タイトリンク特定方法およびタイトリンク特定プログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15862836 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016561232 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15527805 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
REEP | Request for entry into the european phase |
Ref document number: 2015862836 Country of ref document: EP |