WO2013008387A1 - パラメータ推定装置、パラメータ推定方法、及び、パラメータ推定プログラム - Google Patents
パラメータ推定装置、パラメータ推定方法、及び、パラメータ推定プログラム Download PDFInfo
- Publication number
- WO2013008387A1 WO2013008387A1 PCT/JP2012/003804 JP2012003804W WO2013008387A1 WO 2013008387 A1 WO2013008387 A1 WO 2013008387A1 JP 2012003804 W JP2012003804 W JP 2012003804W WO 2013008387 A1 WO2013008387 A1 WO 2013008387A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- communication throughput
- parameter
- time
- parameter estimation
- range
- 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
- H04L43/0888—Throughput
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/04—Processing captured monitoring data, e.g. for logfile generation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/25—Flow control; Congestion control with rate being modified by the source upon detecting a change of network conditions
Definitions
- the present invention relates to a communication throughput prediction apparatus that estimates communication throughput and a control parameter determination apparatus that determines control parameters used for transmitting data.
- a communication throughput prediction device that estimates a communication throughput that is a data size (amount of data) distributed (transmitted) per unit time is known.
- the communication throughput prediction apparatus described in Patent Document 1 transmits test data having a predetermined data size, and completes transmission of the test data. Measure the time required. The communication throughput prediction apparatus predicts future communication throughput based on the measured time.
- TCP / IP Transmission Control Protocol / Internet Protocol
- Etc. End-to-End delay, packet loss, cross traffic, and radio wave intensity in wireless communication.
- the communication throughput prediction apparatus after the transmission of test data is completed, the communication throughput decreases due to an increase in cross traffic, or the communication throughput increases due to an increase in radio wave intensity in wireless communication. There is a case. In such a case, the communication throughput prediction apparatus has a problem that the communication throughput cannot be estimated with high accuracy.
- an object of the present invention is to provide a parameter estimation apparatus that can solve the above-described problem that “accuracy regarding estimation of communication throughput may be excessively low”.
- a parameter estimation apparatus is Communication throughput acquisition means for acquiring communication throughput, which is the amount of data transmitted per unit time; Based on the communication throughput acquired up to the first time point, a function specifying parameter for specifying a probability density function using the communication throughput at the second time point after the first time point as a random variable is estimated.
- the parameter estimation method includes: Get the communication throughput, which is the amount of data transmitted per unit time, Based on the communication throughput acquired up to the first time point, a function specifying parameter for specifying a probability density function using the communication throughput at the second time point after the first time point as a random variable is estimated. It is a method to do.
- a parameter estimation program is In the information processing device, Get the communication throughput, which is the amount of data transmitted per unit time, Based on the communication throughput acquired up to the first time point, a function specifying parameter for specifying a probability density function using the communication throughput at the second time point after the first time point as a random variable is estimated. This is a program for executing processing.
- the present invention can be configured as described above to improve the accuracy of communication throughput estimation.
- the standard deviation calculated based on the probability density function used by the communication throughput prediction apparatus according to the first embodiment of the present invention is compared with the standard deviation calculated based on the communication throughput measured in the actual communication network. It is a graph. It is a block diagram showing the function of the delivery apparatus which concerns on 2nd Embodiment of this invention. It is a block diagram showing the function of the delivery system which concerns on 3rd Embodiment of this invention.
- the communication throughput prediction apparatus (parameter estimation apparatus) 100 is an apparatus that predicts communication throughput.
- Communication throughput is the amount of data transmitted per unit time.
- the communication throughput prediction device 100 is an information processing device.
- the communication throughput prediction apparatus 100 is a personal computer, a mobile phone terminal, a PHS (Personal Handyphone System), a PDA (Personal Data Assistant, a Personal Digital Assistant), a smartphone, a car navigation terminal, or a game terminal.
- PHS Personal Handyphone System
- PDA Personal Digital Assistant
- the communication throughput prediction apparatus 100 includes a central processing unit (CPU) (not shown) and a storage device (memory and hard disk drive (HDD)).
- the communication throughput prediction apparatus 100 is configured to realize functions to be described later when a CPU executes a program stored in a storage device.
- FIG. 1 is a block diagram showing functions of the communication throughput prediction apparatus 100 configured as described above.
- the function of the communication throughput prediction apparatus 100 includes a communication throughput measurement unit (communication throughput acquisition unit) 101, a drift calculation unit 102, a variance calculation unit 103, and a communication throughput prediction unit (communication state estimation unit) 104.
- the drift calculation unit 102 and the variance calculation unit 103 constitute function specifying parameter estimation means.
- the communication throughput measuring unit 101 measures (acquires) the current (current) communication throughput in data transmission (in this example, transmission according to the streaming method) for which communication throughput is predicted.
- the communication throughput prediction apparatus 100 may be configured to acquire the communication throughput by receiving it from another apparatus.
- the transmission of data for which the communication throughput is predicted is the transmission of data representing a moving image from the server device to the client device connected to the server device via the communication network.
- the client device is configured to play a moving image based on the received data.
- the server device transmits data to the client device via the communication network according to TCP / IP (Transmission Control Protocol / Internet Protocol). Therefore, in this example, the communication throughput predicted by the communication throughput prediction apparatus 100 is a communication throughput in a TCP session established between the server apparatus and the client apparatus.
- TCP / IP Transmission Control Protocol / Internet Protocol
- the communication throughput prediction apparatus 100 may be configured to acquire a communication throughput related to data transmitted according to a communication protocol other than TCP / IP and guaranteeing the arrival of data.
- Equation 1 the current communication throughput at a certain time t [sec] is u (t) [bps (bit per second)]
- the current communication throughput u (t) is calculated as shown in Equation 1.
- ⁇ T [sec] is a preset time (for example, 2 [sec]).
- ⁇ S [bit] is the amount of data (data size) transmitted from time t ⁇ T before time t to time t by ⁇ T.
- the communication throughput measuring unit 101 measures the current communication throughput u for each preset time interval (interval) I [sec]. As a result, the communication throughput measuring unit 101 acquires time series data (a series of data measured over time) u (nI) of the communication throughput.
- n is an integer from 1 to N, and N represents the number of times the communication throughput u has been measured up to the first time point (current time in this example).
- the drift calculation unit 102 calculates drift based on the time series data u (nI) of the communication throughput u acquired by the communication throughput measurement unit 101.
- the drift represents a long-term rate of change of the communication throughput u.
- the drift constitutes a part of the function specifying parameter.
- the function specifying parameter is a probability density function (probability density shown in Equation 5 described later) using the communication throughput at a second time point (in the present example, a future time point) after the first time point as a random variable. Function) parameter.
- the drift calculation unit 102 includes the latest m pieces of time series data u (NI), u ((N) of all the time series data u (nI) acquired by the communication throughput measurement unit 101. -1) Calculate the drift based on I),..., U ((N ⁇ m + 1) I).
- the drift calculation unit 102 approximates the time series data u (NI), u ((N ⁇ 1) I),..., U ((N ⁇ m + 1) I) by using the least square method.
- the slope of the linear function is calculated as drift. That is, the drift calculation unit 102 calculates the drift ⁇ as shown in Equation 2.
- ⁇ X k is the total sum of X k when k is an integer from “N ⁇ m + 1” to “N”.
- the drift calculation unit 102 may be configured to calculate the drift by using a weighted least square method in which the communication throughput measured at a time close to the current time is weighted to be larger.
- the drift calculation unit 102 is preferably configured to calculate the drift ⁇ by using the simplest three-point formula.
- the drift calculation unit 102 smoothes the time-series data u (nI) by passing through a low-pass filter (for example, an IIR (Infinite Impulse Response) filter), and smoothed time-series data u (nI) It is preferable that the differential coefficient is calculated based on the above. According to this, the drift can be calculated with high accuracy.
- a low-pass filter for example, an IIR (Infinite Impulse Response) filter
- the variance calculation unit 103 calculates the variance ⁇ 2 of the time series data u (nI) of the communication throughput u acquired by the communication throughput measurement unit 101. Note that the variance constitutes a part of the function specifying parameter.
- the variance calculation unit 103 calculates the unbiased variance as the variance ⁇ 2 as shown in Equation 4.
- u ave is the latest m pieces of time-series data u (NI), u ((N ⁇ 1) among all the time-series data u (nI) acquired by the communication throughput measuring unit 101. I),..., U ((N ⁇ m + 1) I) are average values.
- the communication throughput prediction device 100 is configured to use the current time (that is, the time when the process of calculating the function specifying parameter is executed) as the first time.
- the communication throughput prediction apparatus 100 may use a time point before the current time point as the first time point.
- the communication throughput prediction apparatus 100 uses, as the second time point, a time point before the current time point, a current time point, or a time point after the current time point in a range that is a time point after the first time point. Also good.
- the communication throughput predicting unit 104 is based on the drift ⁇ calculated by the drift calculating unit 102, the variance ⁇ 2 calculated by the variance calculating unit 103, and the probability density function f (u, t) shown in Equation 5.
- the probability that the actual communication throughput at the future time t is within the target range is calculated.
- the target range is a preset communication throughput range.
- the communication throughput prediction unit 104 may be configured to calculate an upper limit value and a lower limit value of a target probability range at a future time t.
- the target probability range is a communication throughput range in which a value obtained by integrating the probability density function f (u, t) over the target probability range is set as a preset target probability value.
- the target probability value is a target probability value.
- the communication throughput prediction unit 104 may be configured to estimate the time required for completing data transmission based on at least one of the calculated upper limit value and lower limit value of the target probability range. Good.
- the communication throughput prediction unit 104 is configured to estimate the amount of data that can be transmitted in a period after the first time point based on at least one of the calculated upper limit value and lower limit value of the target probability range. May be.
- the communication throughput prediction unit 104 includes the drift ⁇ calculated by the drift calculation unit 102, the variance ⁇ 2 calculated by the variance calculation unit 103, and the probability density function f (u, t) expressed by Equation 5. Based on the above, it can be said that the communication throughput u at the future time t is probabilistically predicted.
- Equation 5 is derived by assuming that the communication throughput changes to perform Brownian motion with drift.
- the change of the communication throughput so as to perform a Brownian motion having a drift corresponds to the fact that the stochastic process W using the communication throughput as a random variable is a generalized Wiener process.
- B (t) is a Wiener process (that is, Brownian motion) whose variance is ⁇ 2 , the following conditions (1) to (3) are satisfied.
- B (t) is an independent increment (2)
- B (s + t) ⁇ B (s) follows N (0, ⁇ 2 t) (3)
- N ( ⁇ , ⁇ 2 ) represents a normal distribution in which the expected value is ⁇ and the variance is ⁇ 2 .
- B (t) is an independent increment when B ⁇ t ⁇ t ⁇ s ′ ⁇ t ′ and B (t) ⁇ B (s) and B (t ′) ⁇ B (s ′). And become independent random variables.
- W (t) is a generalized Wiener process (that is, Brownian motion having a drift)
- W (t) is expressed as shown in Equation 6 using B (t). expressed.
- Equation 7 the probability density function f w (x) for the random variable x after t [sec] of the generalized Wiener process W (t) is expressed as shown in Equation 7.
- Equation 5 the probability density function f (u, t) using the communication throughput u at a future time t [sec] as a random variable is expressed as shown in Equation 5. In this way, Equation 5 is derived.
- FIG. 2 is a graph showing an example of a change in communication throughput according to the probability density function f (u, t) shown in Equation 5.
- Equation 5 the variance of the probability density function f (u, t) of the communication throughput u changes in direct proportion to the time t. That is, the standard deviation of the probability density function f (u, t) changes in direct proportion to the square root t 1/2 of the time t.
- the dashed-dotted line in FIG. 2 is a curve representing a value that changes in direct proportion to the square root t 1/2 of time t.
- the fluctuation width (or variation) of the communication throughput obtained by the simulation is within the range of the alternate long and short dash line, and therefore changes in direct proportion to the square root t 1/2 of the time t. I understand.
- FIG. 3 compares the standard deviation calculated based on the probability density function f (u, t) shown in Equation 5 with the standard deviation calculated based on the communication throughput measured in the actual communication network. It is a graph. A broken line is a standard deviation calculated based on communication throughput measured in an actual communication network. The solid line is a standard deviation calculated based on the probability density function f (u, t) shown in Formula 5 (that is, changes in direct proportion to the square root t 1/2 of time t).
- the method for calculating the standard deviation based on the measured communication throughput will be described more specifically. First, every time a preset time ⁇ T elapses, the actual communication throughput is measured.
- Such processing is executed every time ⁇ T elapses and for various prediction times T. Then, a histogram (frequency distribution) of the normalized communication throughput v is acquired for each predicted time T. Then, for each predicted time T, the standard deviation is calculated based on the acquired histogram.
- the standard deviation calculated based on the probability density function f (u, t) shown in Formula 5 is sufficiently close to the standard deviation calculated based on the communication throughput measured in the actual communication network. Has a value. That is, it can be said that the probability density function specified by the communication throughput prediction apparatus 100 represents the communication throughput in the actual communication network with high accuracy.
- the communication throughput prediction apparatus 100 As described above, according to the communication throughput prediction apparatus 100 according to the first embodiment of the present invention, it is possible to improve the accuracy regarding the estimation of the communication throughput. Specifically, according to the communication throughput prediction apparatus 100, the communication throughput (for example, the upper limit value and the lower limit value of the target probability range) with high accuracy based on the probability density function specified by the estimated function specifying parameter. ) Can be estimated.
- the communication throughput for example, the upper limit value and the lower limit value of the target probability range
- the communication throughput prediction apparatus 100 replaces the probability density function f (u, t) shown in Equation 5 with a Brownian motion (that is, a Brownian motion with no drift). ) May be configured to use a probability density function derived by assuming that it changes to perform. Note that the communication throughput changing so as to perform Brownian motion without drift corresponds to the fact that the stochastic process W using the communication throughput as a random variable is a Wiener process.
- the distribution device determines the content distribution rate and the encoding rate of the content to be distributed to the communication throughput prediction device 100 according to the first embodiment based on the point of content distribution and the function specifying parameter. Is different. Accordingly, the following description will focus on such differences.
- the distribution device (parameter estimation device) 200 is a device that transmits (distributes) data representing content.
- the distribution device 200 is an information processing device.
- the distribution device 200 includes a central processing unit (CPU) (not shown) and a storage device (memory and HDD).
- the distribution device 200 is configured to realize functions to be described later when the CPU executes a program stored in the storage device.
- FIG. 4 is a block diagram showing functions of the distribution apparatus 200 configured as described above.
- the functions of the distribution device 200 include a communication throughput measuring unit (communication throughput obtaining unit) 201, a drift calculating unit 202, a variance calculating unit 203, a rate control unit (control parameter determining unit) 204, a content storage unit 205, A data transmission unit 206.
- the drift calculation unit 202 and the variance calculation unit 203 constitute function specifying parameter estimation means.
- the communication throughput measuring unit 201, the drift calculating unit 202, and the variance calculating unit 203 have the same functions as the communication throughput measuring unit 101, the drift calculating unit 102, and the variance calculating unit 103 according to the first embodiment.
- the communication throughput measuring unit 201 measures the communication throughput in data transmission by the data transmitting unit 206 described later.
- the rate control unit 204 is transmitted at a future time t based on the drift ⁇ calculated by the drift calculation unit 202, the variance ⁇ 2 calculated by the variance calculation unit 203, Equation 9, and Equation 10.
- the coding rate r * of the data to be obtained is calculated (determined).
- the coding rate constitutes a control parameter used for transmitting data.
- the rate control unit 204 sets the value of the evaluation function J shown in Equation 10 under the constraint that the remaining reproduction time T p (t) shown in Equation 9 is always larger than 0.
- the code rate r * to be minimized is calculated.
- the remaining playback time T p [sec] is a time during which content can be played back based on a portion of the data stored in the storage device of the client device that has not yet been played back.
- the client device is configured to store the received data in the storage device and reproduce the content based on the stored data while receiving the data representing the content from the distribution device 200 according to the streaming method.
- P [(Dimensionless)] is a playback rate set in accordance with an instruction from the user.
- the playback rate is the playback time of content played back per unit time.
- the playback rate is a value obtained by dividing the playback speed of the content by the playback speed corresponding to the constant speed playback.
- R [bps] is a coding rate of data transmitted by the distribution apparatus 200.
- ⁇ is a coefficient set in advance according to a target probability value which is a target probability value. Details of the coefficient ⁇ will be described later.
- Each of a 1 and a 2 is a preset coefficient. Details of the coefficient a 1 and the coefficient a 2 will be described later.
- r ave is a value obtained by averaging the coding rates of data transmitted by the distribution apparatus 200 up to the present time.
- the rate control unit 204 calculates the code rate r * under the constraint that the remaining reproduction time T p (t) shown in Equation 9 is always larger than 0.
- Equation 9 the range of values that the future communication throughput u can take is directly proportional to the square root t 1/2 of the time t as represented by the probability density function f (u, t) (as shown in FIG. 2). Then spread.
- the target probability range is a range of communication throughput, and is a range in which a value obtained by integrating the probability density function f (u, t) over the range is a target probability value.
- the target probability value is a probability that the actual communication throughput at a future time t is equal to or higher than the lower limit value u inf (t) of the target probability range.
- the target probability value is set to 99.87%.
- the target probability range is a range where ⁇ 3 ⁇ ⁇ u ⁇ + ⁇ . Therefore, the coefficient ⁇ is set to 3.
- the target probability range is a range where ⁇ ⁇ u ⁇ + ⁇ . Therefore, in this case, the coefficient ⁇ is set to 1.
- Equation 12 the differential equation relating to the remaining reproduction time T p is expressed as shown in Equation 12.
- Equation 9 is derived by integrating both sides in the range of time 0 to t. .
- Equation 10 a method for deriving Equation 10 will be described.
- the higher the encoding rate the higher the reproduction quality of the content (image quality when the content is a moving image).
- the fluctuation in the encoding rate of the data received by the client device becomes smaller, the fluctuation in the reproduction quality of the content (if the content is a moving image, flickering in image quality) is suppressed.
- the reproduction quality of the content felt by the user is increased.
- the evaluation function J it is considered that it is preferable to define the evaluation function J so that it has a value that decreases as the encoding rate r increases and decreases as the variation in the encoding rate r decreases.
- the sum of the first term inversely proportional to the square of the coding rate r and the second term proportional to the square of the difference between the average value r ave and the coding rate r is expressed as time 0
- a function integrated over t is defined as an evaluation function J.
- Proportionality coefficient a 1 of the first term, and, each of the proportional coefficient a 2 second term is a positive value.
- the first term has a value that decreases as the encoding rate r increases.
- the second term has a value that decreases as the variation in the encoding rate r decreases. In this way, Equation 10 is derived.
- the rate control unit 204 recalculates the encoding rate r * every time a preset update period elapses.
- the content storage unit 205 stores (accumulates) in advance data obtained by encoding one content (for example, information representing video or audio) at a plurality of different encoding rates.
- the data transmission unit 206 acquires data encoded at the latest encoding rate r * calculated by the rate control unit 204 among the data stored in the content storage unit 205, and the acquired data Is transmitted (distributed) to a client apparatus communicably connected to the distribution apparatus 200.
- the data transmission unit 206 exceeds the encoding rate r * when data encoded at the encoding rate r * calculated by the rate control unit 204 is not stored in the content storage unit 205. Data encoded at the maximum encoding rate in a non-existing range is acquired.
- the data transmitting unit 206 when data encoded at the encoding rate r * calculated by the rate control unit 204 is not stored in the content storage unit 205, the code closest to the encoding rate r *. It may be configured to acquire data encoded at a conversion rate.
- the encoding rate is determined with high accuracy based on the probability density function specified by the estimated function specifying parameter. can do.
- the distribution apparatus 200 determines the control parameter by assuming that the communication throughput is the lower limit value of the target probability range.
- the distribution apparatus 200 distributes data representing content to the client apparatus according to the streaming method, the probability that the reproduction quality of the content in the client apparatus is excessively lowered can be sufficiently reduced.
- the distribution apparatus 200 is configured to determine the encoding rate as the control parameter, but may be configured to determine a control parameter other than the encoding rate.
- the distribution apparatus 200 is configured to determine the control parameter by assuming that the communication throughput is the lower limit value of the target probability range.
- the control parameter may be determined by assuming the upper limit value of the range.
- the distribution system according to the third embodiment is different from the distribution apparatus 200 according to the second embodiment in that the client device determines the encoding rate. Accordingly, the following description will focus on such differences.
- the distribution system 1 includes a client device 300 that reproduces content and a server device 400 that distributes data representing the content.
- the client device 300 and the server device 400 are communicably connected to each other via a communication network (in this example, an IP (Internet Protocol) network).
- a communication network in this example, an IP (Internet Protocol) network.
- Each of the client device 300 and the server device 400 is an information processing device.
- Each of the client device 300 and the server device 400 includes a central processing unit (CPU) (not shown) and a storage device (memory and HDD).
- CPU central processing unit
- storage device memory and HDD
- Each of the client device 300 and the server device 400 is configured to realize functions to be described later when the CPU executes a program stored in the storage device.
- FIG. 5 is a block diagram showing functions of the distribution system 1 configured as described above.
- the functions of the client device 300 include a data reception unit 301, a content reproduction unit 302, a communication throughput measurement unit (communication throughput acquisition means) 303, a drift calculation unit 304, a variance calculation unit 305, and a rate control unit (control parameter). Determining means) 306.
- the drift calculation unit 304 and the variance calculation unit 305 constitute function specifying parameter estimation means.
- the data receiving unit 301 stores the received data in the storage device included in the client device 300 while receiving the data representing the content transmitted (distributed) by the server device 400.
- the content playback unit 302 plays back content based on the data stored in the storage device.
- the content reproduction unit 302 reproduces content via a display and a speaker as an output device.
- the communication throughput measuring unit 303, the drift calculating unit 304, and the variance calculating unit 305 have the same functions as the communication throughput measuring unit 101, the drift calculating unit 102, and the variance calculating unit 103 according to the first embodiment.
- the communication throughput measuring unit 303 measures the communication throughput in data reception by the data receiving unit 301.
- the rate control unit 306 has the same function as the rate control unit 204 according to the second embodiment. Furthermore, the rate control unit 306 transmits the calculated encoding rate r * to the server device 400.
- the function of the server device 400 includes a content storage unit 401 and a data transmission unit 402.
- the content storage unit 401 has the same function as the content storage unit 205 according to the second embodiment.
- the data transmission unit 402 receives the encoding rate transmitted by the client device 300.
- the data transmission unit 402 acquires data encoded at the latest received encoding rate r * among the data stored in the content storage unit 401, and transmits the acquired data to the client device 300. Send.
- the data transmission unit 402 when data encoded at the received encoding rate r * is not stored in the content storage unit 401, the data transmission unit 402 has the maximum in a range not exceeding the encoding rate r * . Data encoded at the encoding rate is acquired.
- the communication throughput prediction apparatus according to the fourth embodiment is a probability derived by assuming that the communication throughput changes to perform geometric Brownian motion with respect to the communication throughput prediction apparatus 100 according to the first embodiment.
- the difference is that a density function is used. Accordingly, the following description will focus on such differences.
- the communication throughput prediction apparatus 500 is an information processing apparatus having the same configuration as the communication throughput prediction apparatus 100 according to the first embodiment.
- FIG. 6 is a block diagram showing the functions of the communication throughput prediction apparatus 500 configured as described above.
- the function of the communication throughput prediction apparatus 500 includes a communication throughput measurement unit (communication throughput acquisition unit) 501, a function specific parameter estimation unit (function specific parameter estimation unit) 502, a communication throughput prediction unit (communication state estimation unit) 503, including.
- the communication throughput measuring unit 501 has the same function as the communication throughput measuring unit 101 according to the first embodiment.
- the function specifying parameter estimation unit 502 estimates a function specifying parameter for specifying a probability density function with a future communication throughput as a random variable based on the communication throughput u measured by the communication throughput measuring unit 501.
- the communication throughput prediction apparatus 500 has a probability density function f (u, t) derived by assuming that the communication throughput changes to perform geometric Brownian motion, as shown in Equation 13. Is used.
- the first parameter ⁇ is also called “percentage drift”.
- the first parameter ⁇ represents the average increase rate with respect to the communication throughput u at the current time.
- the second parameter ⁇ is also called “percentage volatility”.
- the first parameter ⁇ and the second parameter ⁇ constitute a function specifying parameter.
- the function specifying parameter estimation unit 502 calculates the converted communication throughput u ′ based on the mathematical formula 14 and the communication throughput u measured by the communication throughput measurement unit 501.
- the function specific parameter estimation unit 502 calculates the drift ⁇ with respect to the post-conversion communication throughput u ′ by substituting the calculated post-conversion communication throughput u ′ into the communication throughput u in Equation 3.
- the function specific parameter estimation unit 502 calculates the variance ⁇ 2 for the converted communication throughput u ′ by substituting the calculated converted communication throughput u ′ for the communication throughput u in Equation 4.
- the function specifying parameter estimation unit 502 solves the simultaneous equations based on the calculated drift ⁇ , the calculated variance ⁇ 2 , the formula 15 and the formula 16 (that is, the formula 15 and the formula 16).
- the first parameter ⁇ and the second parameter ⁇ are calculated (estimated).
- the communication throughput prediction unit 503 includes the first parameter ⁇ and the second parameter ⁇ calculated by the function specific parameter estimation unit 502, and the probability density function f (u, t) expressed by Equation 13. Based on this, the probability that the actual communication throughput at the future time t is within the target range is calculated.
- the communication throughput prediction unit 503 may be configured to calculate an upper limit value and a lower limit value of a target probability range at a future time t. Further, the communication throughput prediction unit 503 may be configured to estimate the time required for completing data transmission based on at least one of the calculated upper limit value and lower limit value of the target probability range. Good.
- the communication throughput prediction unit 503 is configured to estimate the amount of data that can be transmitted in a period after the first time point based on at least one of the calculated upper limit value and lower limit value of the target probability range. May be.
- the communication throughput prediction unit 503 determines the average value (expected value) E [communication throughput at a future time t based on the first parameter ⁇ calculated by the function specifying parameter estimation unit 502 and Equation 17. u (t)] may be calculated.
- the communication throughput prediction unit 503 also determines the communication throughput at a future time t based on the first parameter ⁇ and the second parameter ⁇ calculated by the function specifying parameter estimation unit 502 and Equation 18.
- the standard deviation V [u (t)] may be calculated.
- the communication throughput prediction unit 503 includes the first parameter ⁇ and the second parameter ⁇ calculated by the function specifying parameter estimation unit 502 and the probability density function f (u, t ) And the communication throughput u at a future time t is probabilistically predicted.
- Equation 13 is derived by assuming that the communication throughput changes to perform geometric Brownian motion. Accordingly, the stochastic differential equation with respect to the communication throughput u is expressed as shown in Equation 19.
- W t represents a Wiener process.
- Equation 20 representing communication throughput u is derived.
- Equation 13 representing a probability density function with the future communication throughput u as a random variable is derived.
- the same operation and effect as the communication throughput prediction apparatus 100 according to the first embodiment can be achieved.
- each of the distribution device 200 according to the second embodiment and the distribution system 1 according to the third embodiment also calculates the probability density derived by assuming that the communication throughput changes so as to perform geometric Brownian motion. It may be configured to use a function.
- a parameter estimation apparatus 600 A communication throughput acquisition unit (communication throughput acquisition means) 601 for acquiring a communication throughput that is an amount of data transmitted per unit time; A function specifying parameter estimating unit (function specifying parameter estimating means) 602 for estimating a function specifying parameter for specifying a probability density function using the communication throughput in the future as a random variable based on the acquired communication throughput; Is provided.
- communication throughput acquisition means communication throughput acquisition means
- a function specifying parameter estimating unit function specifying parameter estimating means 602 for estimating a function specifying parameter for specifying a probability density function using the communication throughput in the future as a random variable based on the acquired communication throughput
- the communication throughput can be estimated with high accuracy based on the probability density function specified by the estimated function specifying parameter.
- control parameters such as a coding rate can be determined with high accuracy based on the probability density function specified by the estimated function specifying parameter.
- the distribution system according to the sixth embodiment is different from the distribution apparatus 200 according to the second embodiment in that a stop time is determined based on a function specifying parameter.
- the stop time is a time for stopping transmission of data representing content. Accordingly, the following description will focus on such differences.
- this portion is data that has been sent from the server device to the client device. There is a possibility that the communication load is unnecessarily increased due to such unnecessarily transmitted data.
- the server device alternately repeats transmission processing for transmitting data representing content to the client device and stop processing for stopping transmission of the data for a predetermined stop time. Execute. At this time, the server device determines the stop time so that the remaining reproduction time approaches the target remaining reproduction time.
- the remaining reproduction time is a time during which the client device can reproduce the content based on a portion of the data received by the client device that has not been reproduced yet.
- the target remaining playback time is a target value for the remaining playback time.
- the distribution system 2 includes a client device 700 that reproduces content, and a server device (parameter estimation device) 800 that distributes data representing content.
- the client device 700 and the server device 800 are connected to be communicable with each other via a communication network (in this example, an IP (Internet Protocol) network).
- a communication network in this example, an IP (Internet Protocol) network.
- Each of the client device 700 and the server device 800 is an information processing device.
- Each of the client device 700 and the server device 800 includes a central processing unit (CPU) (not shown) and a storage device (memory and HDD).
- CPU central processing unit
- storage device memory and HDD
- Each of the client device 700 and the server device 800 is configured to realize functions to be described later when the CPU executes a program stored in the storage device.
- FIG. 8 is a block diagram showing functions of the distribution system 2 configured as described above.
- the functions of the client device 700 include a data receiving unit 701, a buffer unit 702, and a content reproduction unit 703.
- the data receiving unit 701 stores the received data in the buffer unit 702 while receiving the data representing the content transmitted (distributed) by the server device 800.
- the content reproduction unit 703 reproduces the content represented by the data based on the data stored in the buffer unit 702.
- the content reproduction unit 703 reproduces content via a display and a speaker as an output device.
- the functions of the server device 800 are a communication throughput measurement unit (communication throughput acquisition unit) 801, a drift calculation unit 802, a variance calculation unit 803, a stop time determination unit (control parameter determination unit) 804, and a content storage unit 805. And a data transmission unit 806. Note that the drift calculation unit 802 and the variance calculation unit 803 constitute function specifying parameter estimation means.
- the content storage unit 805 has the same function as the content storage unit 205 according to the second embodiment.
- the communication throughput measuring unit 801, the drift calculating unit 802, and the variance calculating unit 803 have the same functions as the communication throughput measuring unit 101, the drift calculating unit 102, and the variance calculating unit 103 according to the first embodiment. Note that the communication throughput measurement unit 801 measures the communication throughput in data transmission by the data transmission unit 806 described later.
- the stop time determination unit 804 calculates (determines) the stop time based on the drift ⁇ calculated by the drift calculation unit 802 and the variance ⁇ 2 calculated by the variance calculation unit 803.
- the stop time constitutes a control parameter used for transmitting data. The detailed function of the stop time determination unit 804 will be described later.
- the data transmission unit 806 stops transmission of data stored in the content storage unit 805 to the client device 700 according to the streaming method, and the transmission of the data for the stop time determined by the stop time determination unit 804.
- the stop process is repeatedly executed alternately.
- the data transmission unit 806 transmits data by a preset data amount in one transmission process. That is, the data transmitted by the data transmission unit 806 in one transmission process is one of a plurality of partial data generated by dividing data representing content for each data amount.
- the stop time determination unit 804 determines a stop time in the stop process to be executed next by the data transmission unit 806 every time the data transmission unit 806 finishes executing the transmission process. .
- the stop time determination unit 804 determines the stop time so that the remaining reproduction time approaches the target remaining reproduction time within a range in which the remaining reproduction time is maintained to be equal to or greater than a preset target remaining reproduction time.
- the remaining playback time is determined by the client device 700 based on a portion of the data stored in the buffer unit 702 of the client device 700 (that is, data received by the client device 700) that has not been played back yet. This is the time when the content can be played.
- the target remaining playback time is a target value (for example, 10 seconds or 30 seconds) of the remaining playback time.
- the stop time determination unit 804 determines the stop time by assuming that the future communication throughput is the lower limit value of the target probability range.
- the stop time determination unit 804 transmits the i th partial data v i. Assume that the stop time a i in the stop process executed immediately before the process is determined. Here, the time t when the data transmission unit 806 finishes executing the transmission process of transmitting the (i ⁇ 1) th partial data v i ⁇ 1 is set to 0.
- the stop time determination unit 804 determines whether Equation 21 is satisfied. It is to be noted that the establishment of Equation 21 means that the time period t is within the range from 0 (current time) to the transmission end upper limit time T sup (that is, the period from the current time to the time after the transmission end upper limit time T sup ) Represents that transmission of the i-th partial data v i can be completed.
- the transmission end upper limit time T sup is expressed by Equation 22.
- the communication throughput boundary time t c is a time when the lower limit value u inf (t) of the target probability range of the communication throughput becomes zero.
- the estimation accuracy upper limit time I is an upper limit time at which the communication throughput estimation accuracy is maintained relatively high.
- the estimated accuracy upper limit time point I is a preset value.
- the estimated accuracy upper limit time point I is preferably set to a value within the range of 10 seconds to several tens of seconds.
- the communication throughput boundary time t c is obtained by solving Equation 23, which indicates that the lower limit value u inf (t) of the target probability range is equal to 0.
- the stop time determination unit 804 calculates the communication throughput boundary time point t c based on Expression 25.
- Equation 28 A discriminant representing whether or not there is a solution in the real number range of Equation 27 is expressed as Equation 28.
- the stop time determination unit 804 sets the communication throughput boundary time t c to ⁇ , as shown in Equation 29.
- the stop time determination unit 804 calculates the communication throughput boundary time point t c based on Equation 30.
- the stop time determination unit 804 determines 0 as the stop time a i when it is determined that Formula 21 is not satisfied. In this case, even if data is transmitted over the entire period from the current time to the transmission end upper limit time T sup , the transmission of the i-th partial data v i cannot be completed. Therefore, it is preferable to start transmission of the i-th partial data v i immediately from the present time. Thereby, it is possible to avoid the remaining reproduction time in the client device 700 from becoming excessively short.
- the stop time determination unit 804 calculates each of the minimum remaining playback time T min (0) and the minimum remaining playback time T min (T sup ).
- the minimum remaining reproduction time T min (a i ) is the minimum value of the remaining reproduction time T p (t) in the period from the current time to the transmission end time b i .
- the transmission end time b i is a time when transmission of the i-th partial data v i ends (completes).
- the minimum remaining reproduction time T min (a i ) is expressed by Equation 31.
- Equation 32 the remaining reproduction time T p (t) at time t is expressed by Equation 32 and Equation 33.
- the transmission end time b i is obtained by Expression 34.
- Equation 34 is derived from Equation 35.
- the stop time a i is also referred to as a transmission start time because transmission of the i-th partial data v i starts at a time after the stop time from the current time.
- the stop time determination unit 804 determines 0 as the stop time a i .
- the minimum remaining reproduction time T min (0) is the target remaining reproduction time T r or more and, if the minimum remaining reproduction time T min (T sup) is smaller than the target remaining reproduction time T r, stop time determiner In step 804, the stop time a i is determined so that the minimum remaining reproduction time T min (a i ) matches the target remaining reproduction time Tr .
- the reproduction time T min (c i ) is calculated, and the temporary stop time c i from which the minimum remaining reproduction time T min (c i ) closest to the target residual reproduction time T r is calculated is determined as the stop time a i. Determine as.
- the stop time determination unit 804 sets the remaining reproduction time T p (t) to the target remaining reproduction time T r within a range in which the remaining reproduction time T p (t) is maintained to be equal to or greater than the target remaining reproduction time T r . It can be said that the stop time a i is determined so as to approach.
- FIG. 9 is a graph showing changes in the remaining reproduction time in the distribution system 2.
- the broken line represents the communication throughput
- the alternate long and short dash line represents the encoding rate
- the solid line represents the remaining reproduction time.
- each partial data v i is the size that can play only 10 seconds the contents based on the partial data v i.
- the communication throughput varies relatively greatly as shown in FIG. Even in such a case, according to the distribution system 2, it is possible to prevent the remaining reproduction time from becoming zero and to prevent the remaining reproduction time from becoming excessively large.
- FIG. 10 is a graph showing a change in the remaining reproduction time in a distribution system (comparison distribution system) as a comparative example.
- This comparative distribution system is different from the distribution system 2 in that the stop time is determined by assuming that the communication throughput has the same value as the current communication throughput in the future.
- the remaining playback time may become zero. That is, when the remaining reproduction time becomes excessively small, the probability that the reproduction quality of the content in the client device is excessively decreased is relatively high.
- the stop time is determined with high accuracy based on the probability density function specified by the estimated function specifying parameter. be able to.
- the server device 800 determines the stop time as the control parameter by assuming that the communication throughput is the lower limit value of the target probability range.
- the server device 800 sets the stop time so that the remaining reproduction time approaches the target remaining reproduction time in a range in which the remaining reproduction time is maintained to be equal to or more than the target remaining reproduction time. decide.
- the distribution system 2 according to this modification is different from the distribution system 2 according to the sixth embodiment in that the method for determining the stop time is simplified. Accordingly, the following description will focus on such differences.
- the stop time determination unit 804 uses the remaining reproduction time T p (b i ) at the transmission end time b i, which is the time when transmission of the i-th partial data v i is completed (completed), as the target remaining reproduction time.
- the stop time a i is determined so as to be close to Tr .
- the target coincidence time which is the time when the remaining reproduction time T p (t) coincides with the target remaining reproduction time T r .
- the target coincidence time b r is calculated by Equation 37 derived from Equation 36. expressed.
- the stop time determination unit 804 calculates the target matching point b r based on Equation 37.
- the stop time determination unit 804 determines the target coincidence time b r as the transmission end time b i .
- the stop time determination unit 804 determines the transmission end upper limit time T sup as the transmission end time b i .
- the stop time determining unit 804 determines the stop time a i based on the determined transmission end time b i and the mathematical formula 34. At this time, if the stop time a i is not a value within the range from 0 to the transmission end time b i , the stop time determination unit 804 determines the stop time a i to 0 again.
- the stop time determination unit 804 the remaining reproduction time T p (b i) as close to the target remaining reproduction time T r in the transmission end time b i, determining a stopping time a i.
- the distribution system 2 according to this modification can also provide the same operations and effects as the distribution system 2 according to the sixth embodiment.
- the parameter estimation device uses the communication throughput acquired up to the first time point as a probability variable with the communication throughput at the second time point after the first time point as a random variable. It was configured to estimate a function specific parameter for specifying the density function.
- the parameter estimation device calculates a probability density function based on the communication throughput acquired in a certain period, with the communication throughput at a point before (past) the period as a random variable.
- the function specifying parameter for specifying may be configured to be estimated.
- each function of the parameter estimation device is realized by the CPU executing a program (software), but may be realized by hardware such as a circuit.
- the program is stored in the storage device, but may be stored in a computer-readable recording medium.
- the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.
- a parameter estimation device comprising:
- the communication throughput can be estimated with high accuracy based on the probability density function specified by the estimated function specifying parameter.
- control parameters such as a coding rate can be determined with high accuracy based on the probability density function specified by the estimated function specifying parameter.
- a parameter estimation device comprising control parameter determining means for determining a control parameter used for transmitting data based on the specified function specifying parameter.
- control parameters such as a coding rate with high accuracy.
- the parameter estimation device (Appendix 3) The parameter estimation device according to attachment 2, wherein The control parameter determining means sets the value obtained by integrating the probability density function specified by the estimated function specifying parameter over the range in which the communication throughput is in the communication throughput range.
- a parameter estimation device configured to determine the control parameter by assuming an upper limit value or a lower limit value of a target probability range, which is a target probability value range that is a probability value of
- the parameter estimation apparatus according to Supplementary Note 2 or Supplementary Note 3, wherein
- the control parameter is an encoding rate of the transmitted data;
- the control parameter determining means sets the value obtained by integrating the probability density function specified by the estimated function specifying parameter over the range in which the communication throughput is in the communication throughput range.
- a parameter estimation device configured to determine the encoding rate by assuming that the target probability value is a lower limit value of a target probability range, which is a range to be a target probability value that is a value of a probability of
- the server device distributes data representing content to the client device according to the streaming method, the probability that the reproduction quality of the content on the client device is excessively lowered can be sufficiently reduced.
- the parameter estimation apparatus according to Supplementary Note 2 or Supplementary Note 3, wherein
- the control parameter is the stop time;
- the control parameter determination means sets a remaining target playback time that is a time during which the content can be played back based on a portion of the data received by the client device that has not been played back, to a preset target residual playback time.
- the communication throughput is within the range of the communication throughput, and the value obtained by integrating the probability density function specified by the estimated function specifying parameter over the range is the target probability.
- a parameter estimation device configured to determine the stop time by assuming that the target probability value is a lower limit value of a target probability range, which is a range that is a target probability value that is a value of.
- the parameter estimation device according to appendix 5 or appendix 6,
- the control parameter determining means is configured to determine the stop time so that the remaining reproduction time approaches the target remaining reproduction time within a range in which the remaining reproduction time is maintained to be equal to or more than the target remaining reproduction time. Parameter estimation device.
- Appendix 8 The parameter estimation device according to any one of appendix 1 to appendix 7, Based on the estimated function specifying parameter, at least one of a communication throughput, a time required for completing data transmission, and an amount of data that can be transmitted in a period after the first time point is estimated.
- a parameter estimation device comprising communication state estimation means.
- the communication throughput on the basis of the probability density function specified by the estimated function specifying parameter, the communication throughput, the time required for data transmission to be completed with high accuracy, and / or after the first time point It is possible to estimate the amount of data that can be transmitted in the period.
- the change in communication throughput is well represented by the Brownian motion. Therefore, according to the parameter estimation device configured as described above, the communication throughput can be estimated with high accuracy, and the control parameter can be determined with high accuracy.
- the change in communication throughput is well represented by Brownian motion with drift. Therefore, according to the parameter estimation device configured as described above, the communication throughput can be estimated with high accuracy, and the control parameter can be determined with high accuracy.
- the change in communication throughput is well represented by geometric Brownian motion. Therefore, according to the parameter estimation device configured as described above, the communication throughput can be estimated with high accuracy, and the control parameter can be determined with high accuracy.
- the parameter estimation method according to attachment 13 comprising: A target in which the communication throughput is a range of the communication throughput, and a value obtained by integrating the probability density function specified by the estimated function specifying parameter over the range is a target probability value.
- a parameter estimation method configured to determine the control parameter by assuming an upper limit value or a lower limit value of a target probability range, which is a range to be a probability value.
- a parameter estimation program according to attachment 17, wherein The processing is as follows: A target in which the communication throughput is a range of the communication throughput, and a value obtained by integrating the probability density function specified by the estimated function specifying parameter over the range is a target probability value.
- a parameter estimation program configured to determine the control parameter by assuming an upper limit value or a lower limit value of a target probability range, which is a range to be a probability value.
- the parameter estimation program according to any one of supplementary notes 16 to 18, comprising: In addition to the information processing apparatus, Based on the estimated function specifying parameter, at least one of a communication throughput, a time required for completing data transmission, and an amount of data that can be transmitted in a period after the first time point is estimated.
- a parameter estimation program for executing processing In addition to the information processing apparatus, Based on the estimated function specifying parameter, at least one of a communication throughput, a time required for completing data transmission, and an amount of data that can be transmitted in a period after the first time point is estimated.
- the present invention relates to a patent application of Japanese Patent Application No. 2011-151349 filed on July 8, 2011 in Japan and a Japanese Patent Application No. 2011-192642 filed on September 5, 2011 in Japan.
- the benefit of the priority claim based on this patent application is enjoyed, and all the contents disclosed in the patent application are included in this specification.
- the present invention can be applied to a communication throughput prediction apparatus that estimates communication throughput, a control parameter determination apparatus that determines a control parameter used to transmit data, and the like.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Data Mining & Analysis (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Information Transfer Between Computers (AREA)
Abstract
Description
単位時間あたりに伝送されるデータの量である通信スループットを取得する通信スループット取得手段と、
第1の時点までに上記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する関数特定パラメータ推定手段と、
を備える。
単位時間あたりに伝送されるデータの量である通信スループットを取得し、
第1の時点までに上記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する方法である。
情報処理装置に、
単位時間あたりに伝送されるデータの量である通信スループットを取得し、
第1の時点までに上記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する、処理を実行させるためのプログラムである。
図1に示したように、第1実施形態に係る通信スループット予測装置(パラメータ推定装置)100は、通信スループットを予測する装置である。通信スループットは、単位時間あたりに伝送されるデータの量である。
図1は、上記のように構成された通信スループット予測装置100の機能を表すブロック図である。通信スループット予測装置100の機能は、通信スループット測定部(通信スループット取得手段)101と、ドリフト算出部102と、分散算出部103と、通信スループット予測部(通信状態推定手段)104と、を含む。なお、ドリフト算出部102、及び、分散算出部103は、関数特定パラメータ推定手段を構成している。
(1)B(t)は、独立増分である
(2)任意のs、t>0に対して、B(s+t)-B(s)は、N(0,σ2t)に従う
(3)B(0)=0、且つ、B(t)は、t=0で連続である
次に、本発明の第2実施形態に係る配信装置について説明する。第2実施形態に係る配信装置は、上記第1実施形態に係る通信スループット予測装置100に対して、コンテンツを配信する点、及び、関数特定パラメータに基づいて、配信するコンテンツの符号化レートを決定する点において相違している。従って、以下、かかる相違点を中心として説明する。
図4は、上記のように構成された配信装置200の機能を表すブロック図である。配信装置200の機能は、通信スループット測定部(通信スループット取得手段)201と、ドリフト算出部202と、分散算出部203と、レート制御部(制御パラメータ決定手段)204と、コンテンツ蓄積部205と、データ送信部206と、を含む。なお、ドリフト算出部202、及び、分散算出部203は、関数特定パラメータ推定手段を構成している。
ところで、将来の通信スループットuが取り得る値の範囲は、確率密度関数f(u,t)により表されるように(図2に示したように)、時刻tの平方根t1/2に正比例して拡散する。
ところで、符号化レートが高いデータほど、コンテンツの再生品質(コンテンツが動画である場合には、画質)が高くなる。また、クライアント装置が受信するデータの符号化レートの変動が小さくなるほど、コンテンツの再生品質の変動(コンテンツが動画である場合には、画質のちらつき)が抑制される。その結果、ユーザが感じるコンテンツの再生品質が高くなる。
このようにして、数式10は導出される。
次に、本発明の第3実施形態に係る配信システムについて説明する。第3実施形態に係る配信システムは、上記第2実施形態に係る配信装置200に対して、クライアント装置が符号化レートを決定する点において相違している。従って、以下、かかる相違点を中心として説明する。
図5は、上記のように構成された配信システム1の機能を表すブロック図である。
クライアント装置300の機能は、データ受信部301と、コンテンツ再生部302と、通信スループット測定部(通信スループット取得手段)303と、ドリフト算出部304と、分散算出部305と、レート制御部(制御パラメータ決定手段)306と、を含む。なお、ドリフト算出部304、及び、分散算出部305は、関数特定パラメータ推定手段を構成している。
次に、本発明の第4実施形態に係る通信スループット予測装置について説明する。第4実施形態に係る通信スループット予測装置は、上記第1実施形態に係る通信スループット予測装置100に対して、通信スループットが幾何ブラウン運動を行うように変化することを仮定することにより導出された確率密度関数を用いる点において相違している。従って、以下、かかる相違点を中心として説明する。
図6は、上記のように構成された通信スループット予測装置500の機能を表すブロック図である。通信スループット予測装置500の機能は、通信スループット測定部(通信スループット取得手段)501と、関数特定パラメータ推定部(関数特定パラメータ推定手段)502と、通信スループット予測部(通信状態推定手段)503と、を含む。
次に、本発明の第5実施形態に係るパラメータ推定装置について図7を参照しながら説明する。
第5実施形態に係るパラメータ推定装置600は、
単位時間あたりに伝送されるデータの量である通信スループットを取得する通信スループット取得部(通信スループット取得手段)601と、
上記取得された通信スループットに基づいて、将来における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する関数特定パラメータ推定部(関数特定パラメータ推定手段)602と、
を備える。
次に、本発明の第6実施形態に係る配信システムについて説明する。第6実施形態に係る配信システムは、上記第2実施形態に係る配信装置200に対して、関数特定パラメータに基づいて停止時間を決定する点において相違している。ここで、停止時間は、コンテンツを表すデータの送信を停止する時間である。従って、以下、かかる相違点を中心として説明する。
これにより、無駄に多くのデータが、サーバ装置からクライアント装置へ送信されることを回避することができる。
図8は、上記のように構成された配信システム2の機能を表すブロック図である。
クライアント装置700の機能は、データ受信部701と、バッファ部702と、コンテンツ再生部703と、を含む。
停止時間決定部804は、データ送信部806が送信処理の実行を終了する毎に、送信処理の実行を終了する時点にて、次にデータ送信部806が実行する停止処理における停止時間を決定する。
停止時間決定部804は、残余再生時間を、予め設定された目標残余再生時間以上に維持する範囲において、当該残余再生時間を当該目標残余再生時間に近づけるように、停止時間を決定する。
ここでは、コンテンツを表すデータが、n個の部分データ(セグメント)vi(i=1,2,…,n)に分割される場合を想定する。各部分データviのデータサイズ(データ量)をsiと置く。また、各部分データviに基づいてコンテンツを再生可能な時間である再生時間をτiと置く。
図9は、配信システム2における残余再生時間の変化を示したグラフである。図9において、破線は、通信スループットを表し、一点鎖線は、符号化レートを表し、実線は、残余再生時間を表す。
そして、目標一致時点brが、0から送信終了上限時点Tsupまでの範囲内の値である場合、停止時間決定部804は、送信終了時点biとして目標一致時点brを決定する。
上記実施形態の一部又は全部は、以下の付記のように記載され得るが、以下には限られない。
単位時間あたりに伝送されるデータの量である通信スループットを取得する通信スループット取得手段と、
第1の時点までに前記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する関数特定パラメータ推定手段と、
を備えるパラメータ推定装置。
付記1に記載のパラメータ推定装置であって、
前記特定された関数特定パラメータに基づいて、データを送信するために用いられる制御パラメータを決定する制御パラメータ決定手段を備えるパラメータ推定装置。
付記2に記載のパラメータ推定装置であって、
前記制御パラメータ決定手段は、前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の上限値又は下限値であることを仮定することにより前記制御パラメータを決定するように構成されたパラメータ推定装置。
付記2又は付記3に記載のパラメータ推定装置であって、
前記制御パラメータは、前記送信されるデータの符号化レートであり、
前記制御パラメータ決定手段は、前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の下限値であることを仮定することにより前記符号化レートを決定するように構成されたパラメータ推定装置。
付記2又は付記3に記載のパラメータ推定装置であって、
ストリーミング方式に従って、コンテンツを表す前記データをクライアント装置へ送信する送信処理と、所定の停止時間だけ当該データの送信を停止する停止処理と、を交互に繰り返し実行するように構成され、
前記制御パラメータは、前記停止時間であり、
前記制御パラメータ決定手段は、前記クライアント装置により受信された前記データのうちの未だ再生されていない部分に基づいて前記コンテンツを再生可能な時間である残余再生時間を、予め設定された目標残余再生時間に近づけるように、前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の下限値であることを仮定することにより前記停止時間を決定するように構成されたパラメータ推定装置。
付記5に記載のパラメータ推定装置であって、
前記制御パラメータ決定手段は、前記送信処理の実行を終了する時点にて、次に実行する前記停止処理における前記停止時間を決定するように構成されたパラメータ推定装置。
付記5又は付記6に記載のパラメータ推定装置であって、
前記制御パラメータ決定手段は、前記残余再生時間を、前記目標残余再生時間以上に維持する範囲において、当該残余再生時間を当該目標残余再生時間に近づけるように、前記停止時間を決定するように構成されたパラメータ推定装置。
付記1乃至付記7のいずれかに記載のパラメータ推定装置であって、
前記推定された関数特定パラメータに基づいて、通信スループット、データの伝送が完了するまでに要する時間、及び、前記第1の時点以降の期間において伝送可能なデータの量、の少なくとも1つを推定する通信状態推定手段を備えるパラメータ推定装置。
付記1乃至付記8のいずれかに記載のパラメータ推定装置であって、
前記確率密度関数は、前記通信スループットがブラウン運動を行うように変化することを仮定することにより導出された関数であるパラメータ推定装置。
付記1乃至付記8のいずれかに記載のパラメータ推定装置であって、
前記確率密度関数は、前記通信スループットが、ドリフトを有するブラウン運動を行うように変化することを仮定することにより導出された関数であるパラメータ推定装置。
付記1乃至付記8のいずれかに記載のパラメータ推定装置であって、
前記確率密度関数は、前記通信スループットが幾何ブラウン運動を行うように変化することを仮定することにより導出された関数であるパラメータ推定装置。
単位時間あたりに伝送されるデータの量である通信スループットを取得し、
第1の時点までに前記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する、パラメータ推定方法。
付記12に記載のパラメータ推定方法であって、
前記特定された関数特定パラメータに基づいて、データを送信するために用いられる制御パラメータを決定する、パラメータ推定方法。
付記13に記載のパラメータ推定方法であって、
前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の上限値又は下限値であることを仮定することにより前記制御パラメータを決定するように構成されたパラメータ推定方法。
付記12乃至付記14のいずれかに記載のパラメータ推定方法であって、
前記推定された関数特定パラメータに基づいて、通信スループット、データの伝送が完了するまでに要する時間、及び、前記第1の時点以降の期間において伝送可能なデータの量、の少なくとも1つを推定する、パラメータ推定方法。
情報処理装置に、
単位時間あたりに伝送されるデータの量である通信スループットを取得し、
第1の時点までに前記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する、処理を実行させるためのパラメータ推定プログラム。
付記16に記載のパラメータ推定プログラムであって、
前記情報処理装置に、更に、
前記特定された関数特定パラメータに基づいて、データを送信するために用いられる制御パラメータを決定する、処理を実行させるためのパラメータ推定プログラム。
付記17に記載のパラメータ推定プログラムであって、
前記処理は、
前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の上限値又は下限値であることを仮定することにより前記制御パラメータを決定するように構成されたパラメータ推定プログラム。
付記16乃至付記18のいずれかに記載のパラメータ推定プログラムであって、
前記情報処理装置に、更に、
前記推定された関数特定パラメータに基づいて、通信スループット、データの伝送が完了するまでに要する時間、及び、前記第1の時点以降の期間において伝送可能なデータの量、の少なくとも1つを推定する、処理を実行させるためのパラメータ推定プログラム。
100 通信スループット予測装置
101 通信スループット測定部
102 ドリフト算出部
103 分散算出部
104 通信スループット予測部
200 配信装置
201 通信スループット測定部
202 ドリフト算出部
203 分散算出部
204 レート制御部
205 コンテンツ蓄積部
206 データ送信部
300 クライアント装置
301 データ受信部
302 コンテンツ再生部
303 通信スループット測定部
304 ドリフト算出部
305 分散算出部
306 レート制御部
400 サーバ装置
401 コンテンツ蓄積部
402 データ送信部
500 通信スループット予測装置
501 通信スループット測定部
502 関数特定パラメータ推定部
503 通信スループット予測部
600 パラメータ推定装置
601 通信スループット取得部
602 関数特定パラメータ推定部
2 配信システム
700 クライアント装置
701 データ受信部
702 バッファ部
703 コンテンツ再生部
800 サーバ装置
801 通信スループット測定部
802 ドリフト算出部
803 分散算出部
804 停止時間決定部
805 コンテンツ蓄積部
806 データ送信部
Claims (19)
- 単位時間あたりに伝送されるデータの量である通信スループットを取得する通信スループット取得手段と、
第1の時点までに前記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する関数特定パラメータ推定手段と、
を備えるパラメータ推定装置。 - 請求項1に記載のパラメータ推定装置であって、
前記特定された関数特定パラメータに基づいて、データを送信するために用いられる制御パラメータを決定する制御パラメータ決定手段を備えるパラメータ推定装置。 - 請求項2に記載のパラメータ推定装置であって、
前記制御パラメータ決定手段は、前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の上限値又は下限値であることを仮定することにより前記制御パラメータを決定するように構成されたパラメータ推定装置。 - 請求項2又は請求項3に記載のパラメータ推定装置であって、
前記制御パラメータは、前記送信されるデータの符号化レートであり、
前記制御パラメータ決定手段は、前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の下限値であることを仮定することにより前記符号化レートを決定するように構成されたパラメータ推定装置。 - 請求項2又は請求項3に記載のパラメータ推定装置であって、
ストリーミング方式に従って、コンテンツを表す前記データをクライアント装置へ送信する送信処理と、所定の停止時間だけ当該データの送信を停止する停止処理と、を交互に繰り返し実行するように構成され、
前記制御パラメータは、前記停止時間であり、
前記制御パラメータ決定手段は、前記クライアント装置により受信された前記データのうちの未だ再生されていない部分に基づいて前記コンテンツを再生可能な時間である残余再生時間を、予め設定された目標残余再生時間に近づけるように、前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の下限値であることを仮定することにより前記停止時間を決定するように構成されたパラメータ推定装置。 - 請求項5に記載のパラメータ推定装置であって、
前記制御パラメータ決定手段は、前記送信処理の実行を終了する時点にて、次に実行する前記停止処理における前記停止時間を決定するように構成されたパラメータ推定装置。 - 請求項5又は請求項6に記載のパラメータ推定装置であって、
前記制御パラメータ決定手段は、前記残余再生時間を、前記目標残余再生時間以上に維持する範囲において、当該残余再生時間を当該目標残余再生時間に近づけるように、前記停止時間を決定するように構成されたパラメータ推定装置。 - 請求項1乃至請求項7のいずれかに記載のパラメータ推定装置であって、
前記推定された関数特定パラメータに基づいて、通信スループット、データの伝送が完了するまでに要する時間、及び、前記第1の時点以降の期間において伝送可能なデータの量、の少なくとも1つを推定する通信状態推定手段を備えるパラメータ推定装置。 - 請求項1乃至請求項8のいずれかに記載のパラメータ推定装置であって、
前記確率密度関数は、前記通信スループットがブラウン運動を行うように変化することを仮定することにより導出された関数であるパラメータ推定装置。 - 請求項1乃至請求項8のいずれかに記載のパラメータ推定装置であって、
前記確率密度関数は、前記通信スループットが、ドリフトを有するブラウン運動を行うように変化することを仮定することにより導出された関数であるパラメータ推定装置。 - 請求項1乃至請求項8のいずれかに記載のパラメータ推定装置であって、
前記確率密度関数は、前記通信スループットが幾何ブラウン運動を行うように変化することを仮定することにより導出された関数であるパラメータ推定装置。 - 単位時間あたりに伝送されるデータの量である通信スループットを取得し、
第1の時点までに前記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する、パラメータ推定方法。 - 請求項12に記載のパラメータ推定方法であって、
前記特定された関数特定パラメータに基づいて、データを送信するために用いられる制御パラメータを決定する、パラメータ推定方法。 - 請求項13に記載のパラメータ推定方法であって、
前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の上限値又は下限値であることを仮定することにより前記制御パラメータを決定するように構成されたパラメータ推定方法。 - 請求項12乃至請求項14のいずれかに記載のパラメータ推定方法であって、
前記推定された関数特定パラメータに基づいて、通信スループット、データの伝送が完了するまでに要する時間、及び、前記第1の時点以降の期間において伝送可能なデータの量、の少なくとも1つを推定する、パラメータ推定方法。 - 情報処理装置に、
単位時間あたりに伝送されるデータの量である通信スループットを取得し、
第1の時点までに前記取得された通信スループットに基づいて、当該第1の時点よりも後の第2の時点における通信スループットを確率変数とした確率密度関数を特定するための関数特定パラメータを推定する、処理を実行させるためのパラメータ推定プログラム。 - 請求項16に記載のパラメータ推定プログラムであって、
前記情報処理装置に、更に、
前記特定された関数特定パラメータに基づいて、データを送信するために用いられる制御パラメータを決定する、処理を実行させるためのパラメータ推定プログラム。 - 請求項17に記載のパラメータ推定プログラムであって、
前記処理は、
前記通信スループットが、当該通信スループットの範囲であり、且つ、当該範囲に亘って、前記推定された関数特定パラメータにより特定される確率密度関数を積分した値を、目標となる確率の値である目標確率値とする範囲である、目標確率範囲の上限値又は下限値であることを仮定することにより前記制御パラメータを決定するように構成されたパラメータ推定プログラム。 - 請求項16乃至請求項18のいずれかに記載のパラメータ推定プログラムであって、
前記情報処理装置に、更に、
前記推定された関数特定パラメータに基づいて、通信スループット、データの伝送が完了するまでに要する時間、及び、前記第1の時点以降の期間において伝送可能なデータの量、の少なくとも1つを推定する、処理を実行させるためのパラメータ推定プログラム。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2013523780A JP5804060B2 (ja) | 2011-07-08 | 2012-06-12 | パラメータ推定装置、パラメータ推定方法、及び、パラメータ推定プログラム |
US14/129,034 US9019835B2 (en) | 2011-07-08 | 2012-06-12 | Parameter estimation device, parameter estimation method, and parameter estimation program |
EP12811739.7A EP2731302B1 (en) | 2011-07-08 | 2012-06-12 | Parameter estimating device, parameter estimating method, and parameter estimating program |
KR1020147000446A KR101511921B1 (ko) | 2011-07-08 | 2012-06-12 | 파라미터 추정 장치, 파라미터 추정 방법, 및 파라미터 추정 프로그램 |
HK14106596.9A HK1194874A1 (zh) | 2011-07-08 | 2014-06-30 | 參數估計裝置、參數估計方法及參數估計程式 |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011151349 | 2011-07-08 | ||
JP2011-151349 | 2011-07-08 | ||
JP2011192642 | 2011-09-05 | ||
JP2011-192642 | 2011-09-05 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013008387A1 true WO2013008387A1 (ja) | 2013-01-17 |
Family
ID=47505698
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/003804 WO2013008387A1 (ja) | 2011-07-08 | 2012-06-12 | パラメータ推定装置、パラメータ推定方法、及び、パラメータ推定プログラム |
Country Status (6)
Country | Link |
---|---|
US (1) | US9019835B2 (ja) |
EP (1) | EP2731302B1 (ja) |
JP (1) | JP5804060B2 (ja) |
KR (1) | KR101511921B1 (ja) |
HK (1) | HK1194874A1 (ja) |
WO (1) | WO2013008387A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020158094A1 (ja) * | 2019-02-01 | 2020-08-06 | 株式会社Nttドコモ | 情報処理装置、サーバ装置及び情報処理システム |
US10749781B2 (en) | 2014-11-18 | 2020-08-18 | Nec Corporation | Setting device, setting method, recording medium to which setting program is recorded, communication system, client device, and server device |
JP7462547B2 (ja) | 2020-12-18 | 2024-04-05 | 本田技研工業株式会社 | 情報処理装置、移動体、プログラム及び情報処理方法 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6163437B2 (ja) * | 2014-02-14 | 2017-07-12 | 株式会社日立製作所 | 電力系統解析支援システム、電力系統の解析支援装置、電力系統の解析支援方法、および電力系統用の計測装置 |
WO2016099356A1 (en) * | 2014-12-18 | 2016-06-23 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and content management module for managing content in a content distribution network |
US11122467B2 (en) * | 2018-09-07 | 2021-09-14 | Vmware, Inc. | Service aware load imbalance detection and root cause identification |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003037649A (ja) | 2001-07-24 | 2003-02-07 | Nippon Telegr & Teleph Corp <Ntt> | コンテンツ配信完了時刻の推定方法及び記録媒体並びにプログラム |
WO2012066748A1 (ja) * | 2010-11-17 | 2012-05-24 | 日本電気株式会社 | コンテンツ提供装置 |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7391740B2 (en) | 2003-04-17 | 2008-06-24 | University Of Maryland | Method for quantifying reponsiveness of flow aggregates to packet drops in a communication network |
-
2012
- 2012-06-12 EP EP12811739.7A patent/EP2731302B1/en active Active
- 2012-06-12 US US14/129,034 patent/US9019835B2/en active Active
- 2012-06-12 JP JP2013523780A patent/JP5804060B2/ja active Active
- 2012-06-12 WO PCT/JP2012/003804 patent/WO2013008387A1/ja active Application Filing
- 2012-06-12 KR KR1020147000446A patent/KR101511921B1/ko active IP Right Grant
-
2014
- 2014-06-30 HK HK14106596.9A patent/HK1194874A1/zh unknown
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003037649A (ja) | 2001-07-24 | 2003-02-07 | Nippon Telegr & Teleph Corp <Ntt> | コンテンツ配信完了時刻の推定方法及び記録媒体並びにプログラム |
WO2012066748A1 (ja) * | 2010-11-17 | 2012-05-24 | 日本電気株式会社 | コンテンツ提供装置 |
Non-Patent Citations (3)
Title |
---|
HIROSHI YOSHIDA ET AL.: "TCP Throughput no Kakuritsuteki Kakusan Yosoku ni Motozuku Eizo Haishin Seigyo", INTERNET CONFERENCE 2011, 28 October 2011 (2011-10-28), pages 57 - 66, XP008172674 * |
See also references of EP2731302A4 * |
TAKAYUKI GOTO ET AL.: "Lightweight TCP Throughput Estimation Method in CDMA2000 lxEV-DO Network", THE TRANSACTIONS OF THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS (J93-B), vol. J93-B, no. 7, 1 July 2010 (2010-07-01), pages 977 - 986, XP008172690 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10749781B2 (en) | 2014-11-18 | 2020-08-18 | Nec Corporation | Setting device, setting method, recording medium to which setting program is recorded, communication system, client device, and server device |
WO2020158094A1 (ja) * | 2019-02-01 | 2020-08-06 | 株式会社Nttドコモ | 情報処理装置、サーバ装置及び情報処理システム |
JPWO2020158094A1 (ja) * | 2019-02-01 | 2021-11-18 | 株式会社Nttドコモ | 情報処理装置、サーバ装置及び情報処理システム |
JP7253574B2 (ja) | 2019-02-01 | 2023-04-06 | 株式会社Nttドコモ | 情報処理装置、サーバ装置及び情報処理システム |
JP7462547B2 (ja) | 2020-12-18 | 2024-04-05 | 本田技研工業株式会社 | 情報処理装置、移動体、プログラム及び情報処理方法 |
Also Published As
Publication number | Publication date |
---|---|
US20140133316A1 (en) | 2014-05-15 |
HK1194874A1 (zh) | 2014-10-24 |
KR101511921B1 (ko) | 2015-04-13 |
US9019835B2 (en) | 2015-04-28 |
JP5804060B2 (ja) | 2015-11-04 |
EP2731302A4 (en) | 2015-01-14 |
JPWO2013008387A1 (ja) | 2015-02-23 |
EP2731302B1 (en) | 2016-08-10 |
KR20140026606A (ko) | 2014-03-05 |
EP2731302A1 (en) | 2014-05-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5804060B2 (ja) | パラメータ推定装置、パラメータ推定方法、及び、パラメータ推定プログラム | |
JP5673538B2 (ja) | 配信システム | |
EP2775673B1 (en) | Content reproduction information estimating device, method and program | |
WO2014133745A2 (en) | Multi-stream optimization | |
JP2004253883A (ja) | 音声・画像リアルタイム通信におけるビットレート制御を実行するデータ通信装置 | |
US20170054648A1 (en) | Data transfer apparatus, data transfer controlling method and data stream | |
JP5938015B2 (ja) | チャンクダウンロード完了判定装置、チャンクダウンロード完了判定方法、及びプログラム | |
KR102037413B1 (ko) | 스트리밍 서비스를 제공하기 위한 동적 버퍼링 방법 및 그 장치 | |
JP2008005108A (ja) | 映像品質推定装置、映像品質管理装置、映像品質推定方法、映像品質管理方法、およびプログラム | |
CN104022845A (zh) | 调整数据块比特率的方法、装置和系统 | |
EP2928145A1 (en) | Method for estimating a bandwidth associated with a connection between a client terminal and at least one server, corresponding client terminal | |
EP2393294B1 (en) | Delivery system, delivery method, server device, program and client device | |
JP2011061533A (ja) | コンテンツ配信システム、体感品質推定装置、方法、及び、プログラム | |
US9258240B2 (en) | Available bandwidth estimating device | |
JP5532045B2 (ja) | 通信システム | |
JP6227443B2 (ja) | 送信レート算出装置、通信装置、送信レート算出方法およびコンピュータプログラム | |
JP5987798B2 (ja) | 情報処理システム、情報処理装置及びプログラム | |
WO2020170869A1 (ja) | エンゲージメント推定装置、エンゲージメント推定方法及びプログラム | |
Iwamoto et al. | Rate adaptation with Bayesian attractor model for MPEG-DASH | |
JP2013046113A (ja) | 基本GoP長を用いた映像品質推定装置及び方法及びプログラム | |
EP2786567A1 (en) | Video quality measurement considering multiple artifacts | |
JP6954289B2 (ja) | ビットレート指示装置、ビットレート指示方法、及び、ビットレート指示プログラム | |
JP6724517B2 (ja) | ビットレート指示装置、ビットレート指示方法、及び、ビットレート指示プログラム | |
JP2019121847A (ja) | 品質推定装置、品質推定方法及びプログラム | |
WO2017145781A1 (ja) | ペーシング制御装置、ペーシング制御方法、及びプログラム |
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: 12811739 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2013523780 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14129034 Country of ref document: US |
|
REEP | Request for entry into the european phase |
Ref document number: 2012811739 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012811739 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 20147000446 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |