JP2513049B2 - Call connection control method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a call connection control method for preventing congestion in a network in a high speed packet switching method.

(Prior Art) The network handled here as an example of a packet switching network is a paper "Standardization trend of high-speed broadband ISDN centering on ATM" published by Kawaharazaki and Okada in IEICE technical research report information network IN87-110. It shall be transmitted and exchanged by the ATM method described in. ATM is a method of transmitting and switching packets called cells, and is characterized by using fixed-length packets.

Asynchronous Transfer Mode (ATM) is a store-and-forward system based on the principle of cell statistical multiplexing, which may cause quality degradation due to congestion. On the other hand, however, the protocol in the network is simplified to perform high-speed exchange, and there is no congestion control means in the network. Therefore, this ATM requires call connection control that predicts the input load and regulates the traffic input before accepting a call so as to prevent congestion.

This call connection control predicts whether or not the quality required for the network (referred to as specified quality) will be satisfied when the connection of the call with the call setup request is permitted and the load of the call flows into the network. However, if the condition is satisfied, the call setting is permitted, and if the condition is not satisfied, the call setting is rejected. In this call setting control, various traffic parameters of each multiplexed call are necessary in order to predict the communication quality of the network. Typical parameters include the maximum call sending speed and average sending speed (called AVE), and each call is required to declare these parameters to the network side at the time of call setup.

As an example of a typical method for determining whether or not a call can be set up by using these parameters, Watanabe et al. Review. " This determines a virtual band that is equal to or higher than the average transmission speed and equal to or lower than the maximum transmission speed for each call. Generally, the virtual band is given in the form of a function of a plurality of parameters such as the maximum sending speed of each call, the average sending speed, the transmission line speed, and the required quality. For each call, the transmission line secures a band in this virtual band. At this time, whether or not the call can be set is determined so that the total sum of the virtual bandwidths of the calls to be set becomes equal to or less than a certain value (for example, the transmission line capacity X percent).

Therefore, in call connection control, in order to guarantee the communication quality of the network, it is necessary to declare the traffic parameters to be declared as accurately as possible and to control the traffic to flow into the network within the declared value. It becomes important. Among the above-mentioned traffic parameters, the maximum transmission rate is uniquely determined from the hardware conditions of the packet terminal, but the problem is the average transmission rate.

Here, the average sending speed AVE is defined as follows. Second
The figure shows the sequence from the start of information transmission, the end of information transmission, and the release of the call (time B) after the call setup is completed (time A). Assuming that the call hold time from time A to time B is T and the total information transmission amount during that time is M, the average transmission speed of this call during call hold is defined by AVE = M / T.

Thus, although the average sending speed is not known until the end of the call, as described above, this value must be declared at the time of call setup request prior to transmission. Since the amount of information to be sent largely depends on the contents of communication, it is generally extremely difficult to predict this average sending speed. If the declared AVE does not match the actual AVE, especially if the actual AVE becomes larger than the declared value AVE, the specified communication quality is guaranteed even if call setting control is performed. become unable.

On the other hand, a control method has been proposed in which the traffic input is restricted for each call at the entrance of the network so that the traffic is not input into the network beyond the declared average transmission rate. For example, by Turner, International Zurich Semina '86 (International Zurich
Sminar'86) "New Directions in Communications" (New Directions in
Leaky bucket method (Leaky)
Bucket). This input regulation law is to measure an average value for each fixed small section during communication. However, how to determine the length of the small section becomes a big problem that determines the control performance. It is not easy to select this small section length for different communication requests for each call. In addition, a transmission limiting mechanism for the terminal side is required to control the input of traffic that exceeds the declared average speed, and control cannot be complicated.

(Problems to be Solved by the Invention) As described above, in the conventional high-speed packet switching, each call declares an average transmission rate in advance at the time of call setup, and the network uses this to allocate resources (bandwidth) in the network. Then, it is judged whether or not the call is set up so that the communication quality is guaranteed. However, the average delivery rate used there is
Actually, it can be defined only when the call ends, and cannot be accurately reported at the time of call setup. Calls that intentionally declare the average transmission rate small may be considered. In contrast,
When the average transmission rate of packets input to the network exceeds the declared value, it becomes extremely difficult for the network to guarantee the specified quality.

It is an object of the present invention to provide a call connection control method in which communication quality is guaranteed by making the average transmission rate declared at the time of call setup and the actual average transmission rate as close as possible in the call connection control method. To do.

(Means for Solving the Problem) In order to solve the above problems, the first invention is to declare in advance the average packet transmission rate of the packet call source when a call setup request is made in a packet communication network. Then, in the call connection control method for determining whether or not the call setup of the packet call is performed using the declared average transmission rate, the ratio of the total information amount transmitted in the network during the call hold period and the call hold time The given average transmission rate is measured, and if the measured average transmission rate is greater than the declared average transmission rate at the time when the call release request is made, the call hold time is extended. Is characterized by.

In a second aspect based on the first aspect, the call hold extension is extended according to a value obtained by dividing the total amount of information transmitted in the network until the call release request is generated by the declared average transmission rate. Characterized by determining the time.

A third invention is characterized in that, in the first invention, the call holding time is charged.

A fourth invention is characterized in that, in the third invention, when the measured average transmission speed is higher than the declared average transmission speed, an additional charge is added to the billing.

A fifth invention is, in a packet communication network, causing a packet average transmission rate of the packet call source to be declared in advance at the time of a call setting request, and using the declared average transmission rate, transmission within the network through which the packet call passes. In a call connection control method for determining a band to be secured in a path, the average transmission rate of each call given by the ratio of the amount of information transmitted in the network during the call hold period and the band securing time is measured, When the measured average transmission rate is higher than the declared average transmission rate at the end of the call, the bandwidth reservation time of the call to the intra-network transmission line is extended.

In a sixth aspect based on the fifth aspect, the extension time for securing the bandwidth is determined according to a value obtained by dividing the total amount of information transmitted in the network during the call holding period by the declared average transmission rate. It is characterized by doing.

A seventh invention is characterized in that, in the fifth invention, a fee is charged for the band securing time.

An eighth invention is characterized in that, in the seventh invention, when the measured average transmission speed is higher than the declared average transmission speed, an additional charge is added to the billing.

According to a ninth aspect of the present invention, in a packet communication network, when a call setup request is made, the packet average transmission rate of the packet call source is declared in advance, and whether or not the packet setup of the packet call can be set is judged using the declared average delivery rate. In the call connection control method to be performed, the average transmission rate during the call hold period is measured, and when the call ends the information transmission, the measured average transmission rate is higher than the declared average transmission rate. Is characterized by adding an additional charge.

(Operation) The principle of the present invention will be described below.

According to the invention described in claim (1), when the measured average transmission speed is higher than the declared average transmission speed at the end of the call information transmission, the call hold time is extended to increase the information transmission amount / It has the effect of reducing the average transmission rate defined by the call hold time. Specifically, the call hold time may be extended until the measured average sending speed reaches the declared average sending speed. At this time, the call holding state continues even after the transmission of the call information is completed, that is, virtual network resources (bandwidth used on the transmission path) are continuously secured during that period. Therefore, it is possible to regulate the input at the level of call connection control for a new call request. In other words, overall, all calls use network resources at the declared average transmission rate,
It is possible to guarantee reliable communication quality by appropriate network resource allocation. Further, if the call hold time is extended until the measured average transmission speed becomes lower than the declared transmission speed, the call setting regulation for the network can be further strengthened.
It is also possible to improve quality.

In the invention described in claim (2), the extension time of the call holding time of claim 1 is explicitly given at the time when the transmission of the call information is completed. In order to prevent excessive quality and quality deterioration, the call hold time may be determined so that the declared average transmission rate and the measured average transmission rate match. Therefore, from the relational expression of declared average transmission rate = total transmitted information volume / call hold time, call hold time = total transmitted information volume / reported average transmission rate.

According to the inventions of claims (5) and (6), the extension of the securing time of resources in the network (bandwidth of the transmission path) and the call process are separated from the items (1) and (2). . Item (1),
In the invention described in the item (2), even when the call information transmission is completed,
Call hold time is extended and the call process is not terminated. Therefore, by extending the time to secure the resources in the network (bandwidth used by the transmission path) and separating the call process, the call is released when the call information is sent out, and only the minimum required bandwidth in the network continues. To be done.

The inventions of claims (3) and (7) are defined by claims (1),
In the invention of the item (5), it is a matter that automatically imposes a penalty on a call that intentionally declares a small average transmission speed. In other words, billing is performed even during the extended call holding time or the band securing time. By charging even for the extended call holding time, it becomes possible to automatically charge as a penalty for the call for which the average transmission speed is declared small even after the transmission of the information is completed.

The invention of claim (9) is clearer than the inventions of claims (1) and (5) in comparison with the inventions of (3) and (7) with respect to calls for which the average transmission speed is declared small. It is possible to impose a penalty on. Claim (3),
In the invention according to (7), if the call holding time or the bandwidth reservation time is the same for both the call whose measured average value is larger than the declared average value and the measured value which is smaller than the declared value, the same charge is applied. become. Therefore, as in claim (9), if a higher charge than the normal charge is applied only to a call whose measured average transmission speed is higher than the declared average transmission speed, the penalty can be further clarified. Claim (4),
The invention of (8) is a combination of the invention of item (9) with the inventions of items (3) and (7).

In the present invention, the input transmission is not controlled for each packet transferred at high speed, but the average transmission speed is controlled at the level of call connection control, so that the realization is easy.

(Example) Next, an example of such an embodiment of the present invention will be described with reference to the drawings.

A. Outline Description FIG. 1 is a diagram for explaining an example of a call connection control method embodying the present invention. 100 is a packet network, and terminal 101-
103 are housed. Each terminal is a call connection controller 110-11
2 to the switching nodes 120-122. The switching nodes are connected to each other by transmission lines 140-142. Each call connection control device 110-112 is connected to the network resource management device 150. Each call connection control device 110-112 discriminates the call control packet and monitors the average packet transmission rate from the terminal. The network resource management device 150 manages the bandwidth of the transmission path in the network and performs the routing process for each call.
The switching control signal is notified to each switching node 120-122. In addition, charging processing is performed according to the network resources used by each call.

In this embodiment, call connection control devices 110-112 are distributed and installed for each terminal, and a network resource management device 150 is centrally installed. However, although other forms of load distribution and function distribution, such as a configuration in which both the call connection control device and the network resource management device are provided on each switching node side, the present invention is also applicable to other forms. .

Taking the case where the terminal 101 communicates with the terminal 102 as an example,
The outline of this embodiment will be described. The terminal 101 sends a call setup request to the terminal 102 to 110 before communication. At this time, the maximum transmission rate and the average transmission rate are declared to 110 as predicted traffic parameters during communication. This call setup request is made by a call control packet. The call connection control device 110 discriminates the content of the call control packet and further transfers a call control signal to the network resource management device 150 so that the call is set up on the transmission path to the target terminal.

When the network resource management device 150 receives the call control signal, it determines the transmission path (140 in this case) to the target terminal by the routing process, and the virtual band to be secured on the transmission path is notified of the virtual band. Calculated from the maximum delivery speed and average delivery speed. The virtual band has a value equal to or higher than the average transmission speed and equal to or lower than the maximum transmission speed. If the required bandwidth can be secured in the transmission path, the call setup is permitted. In this case, the switching control information (virtual circuit number, updating of the switching control table in the switching node, etc.) relating to this call is notified to each switching node 120, 121 in the form of a switching control signal. If bandwidth cannot be secured, call setup is rejected.

After the call is set up in this way, the terminal sends a data packet, which is sent by the switching node 12.
It is transferred to the terminal 102 via 0,121. At this time, the call connection control device 110 measures the average of the packet transmission speed of the call and monitors whether it is higher or lower than the declared average transmission speed.

When the transmission of the series of data packets is completed, the terminal 101 transfers the call release request packet to the call connection control device 110. The call connection controller 110 determines the contents of the call release packet and passes the contents to the network resource manager 150 in the form of a call control signal. At this time, the call connection control device 110 compares the declared average packet transmission rate with the measured average transmission rate, and passes the size determination result at the same time.

At this time, when the network resource management device 150 receives the call release request signal, the network resource management device 150 performs band securing extension processing and billing processing. If the measured average value is larger than the declared average value, the call hold time or the bandwidth reservation time is extended so that the measured average value matches the declared average value. When the extension of the band securing is performed in this manner, the fee is also charged for the extension. After the appropriate extension processing is completed, the reserved bandwidth of the call secured on the transmission path in the network is released.

The above is a rough call connection control sequence according to an embodiment of the present invention.

B. Sequence Example of Call Hold, Bandwidth Extension Processing, and Charge Processing In the following, the call hold or bandwidth extension processing and charge processing will be further described in detail.

FIG. 3 shows a call connection control sequence which implements the inventions of claims (1) and (2). In other words, the case where the measured value is larger than the declared average value and the processing of releasing the call itself is extended is explained.

In this figure, it is assumed that the horizontal axis is the time axis, the call is set at point A, packets are sent to b1, b2, and b3, and a call release request occurs at point B. b1-b3 mean the information amount (the number of bits) of each transmitted packet. The average transmission speed AVE 'declared at the time of call setup is shown by a solid line, and the average value measured by the call connection control device 110 is shown by a dotted line. The actual measurement value is represented by Σb / t (t is the elapsed time from the call setting), and increases when there is packet transmission and decreases when there is no packet transmission.

At point B when the call release request is made, the measured average value and the declared value are compared. When the actual measured value is larger than the declared value as shown in FIG. 3, the call release process is extended to the point B '. At point B ', the measured average value matches the declared average value. The band secured on the transmission path in the network for this extension time τ is continuously reserved. Therefore, during this period, it is possible to indirectly restrict new call setup of subsequent calls, and eventually traffic does not flow into the area between points A and B'above the average bandwidth declared for each transmission line. Become. In other words, the entire transmission line can be controlled so that packets can be sent within the declared average bandwidth over a long period of time, so the specified communication quality (delay, discard rate) can be controlled by appropriate bandwidth allocation management.
Will be guaranteed.

In the invention described in claim (1), the extension time τ is not specified. In other words, by extending this hold longer than the point B ', it is possible to hold the band on the transmission line for a longer time and further strengthen the regulation of the call setting of the subsequent call.

In the second aspect of the invention, the extension time τ is given so that the declared average value and the actually measured average value match. AV here
Since E '= Σb / (T + τ), the extended call hold time can be explicitly obtained by τ = Σb / AVE'-T.
This release may be performed after the time τ thus obtained has elapsed.

FIG. 4 shows a call control sequence which implements the inventions of claims (5) and (6). That is, the case is explained in which when the measured average value is larger than the declared average value, the call is released immediately and only the process of releasing the band secured on the transmission path is extended.

In this figure, as in FIG. 3, the horizontal axis is the time axis, and call setup is performed at point A and packets b1, b2, b3 are transmitted. Here, a call release request occurs at point B. b1-b3 are the information amount (the number of bits) of each transmitted packet, and the declared average transmission speed AVE 'and the actually measured average value are also shown. Here, as soon as a call release request occurs (point B),
Call release processing is performed. By this release processing, the virtual circuit of this call is released and the table in each switching node is updated. At this time, the measured average value and the declared average value are compared, and if the measured value is larger than the declared value, the band securing time is extended to point B '.
At point B ', the measured average value matches the declared average value.

Next, the charging processing operation will be described in detail with reference to FIG.
Similar to the case of FIGS. 3 and 4, FIG. 5 shows a sequence when the call is set up at the point A, the packet is transmitted up to the point B, and the call holding or bandwidth reservation is extended up to the point B ′. . At this time, according to the conventional time billing method, billing is only applied to the section T1. On the other hand, when the inventions of claims (3) and (7) are carried out, the time period during which the bandwidth of the transmission path in the network is substantially secured is extended, so section T2 = (total transmission amount Σb) /
For the (declared value AVE '), time is charged. As a result, for calls whose declared value is smaller than the actual measured value, the larger the amount sent, the smaller the declared value, the longer the charging section,
You will be charged a high fee automatically. That is, it has the effect of preventing calls that intentionally reduce the declared value.

When the inventions of claims (4), (8) and (9) are implemented, a penalty charge higher than the normal charge is added when extension of bandwidth reservation occurs. This is for example section T3
It is realized by carrying out a time charge higher than the section T1. As a result, even if the time for securing the band on the transmission path is the same, it is possible to charge the billing by distinguishing between those whose bandwidth securing time has been extended due to the declared value being exceeded and those whose bandwidth securing time is not. Therefore, the effect of preventing a call in which the declared value is intentionally reduced becomes large.

C. Configuration Examples of Call Connection Control Device and Network Resource Management Device Next, specific configuration examples of the call connection control device 110 and the network resource management device 150 according to the embodiment of the present invention will be described with reference to the block diagrams shown in FIGS. 6 and 7. Described in.

FIG. 6 is a block diagram of the call connection control device 110. 608
Receives the call control packet among the packets from the terminal and monitors the passing data packet. A call control packet such as a call setting request is processed by the call processing unit 600, and is notified to the network resource management device 150 as a call control signal so as to drive routing processing, bandwidth management processing, and accounting processing in the network.

Reference numeral 601 is a counter for measuring the amount of information in the transmitted data packet. Initialized at call setup, it keeps counting transit data packets until a call release request occurs. This is an object for measuring Σb in FIGS. 3 and 4. Reference numeral 602 is a timer for measuring the elapsed time from the call setup. From the output value τ of the timer 602 and the output value σb of the counter 601, a divider 605 obtains a measured average delivery speed Σb / T. The comparator 606 determines the magnitude of the declared average value AVE ′ held in the register 603 and the actually measured average sending speed Σb / T. This size determination result is notified to the network resource management device 150. The extension processing time τ of the band securing time is given by τ = Σb / AVE'-T and calculated by the divider 604 and the subtractor 607. The required extension time τ is also notified to the network resource management device 150.

FIG. 7 is a block diagram of the network resource management device 150. 700
Is a call processing unit such as intra-network routing processing, which notifies each switching node of a switching control signal. The switching control signal is used to release the setting of the virtual circuit of the call and to update the switching control table in the switching node. Processing unit
In response to the call setup / release request, 700 requests the bandwidth allocation processing unit 701 to secure / release the used bandwidth in the transmission path through which the call passes. The bandwidth allocation processing unit 701 is
The virtual bandwidth to be secured in the transmission line is calculated from the average transmission rate declared by each call. Bandwidth is secured / released on the transmission path with the obtained virtual bandwidth. At this time, if the bandwidth of the transmission path cannot be secured, the call processing unit 700 is notified to reject the call connection.

When a call release request occurs, the call connection control device 110 notifies the call processing unit 700 of the magnitude comparison result (extension command) of the declared average transmission rate and the measured average transmission rate. In response to this, the call processing unit 700 performs the processing of holding the call or extending the bandwidth reservation as described with reference to FIGS. 3 and 4. This extension processing may be performed only by extension processing of bandwidth reservation as shown in FIG. 4 and preceded by call release processing. The extension time is determined according to the time τ notified from the call connection control device 110. In addition, it is possible to further strengthen the network input restriction by holding the bandwidth reservation for a time τ or more.

The billing process is performed by the billing processing unit 702 and the penalty addition processing unit 703 when receiving the above-mentioned extension time, the declared average sending speed, and the magnitude comparison result of the measured value from the call connection control device 110. The charging processing unit 702 performs processing so that charging is also performed for the extended call holding time or bandwidth reservation time. Penalty addition processing unit 703, when the call release request, when the measured value is larger than the declared average transmission speed,
The charging processing unit 702 is notified to charge a higher charge as a penalty. For example, the extended holding time is charged higher than the normal holding time. Further, the additional fee can be fixedly added to the extended time.

(Effect of the Invention) According to the present invention, at the time of a call release request, when the measured average transmission rate is higher than the declared average transmission rate, the call holding time or the band securing time is extended.
This extension time is chosen so that the declared average delivery rate is equal to the measured average delivery rate. The extension time is explicitly given when the call information transmission is completed.

Suspending the bandwidth reservation of the transmission path means indirectly limiting the call setup request to other calls. Therefore, in the long term, the resources in the network are used at the average transmission rate declared by all calls, and the reliable communication quality can be guaranteed by appropriate network resource allocation.

In addition, by separating the extension of the time to secure the used bandwidth of the transmission path and the call release processing and advancing the call release processing first, the present invention can be easily implemented without limiting the call processing procedure.

In addition, by charging for the extended call hold time and bandwidth reservation time, it is possible to automatically impose a penalty fee on calls that declare the average transmission speed to be smaller than the actual average transmission speed. It has the effect of preventing calls that are intentionally declared smaller. For the extension time,
The above effect can be increased by adding a higher charge than the normal charge.

In the present invention, input control is not performed for each packet transferred at high speed, but control is performed so that the average transmission rate declared at the level of call connection control matches the actual average transmission rate, so the configuration is simple. Therefore, it is possible to easily guarantee the communication quality of the network.

[Brief description of drawings]

FIG. 1 is a block diagram of a packet network for explaining an example of a call connection control system embodying the present invention. FIG. 2 is a diagram showing a call connection control sequence in the prior art. FIG. 3 is a diagram showing a call connection control sequence implementing the inventions of claims (1) and (5), and FIG. 4 is a diagram showing a call connection control sequence implementing the inventions of claims 2 and 6. FIG. 5 is a block diagram of a configuration example of the call connection control device 110 in an example of the call connection control system embodying the present invention. FIG. 6 is a block diagram of a configuration example of the network resource management device 150 in an example of the call connection control system embodying the present invention. FIG. 7 is a diagram for explaining sections to be charged in the embodiments of the inventions of claims (3), (4), (7), (8) and (9). In the figure, 100 is a packet network, 101-103 are terminals, 110-112 are call connection control devices, 150 is a network resource management device, 120-122 are switching nodes, and 140-142 are transmission lines.

Claims (9)

(57) [Claims] 1. In a packet communication network, when a call setup request is made, the packet average transmission rate of the packet call source is declared in advance, and whether or not the call setup of the packet call is possible is determined using the declared average delivery rate. In the call connection control method, the average transmission rate given by the ratio of the total information amount transmitted in the network during the call hold period and the call hold time is measured, and the measured when the call release request occurs. If the average delivery rate is greater than the declared average delivery rate,
Call connection control method characterized by extending call hold time. 2. The call connection control system according to claim 1,
Call connection control characterized in that the call hold extension time is determined according to a value obtained by dividing the total amount of information transmitted in the network by the call release request by the declared average transmission speed. method. 3. The call connection control system according to claim 1, wherein the call hold time is charged. 4. The call connection control system according to claim 3,
A call connection control method, wherein when the measured average transmission rate is higher than the declared average transmission rate, an additional charge is added to the billing. 5. In a packet communication network, when a call setup request is made, the packet average transmission rate of the packet call source is declared in advance, and the declared average transmission rate is used to transmit the packet call in the network. In the call connection control method for determining the bandwidth to be secured in step 1, the average transmission rate of each call given by the ratio of the amount of information transmitted in the network during the call hold period and the time to secure the bandwidth is measured, A call connection control method, characterized in that, when the measured average transmission rate is higher than the declared average transmission rate at the end point, the bandwidth reservation time of the call to the intra-network transmission line is extended. 6. The call connection control system according to claim 5, wherein the total amount of information transmitted in the network during the call holding period is divided by the declared average transmission speed. A call connection control method characterized by determining the extension time of bandwidth reservation. 7. The call connection control system according to claim 5, wherein the bandwidth reservation time is charged. 8. The call connection control system according to claim 7, wherein when the measured average transmission rate is higher than the declared average transmission rate, an additional charge is added to the billing. Call connection control method. 9. In a packet communication network, when a call setup request is made, the packet average transmission rate of the packet call source is declared in advance, and whether the call setup of the packet call is possible or not is judged using the declared average delivery rate. In the call connection control method, the average transmission rate during the call holding period is measured, and when the call ends the information transmission, the measured average transmission rate is higher than the declared average transmission rate. A call connection control method characterized by adding an additional charge.