KR20030034269A - Dynamic rt-VBR traffic characterization scheme - Google Patents

Abstract

PURPOSE: A method for characterizing data traffic having a real-time dynamic variable bit rate is provided to transmit real-time data having a variable bit rate with a small number of network resources and minimum period of time required for initial characterization. CONSTITUTION: Video data having a variable bit rate, inputted in real time, is transmitted with a traffic peak rate(301). Traffic pattern variation and burst data are detected using subsum composed of a predetermined number of video data, and a traffic characterization information updating point message including the detected traffic pattern variation and burst data are transmitted(302). A transmission rate and buffer size are searched based on the traffic characterization information updating point message to generate a new traffic parameter update value(306). Connection resetting is attempted according to transmission rate of the update value to transmit video data with the updated traffic parameter.

Description

Characterization method of data traffic with real-time dynamic variable bit rate {Dynamic rt-VBR traffic characterization scheme}

The present invention relates to a method of characterizing data traffic having a real-time dynamic variable bit rate, and more particularly, to start transmission at the maximum bandwidth that can service the maximum burst for packet loss without calculating the initial equivalent bandwidth, It is a traffic parameter that can efficiently share network resources, and can be read by a computer that records a method for characterizing data traffic having a real-time dynamic variable bit rate for resetting a connection by appropriately changing the characteristics of source traffic. To a recording medium.

In general, the variable bit rate traffic characterization used in an ATM network or an IP network is fixedly described as a maximum service rate (peak rate) for a transmission service, or additionally described by processing a randomly assigned average cell rate (SCR).

Such a scheme has been disclosed in Korean Patent Application No. 1998-051629 (connection admission control apparatus and method thereof supporting multiple quality of service) (hereinafter, referred to as a prior invention).

The main subject of this prior invention is traffic parameters such as maximum cell rate (PCR), average cell rate (SCR), maximum burst size (MBS), minimum cell rate (MCR), etc. Connection acceptance is controlled by considering delay characteristics and quality of service.

However, the above-described invention is inelastic as to whether network resources are available to nodes on a transmission path by allocating an equivalent bandwidth according to an initial predetermined equation, and requires only an equivalent bandwidth of a size determined from the equation even if there is a margin of delay. Therefore, there was a problem that is not adaptive to the situation.

Accordingly, the present invention has been proposed to solve the above-mentioned conventional problems, and it is possible to use data having a real time variable bit rate that is sensitive to delay more quickly and without loss of a maximum transmission rate or transmission packet so as to efficiently use network resources. Allows more connections to share finite network resources, minimizes the time required for less network resources and initial characterization, and provides data with real-time dynamic variable bit rates for the transmission of real-time data with variable bit rates that are delay sensitive. It is an object of the present invention to provide a method for characterizing traffic and a computer-readable recording medium having recorded thereon a program for realizing the method.

1 is a configuration diagram of an embodiment of an apparatus for characterizing data traffic having a real-time dynamic variable bit rate applied to the present invention.

Figure 2 is a change in traffic characterization point for explaining the real-time dynamic traffic characterization process according to the present invention.

3 is a flow diagram of an embodiment of a method of characterizing data traffic having a real-time dynamic variable bit rate in accordance with the present invention.

FIG. 4 is a detailed flowchart illustrating an embodiment of a UP (detection point) detection process illustrated in FIG. 3.

FIG. 5 is a detailed flowchart of an embodiment of an update value generation process of the traffic parameter shown in FIG. 3; FIG.

<Description of the symbols for the main parts of the drawings>

101: MPEG decoder 102: dynamic variable bit rate transmission module

103: traffic pattern monitoring module 104: transmission rate / buffer size update module

105: dynamic network resource reservation module 106: network interface card

In order to achieve the above object, the present invention provides a method of characterizing data traffic applied to an apparatus for characterizing data traffic having a real-time dynamic variable bit rate, and connecting image data having a variable bit rate input in real time at a maximum traffic rate. A first step of requesting and transmitting; A second step of detecting burst data through traffic pattern transformation using a sub-sum consisting of a predetermined number of image data and transmitting a traffic characterization information update time message; Generating a new traffic parameter update value by searching a transmission speed and a buffer size based on the received traffic characterization information update time message; And a fourth step of attempting to reestablish a connection according to the transmission rate of the generated update value and transmitting the image data with the updated traffic parameter.

On the other hand, the present invention, to characterize the data traffic, an image device having a processor, comprising: a first function of connecting and transmitting video data having a variable bit rate input in real time at the maximum traffic rate; A second function of detecting burst data through traffic pattern transformation by using a sub-sum consisting of a predetermined number of image data and transmitting a traffic characterization information update time message; A third function of searching for a transmission rate and a buffer size based on the received traffic characterization information update time message and generating an update value of a new traffic parameter; And a computer readable recording medium having recorded thereon a program for realizing a fourth function of attempting to reestablish a connection according to the transmission rate of the generated update value and transmitting image data with the updated traffic parameter.

The objects, features and advantages described above will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, the terms used in the present invention are defined as follows.

-Moving Picture Expert Group (MPEG): A video compression standard for video communication established by the MPEG Committee, a technical standard organization for compressing and restoring video and related audio signals.

DVBR (Dynamic VBR): A method for traffic characterization of real-time video data proposed by the present invention.

-NIC (Network Interface Card): It is a card for network connection.

RSVP (Resource Reservation Protocol): One of the protocols for stream type data transmission. RSVP is mounted in routers, PCs, etc. to realize bandwidth reservation between IP routers.

-PR (Peak Rate): Refers to the minimum bandwidth (transmission rate) required to transmit video data without delay and loss. Transmission at rates below this bandwidth can result in queuing or some loss of data.

-AR (Average Rate): This is an abbreviation introduced for explanation in the present invention. AR is a value obtained by dividing the total video data amount by the total transmission time. DVBR uses AR (j) until j-th image data transmission because AR value cannot be obtained during traffic characterization.

b _min (minimum buffer size): This is the minimum buffer size required for transmitting image data through PR.

b _max (maximum buffer size): The maximum buffer size defined by the delay constraint. Because buffering causes queuing delays, the buffer allocation size is limited by the delay timeout.

-SSTM (Subsum Trace Monitoring): A method for monitoring the characteristics of an image data stream proposed by the present invention. In this way, the maximum burst information and the timing of the traffic change can be inferred.

-UP (Updating Point): It is time to reset the connection. SSTM module is to find UP.

Traffic specification: A set of parameters with values representing the characteristics of traffic. Speed (r) and buffer size (b) are among the traffic technologies.

SS (Subsum): Sub-sum of video data.

SO (Subsum order): SS is made up of the sum of SO video data.

SS _current : The current SS value.

SS _before : Previous SS value.

SS _state : A state value indicating whether a change in SS values tends to increase or decrease.

-MaxSS: Maximum SS value.

1 is a configuration diagram of an embodiment system for characterizing data traffic having a real-time variable bit rate applied to the present invention.

As shown in FIG. 1, the configuration includes an MPEG encoder 101 for encoding image data received from a digital camera or another type of image source, and a dynamic variable bit rate for transmitting the encoded image data in real time by applying the DVBR method. A transmission module 102 and a network interface card 106 for connecting and transmitting the image data to which the dynamic variable bit rate is applied are connected to the router 107 of the external network 108.

Here, the dynamic variable bit rate transmission module 102 monitors the traffic pattern of the input image data and then provides the traffic pattern monitoring module 103 for providing the input image data to the network interface 106, and the traffic pattern. Information determined by the transmission rate / buffer size updating module 104 and the transmission rate / buffer size updating module 104 for determining the transmission speed and the size of the buffer according to the traffic information of the image data monitored by the monitoring module 103. Dynamic resource reservation module 105 for transmitting a network resource reservation message to be updated based on the network interface card 106.

Hereinafter, the transmission operation of the image data based on the above configuration will be described in detail.

First, when actual video data is input to the MPEG encoder 101 through a camera or the like, the MPEG encoder 101 encodes the input video data in the form of a bit stream to generate a traffic pattern of the dynamic variable bit rate transmission module 102. Transmit to the monitoring module 103.

Here, the image data in the form of a bit stream output from the MPEG encoder 101 may be slightly different depending on the setting value of the MPEG encoder 101, but may be a frame stream of an IBBPBBPBBPBBPBB form or a standard according to a form such as PS or TS. Compressed stream. Hereinafter, the 'image data' referred to in the present specification refers to a frame stream, a PS, a TS, or any modified form of image information transmitted to a network interface card (NIC).

At this time, the traffic pattern monitoring module 103 of the dynamic variable bit rate transmission module 102 provides the received video data in the form of a bit stream to the network interface card 106, and continuously provides the traffic pattern to SSTM (SubSum Trace Monitoring) A monitoring message is sent to the transmission rate / buffer size update module 104 informing the timing of updating the traffic characterization information. Then, the transmission rate / buffer size update module 104 receives a message indicating the time point of updating the traffic characterization information, obtains a new transmission rate and a buffer size value through a convergence algorithm, and obtains a dynamic network resource reservation module ( 105).

The dynamic network resource reservation module 105 updates the router 107 supporting RSVP (Resource Reservation Protocol) of an external network based on the received new transmission rate value and buffer size value. The network resource reservation message and the video data from the traffic pattern monitoring module 103 are transmitted through the network interface card 106.

The flow of video data and control messages are as follows.

The raw image data is compressed through the MPEG encoder 101 and sent to the dynamic variable bit rate transmission module 102 in the form of an MPEG stream. Traffic characteristic information is extracted from the traffic pattern monitoring module 103, and the image data is sent to the network interface card 106 and transmitted to the external network 108 without changing its form.

When UP (Updating Point) information and the maximum burst size information are sent to the transmission rate / buffer size update module 104 where the convergence algorithm is performed, the transmission rate / buffer size update module 104 generates the traffic description information to generate the dynamic description. This is sent to the dynamic network resource reservation module 105 for reestablishing a connection through Resource Reservation Protocol (RSVP). In this case, the dynamic network resource reservation module 105 transmits an RSVP signaling message through the network interface card 106.

2 is a view illustrating a change in traffic characterization point for explaining a dynamic traffic characterization process according to the present invention.

Here, the traffic characterization of the variable bit rate refers to a process of determining the values of the traffic specifications promised to understand the variability of the corresponding traffic in the network. All variable bit rate traffic may be characterized by an ordered pair of transmission service rate (r) and buffer size (b) values, which are the most important parameters describing the traffic.

The dynamic VBR (DVBR) method of dynamically characterizing traffic reduces a transmission service rate value by dynamically estimating a (r, b) curve of source traffic targeted for traffic characterization, and a buffer size of QoS (Quality). We change the traffic characterization toward the maximum at a range smaller than the maximum limit that can guarantee the of service.

Here, the curve (r, b) means an optimal relationship curve between the transmission service rate r and the required buffer size b that the source traffic transmitted at a real time variable bit rate may ideally have. The (r, b) curve is ideally obtained only in the case of stored variable bit rate source traffic, but the transmission rate / buffer size update is dynamically performed in the dynamic variable bit rate transmission module 102 during transmission. Estimate using 104).

As shown in FIG. 2, when the vertical axis represents the transmission service rate r and the horizontal axis represents the required buffer size b in the curve (r, b), the vertical axis of the transmission service rate r represents the speed. It has an upper limit (PR: Peak Rate) and a lower limit of speed (AR: Average Rate). The horizontal axis of the buffer size b has an upper limit b _max of the buffer and a lower limit b _min of the buffer.

Here, the upper limit b _{max of} the buffer may be obtained as the maximum limit of the sum of the queuing delays of the queues for which the required buffer size b is allocated to all nodes on the transmission path as shown in Equation (1). .

,

I is one of all intersections.

On the other hand, the lower limit (b _min ) of the buffer is the size of the buffer required at the start position 201 of the transmission rate service at the peak rate (Peak rate) of the initial rate. That is, the start position 201 is a case of transmitting at a peak rate of speed, and requires a very large transmission service speed and a minimum buffer size. Therefore, there is no characterization delay required for the characterization of traffic, so it is suitable for the transmission of data having a real-time variable bit rate, but it is very inefficient for sharing and using limited network resources because an excessively large bandwidth allocation is required for the actual amount of data. .

The location 202, the characterization point of the DVBR, uses less network resources than that specified at the start location 201 or the lower speed limit location 204. However, since it is impossible to determine the position 203 in real time, the connection parameters are reset while starting from the transmission start position 201 and dynamically changing traffic parameters to the position 202 estimated as the position 203 on the way. This process uses the method of monotonically decreasing the transmission rate (r) and the buffer size (b) of the monotonically increasing method. Make it easier and more consistent.

3 is a flowchart illustrating a method for characterizing data traffic having a real-time variable bit rate according to the present invention.

As shown in FIG. 3, first, the connection request is performed at the maximum transmission rate in order to satisfy the QoS requirements of the delay sensitive real-time application program, and the transmission is started with the maximum bandwidth that can service the maximum burst for packet loss. 301).

In operation 302, a burst degree of data traffic is started by starting a SSTM (SubSum Trace Monitoring) mode for detecting a time point of a traffic parameter change. At this point, the setup parameters for the SSTM are initialized.

Thereafter, it is determined whether there is video data to be transmitted (303). As a result of the determination, the loop is terminated when there is no video data to be transmitted, and UP (updating point) detection is performed (304) when there is video data to be transmitted. This is equivalent to burst detection.

Here, the UP detection will be described in detail with reference to FIG. 4 to be described later.

Then, UP detection is performed through the UP detection process 304, and if the UP detection (burst data detection) succeeds (305), and if it does not succeed, the process proceeds to the step 303, and only when the UP detection is successful. An update value of the traffic parameter is newly generated through the convergence formula (306). Thereafter, it is determined whether the connection reset request through the generated traffic technology update value is successful (307).

As a result of the determination in step 307, if the connection reestablishment request through the traffic descriptor update value is successful, that is, when all intermediate nodes on the transmission path accept the update value of the traffic parameter, the transmission rate r and the buffer are updated with the updated traffic description. The size b value is updated (308), and the process returns to step 303 to determine whether there is data to be transmitted, and the loop is repeated.

In addition, if the determination result in step 307, the connection reset request through the traffic descriptor update value is not successful, the process returns to step 303 to determine whether there is data to be transmitted and repeats the loop.

FIG. 4 is a detailed flowchart illustrating an UP detection process illustrated in FIG. 3, and is a flowchart of determining a timing of a traffic parameter change.

Here, when the procedure of FIG. 4 is referred to as SSTM (SubSum Trace Monitoring), the initial parameters are started in the state initialized as follows.

That is, the image data index value I is initialized to 1, the current subsum SScurrent is 0, the previous subsum SSbefore is 0, and the increase / decrease trend SSstate is 'decreasing', and the maximum subsumption The size is 0, the initial speed r ₀ is PR, the buffer size b ₀ is the minimum buffer size, and the sub-sum order is initially determined (the amount that can include the sum of the image data corresponding to at least two GOPs). Initialize with

The procedure of FIG. 4 starts when there is image data to be transmitted as a result of the inspection in the above process (step 303 of FIG. 3), and transmits the i-th image data in step 401.

In operation 402, SS _current is generated. SS _current determines whether the i-th image data is a multiple of SO that can newly create an SS value, and when the i-th image data is a multiple of SO that can newly create an SS value, data subsequent to SO can be transmitted. Each time we add their magnitude to get the current SS _current . That is, one SS _current is generated for SO data transmission. (If the value of i is a multiple of SO, SS _current is generated.) On the other hand, if SS _current is not generated, UP (Updating Point) detection failure processing ( 412) and proceed to step 305.

Steps 404 and 405 are for determining the burst degree of the image data, and are for updating MaxSS by comparing the current SS value and the MaxSS value. That is, SS _current that when the generated Compare MaxSS and SS _current (404), MaxSS the SS _current greater than or equal to, and then proceeds to the next step (406) MaxSS is substituted for SS _current in the (404) MaxSS less than SS _current 405 proceeds to the next step 406.

Steps 406-409 are the processes for finding the UP. UP is the time when the trend of SS transitions from increasing trend to decreasing trend.

First, check whether SSstate is 'increasing' to find UP (406), and if SSstate is not 'increasing', substitute SS _current into SSbefore (403) and process UP detection failure (413) to process (305). Proceed to

On the other hand, it is checked whether SSstate is 'increasing' (406), and if SSstate is 'increasing', it is checked whether SSbefore is greater than SScurrent (407).

As a result, if SSbefore is not greater than SScurrent, SSstate is set to 'increasing' (408), SS _current is substituted into SSbefore (403), and UP detection failure processing (413) is performed to next process (305).

As a result, if SSbefore is greater than SScurrent, set SSstate to 'increasing' (409), notify UP and MaxSS (410), substitute SS _current to SSbefore (411), process UP detection success (414) Proceed to 305).

In the step 403, the SScurrent is generated after the i-th image data transmission, but the SSbefore is updated when the traffic descriptor update time (UP) is not reached. The step 411 is performed according to the success of the UP detection. After notifying the traffic descriptor update time, the SSbefore value is updated before going to the ready mode for recalculating the SS value.

FIG. 5 is a detailed flowchart of an embodiment of a process of generating an update value of the traffic parameter illustrated in FIG. 3.

As shown in Fig. 5, when a message informing the updating point of SSTM in the air is transmitted to the convergence algorithm, the average speed (AR (j) obtained by dividing the total size of the traffic (j image data) transmitted by the transmission time so far ): An average rate for the j video data is obtained (501).

Here, the average speed can be obtained through Equation 2 below.

AR (j): = TrafficSum (j) / IntervalSum (j)

Here, AR (j) is the average speed until the j th data is recognized by the traffic monitoring module, TrafficSum (j) is the sum of the total data sizes until the j th data is processed, and IntervalSum (j) is The time from the first data until the jth data is recognized by the traffic pattern monitoring module. Here, interval (I) refers to the time difference between the data monitored by the traffic monitoring module.

At this time, assuming an ideal case, the speed should be reduced to the average speed (AR). However, the present invention contemplates only real-time variable bit rate data transmission, which is sensitive to the amount of delay, so that the speed reduction up to the actual AR becomes inadequate due to the enormous queuing delay. Therefore, in the present invention, it is assumed that the buffer size corresponding to the b _max position of FIG. 2, which is defined as the limit of the queuing delay required by the upper application, is located before the average of PR and AR.

After calculating the average speed, the risk is calculated from the MaxSS value (502) to prevent the sudden change of speed. In this case, the calculation of risk is shown in Equation 3 below.

h (j) = ((maxSS (j)) / SO) * α

(However, α is a constant for controlling the reflection of the risk in calculating the risk.)

Then, the change in the rate of decrease through the risk is calculated, and the change is calculated as shown in Equation (4).

ΔRate (j): = (CR (j)-AR (j)) / (1 + h (j) / j)

Then, using the above Equation 3 and Equation 4, it is possible to obtain a traffic parameter to obtain the current rate to be updated.

Here, the current rate to be updated can be obtained through Equation 5 below.

CR (j + 1): = PR -ΔRate (j)

In addition, the new buffer size can be obtained through Equation 6 below through the speed update obtained through Equation 5.

r ₀ = CR (0) = PR,

b ₀ = I / PR,

b _j = b _j-1 + (CR (j-1)-CR (j)) * I

Here, r ₀ and b ₀ denote initial values, I denotes an image data arrival time interval, and b _j denotes a buffer size (ie, a previous buffer size) when transmitting a j-th image packet.

Equation 5 and Equation 6 are used to calculate the new speed update value and the buffer size according to the speed change (503).

Thereafter, it is determined whether or not the current speed CR (Current Rate) calculated in Equation 5 continues to monotonously decrease compared to the previous speed (504).

As a result of the determination in step 504, if the monotonic reduction does not continue compared to the previous speed, after discarding the update values of r and b currently obtained (507), the traffic descriptor is treated as not being updated.

On the other hand, as a result of the determination in step 504, if the monotonic reduction is continued compared to the previous speed, it is checked whether the updated speed (r) and the buffer size (b) meet the QoS (505). If it is unsatisfactory, it discards the currently obtained r and b update values (507), and if the traffic descriptor is not updated, it is processed. If satisfied, the updated values of the transmission rate (r) and the buffer size (b) are obtained. Confirm 506, and proceed to process 307 of FIG. 3.

The method of the present invention as described above may be implemented as a program and stored in a computer-readable recording medium (CD-ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.).

The present invention described above is not limited to the stated embodiments and the accompanying drawings, and it is common in the art that various substitutions, modifications, and changes can be made without departing from the technical spirit of the present invention. It will be evident to those who have knowledge of.

As described above, the present invention provides a very fast advantage for real-time variable bit rate data traffic because there is no time delay required for initial characterization when the connection having the real-time variable bit rate is compressed through an encoder and then immediately connected to the network. In addition, it is possible to efficiently use the network resources by designing the load in the network to be small while dynamically changing the traffic parameters according to the traffic characteristics from the time point to satisfy the required transmission delay.

Claims (7)

A characterization method of data traffic applied to an apparatus for characterizing data traffic having a real-time dynamic variable bit rate, A first step of connecting and transmitting video data having a variable bit rate input in real time at a maximum traffic rate; Detecting a traffic pattern change and burst data by using a sub-sum consisting of a predetermined number of image data, and transmitting the traffic characterization information update time message including the contents; Generating a new traffic parameter update value by searching a transmission speed and a buffer size based on the received traffic characterization information update time message; And A fourth step of attempting to reestablish a connection according to the transmission rate of the generated update value and transmitting image data with the updated traffic parameter; Characterization method of data traffic having a real-time dynamic variable bit rate comprising a. The method of claim 1, In the first step, the connection request at the maximum data rate is A method of characterizing data traffic having a real-time dynamic variable bit rate, characterized in that for removing the time delay required for optimal characterization of traffic having a variable bit rate. The method of claim 1, In the second step, the sub-sum is The data sum having a real-time dynamic variable bit rate, wherein the sub-sum is greater than the number of frames constituting at least two groups of pictures (GOP). Method of characterization. The method of claim 1, The third step, A fifth step of calculating an average rate (AR) and a risk of dividing the total size of the traffic transmitted by the transmission time if a message informing the convergence algorithm of the update time of the buster data detection (SSTM) is transmitted; And Calculating a speed change amount due to the risk, and adjusting a buffer size according to the speed change when the calculated current speed CR continuously decreases monotonically compared to the previous speed; And A seventh step of determining whether the adjusted speed and the size of the buffer satisfy a required quality of service (QoS) condition; Characterization method of data traffic having a real-time dynamic variable bit rate comprising a. The method of claim 4, wherein And calculating the buffer size increment by multiplying the speed reduction by the video data arrival time interval. The method of claim 4, wherein A method of characterizing data traffic having a real-time dynamic variable bit rate, wherein the change in speed occurs in a monotonically decreasing manner to facilitate network resource management. An imaging device having a processor for characterizing data traffic, A first function of connecting and transmitting video data having a variable bit rate input in real time at a maximum traffic rate; A second function of detecting a traffic pattern change and burst data by using a sub-sum consisting of a predetermined number of image data, and transmitting the traffic characterization information update time message including the contents; A third function of searching for a transmission rate and a buffer size based on the received traffic characterization information update time message and generating an update value of a new traffic parameter; And A fourth function of attempting to reestablish a connection according to the transmission rate of the generated update value and transmitting image data with the updated traffic parameter A computer-readable recording medium having recorded thereon a program for realizing this.