KR19990005224A - Traffic monitoring based fast call admission control method and apparatus - Google Patents

Abstract

The present invention relates to a traffic-monitoring-based high-speed call admission control apparatus, in which, when a call request with a maximum cell rate is requested, the maximum cell rate in the class is compared with the available available extra bandwidth, At least one sub-call admission control means for determining a link capacity value and outputting a link capacity value, at least one buffer means for storing a link capacity value obtained by the sub-call admission control means of the corresponding class, A server means for obtaining a monitoring value for traffic corresponding to each class at a predetermined period on the basis of the selected link capacity value and a means for calculating a cell distribution function based on the obtained traffic monitoring value, The capacity allocated to the sub call admission control means It is done on a per-control means, while the real-time control using a call admission control techniques that are based on the probability distribution function of the maximum selyul and traffic can be effectively used for any link traffic sources.

Description

A method and apparatus for fast call admission control based on traffic monitoring.

The present invention is based on traffic monitoring using a parallel controller for control of homogeneous traffic in a connection control unit of an asynchronous transfer mode (ATM) exchanger, : PCR), and performs a call admission control so as to make both the real time increase of the control and the reduction of the error compatible through the quick comparison operation.

In general, as a traffic control technique for implementing an asynchronous transfer mode (ATM) exchange scheme, which is a recommended method for providing an Integrated Service Digital Network (ISDN) service, Algorithm. In fact, these algorithms are commercialized in foreign countries and installed in their exchangers, or their control algorithms are commercialized and sold.

Typical call admission control (CAC) schemes currently in use include call admission control schemes based on equivalent bandwidth calculation and call admission control schemes based on burst modeling.

In the call admission control scheme based on the former equivalent bandwidth calculation, this technique approximates the bit rate generated in the multiplexed connection and obtains the equivalent bandwidth. Then, if there is a connection request of a new call, It is judged whether or not the remaining bandwidth is exceeded.

The equivalent bandwidth refers to a minimum bandwidth satisfying a requirement for a quality of service of the call and is larger than an average cell rate and is smaller than a peak cell rate.

There are various methods for calculating the equivalent bandwidth, and representative methods are as follows.

This is a method of calculating by using only the characteristics of traffic irrespective of the capacity of the entire physical link. The equivalent bandwidth for each call is obtained and an equivalent bandwidth for the entire traffic is obtained based on the queuing analysis.

The equivalent bandwidth of each call with a cell loss requirement threshold of < RTI ID = 0.0 > epsilon < / RTI >

Where a = 1n ε, Rp is the peak cell rate, ρ is the ratio of the average cell rate to the maximum cell rate, b is the average burst length, and x is the size of the buffer.

At this time, when n connections are multiplexed, the required total bandwidth C is calculated as shown in Equation (2).

Assuming a Gaussian distributlon for the aggregated traffic in order to consider the multiplexing effect, the total bandwidth size C is obtained by Equation (3).

Where m _i is the average bit rate and σ i ² is the variance.

In addition, in the call admission control scheme based on the latter burst modeling, this technique is called a Peak Bit Rate (PBR) scheme regardless of the distribution of on / off periods of a cell arrival processor. ) And an average bit rate (ABR).

The probability that the number of bursts in an ON state at any moment when the n virtual channels with the maximum bit rate (RBR) = R and the average bit rate (ABR) = a are multiplexed is k (4) "

Therefore, the probability that the size of bandwidth required when k virtual channels are multiplexed in the link is kR is assumed to be P (n, k).

At this time, if the utilization rate of the bandwidth used by the connected calls is determined as the probability of exceeding the limit, the acceptance of the new call can be determined as shown in Equation (5) below.

This means that if the physical link bandwidth is C and the total number of concatenated calls containing the new call is n then the probability that the size of the total bandwidth to be used by the connected calls after a new call is connected does not exceed 90% Appropriate level. That is, if it is smaller than (1 - ε), the connection is allowed, otherwise it is rejected.

However, in the call admission control scheme based on the equivalent bandwidth calculation used in the asynchronous transfer mode (ATM) exchange scheme described above, it is difficult to calculate the exact equivalent bandwidth of individual calls in advance, and when the number of traffic sources is small, And the time and accuracy of calculating the total equivalent bandwidth is problematic, so that it is difficult to control in real time.

In the call admission control scheme based on the burst modeling, it is easier than the call admission control scheme based on the equivalent bandwidth calculation, but the calculation becomes complicated under a heterogeneous traffic environment and the relationship between the burst traffic characteristic and the service quality is unclear There is a drawback that the structure of the call admission control section is complicated and the implementation and real-time control become difficult.

Therefore, the present invention excludes the degradation of the link utilization efficiency and the computational complexity in a heterogeneous traffic environment, which occurs when the number of traffic sources is small, in the above-described conventional techniques. As one aspect of the present invention, The present invention provides a method and apparatus for fast call admission control based on traffic monitoring that enables efficient use of a link for an arbitrary traffic source in a real-time control system using a call admission control scheme based on a distribution function.

Another object of the present invention is to increase the real-time performance of the control based on traffic monitoring and using only the maximum cell rate as a parameter provided by the source.

Another object of the present invention is to reduce the error of the controller and improve the utilization efficiency of the link, rather than the call admission control scheme based on the burst modeling.

As another aspect of the present invention, there is an object to prevent an overload of an exchange by preventing traffic and entry in advance.

As another aspect of the present invention, there is a need to improve the communication quality by reducing the control time required for entering a call.

Brief Description of the Drawings Fig. 1 is a block diagram illustrating an embodiment of the present invention,

2 is a block diagram illustrating the sub-call admission control unit of FIG. 1 in more detail;

3 is a signal flow diagram for calculating an equivalent bandwidth in an exchange using the asynchronous transmission scheme of FIG.

Figure 4 shows a signal flow diagram for the call admission control of Figure 1;

DESCRIPTION OF THE REFERENCE NUMERALS

100: first sub-call permitting control section 101: second sub-

102: M sub-call permission control unit 103: First buffer unit

104: second buffer unit 105: M-buffer unit

106: switching unit 107: server unit

According to an aspect of the present invention, there is provided a traffic-monitoring-based fast call admission control apparatus for managing a maximum cell rate and a usable available bandwidth in a class, At least one sub-call admission control means for determining whether or not to accept a connection and outputting a link capacity value, at least one buffer means for storing a link capacity value obtained by the sub-call admission control means of the corresponding class, Server means for obtaining a monitoring value for traffic corresponding to each class in a predetermined period of time from a link capacity value selected by the switching means, Based on this distribution function, Calculating the bandwidth to be characterized by including a capacity allocation control means to allocate the sub-call admission control means.

In the above-described traffic-monitoring-based fast call admission control apparatus according to the present invention, the sub-call admission control means may compare a maximum cell rate inputted and a size of an available bandwidth allocated by the capacity allocation control means And an available bandwidth calculating means for calculating a new available bandwidth with an available bandwidth and a maximum cell rate obtained by the comparing means.

The capacity allocation control means may include a cell rate distribution determination means for obtaining a cell distribution function based on the traffic monitoring value obtained by the server means, The available bandwidth determining means for determining the available bandwidth and assigning the available bandwidth to any one of the sub-call admission control means.

The traffic monitoring-based fast call admission control apparatus according to the present invention is characterized in that the available bandwidth in the available bandwidth calculation means is determined by subtracting the sum of the equivalent bandwidths of the respective classes from the total link capacity.

According to another aspect of the present invention, there is provided a fast call admission control method based on traffic monitoring, the method comprising: receiving parameters necessary for calculating an equivalent bandwidth for traffic of a certain class; Setting an intermediate value between an upper limit value and a lower limit value of the equivalent bandwidth, calculating a cell loss rate for traffic of an arbitrary class in the corresponding update period of the equivalent bandwidth, calculating an upper limit value and a lower limit value of the equivalent bandwidth, Determining whether the cell loss rate is equal to or less than a cell loss rate requirement limit value, outputting the calculated equivalent bandwidth if the cell loss rate requirement limit is less than or equal to the cell loss rate requirement limit value, , The added cell loss rate and the cell loss rate requirement Designating the presently designated equivalent bandwidth as an upper limit value if the added cell loss rate is equal to or lower than the cell loss rate requirement; Determining whether the number of cells is greater than the designated value; and repeating the steps after the upper and lower limit designation steps if the number is less than the designated value, and outputting the upper limit value of the currently designated equivalent bandwidth if it is greater.

In order to achieve the objects of the present invention, there is provided a traffic monitoring based fast call admission control method, comprising: when a connection request for a new call corresponding to a class is received, the available bandwidth available to a user in a communication network, Comparing the available bandwidth with the maximum cell rate, and if the available bandwidth is less than the maximum cell rate, rejecting the call connection; and if the available bandwidth is greater than the maximum cell rate, allowing the connection of the call and updating the available bandwidth .

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of a traffic monitoring based fast call admission control apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

Prior to the description of the present invention, the call admission controller is required to be capable of real-time control and efficient link utilization for any traffic source.

In the present invention, a controller having a parallel structure is proposed to satisfy both of these conditions, and a call admission control scheme based on a maximum cell rate (PCR) and a probability distribution function of traffic is proposed.

The proposed parallel controller has M first to Mth sub-call admission control units (100 to 102).

The call requesting connection is determined by the controller of the corresponding class among the M control units whether or not the call is accepted for connection and the cells generated in the corresponding class are outputted to the outputs of the first to Mth sub- And are stored in first to Mth buffer units 103 to 105 connected to the first to Mth buffer units 103 to 105, respectively.

The proposed method uses only the maximum cell rate as a traffic parameter provided by the user.

This is because when the average cell rate is used, it is impossible for the traffic sources to accurately estimate the average cell rate in the case of a specific traffic, and thus it is not suitable as a traffic parameter for the call admission control.

On the other hand, the maximum cell rate can be estimated relatively accurately.

However, when the call admission control is performed based on only the maximum cell rate except for the average cell rate, information on the traffic becomes insufficient. Therefore, a probability distribution function for the number of cells arriving during a certain measurement period is obtained through actual measurement of traffic.

And we estimate the equivalent bandwidth used by current traffic from the obtained probability distribution function.

The proposed scheme accepts a connection when the maximum call rate of the call is less than the available extra bandwidth when a new call makes a connection request, or rejects the connection if it is not.

The extra bandwidth is determined by subtracting the sum of the equivalent bandwidth of each class from the total link capacity.

The technique of the present invention is a parameter provided by a traffic source and performs call admission control using only the maximum cell rate so that the structure of the controller is simplified compared to the call admission control scheme based on the equivalent bandwidth calculation, .

Also, compared to the call admission control scheme based on burst modeling, the efficiency of link utilization can be improved by reducing the error of the controller.

In addition, congestion in a computer network occurs when a larger amount of traffic flows into the network than can be handled by the network, which can be caused by unexpected fluctuations in the traffic flow or obstacles in the network.

In a high-speed communication network environment such as an asynchronous transfer mode (ATM) network in which an error rate is extremely low, a cell loss due to buffer overflow is the biggest cause of errors.

Congestion in an ATM network can seriously degrade service quality. A series of actions taken to minimize the impact of congestion is called congestion control.

Congestion control schemes that can be used in ATM networks are divided into two categories: Preventive Congestion Control Scheme and Reactive Congestion Control Scheme depending on the time of control application.

The preventive congestion control technique is a technique for predicting traffic conditions of a communication network and performing appropriate control before traffic congestion occurs.

In high - speed transmission protocols such as ATM, proactive congestion control techniques have been reported to perform better than adaptive congestion control schemes.

One of the proactive congestion control schemes is the call admission control scheme.

Call admission control is defined as a series of actions performed by a communication network to control a virtual channel connection (VCC) or a virtual path connection (VPC) during a call connection process.

The purpose of call admission control is to ensure that when there is a connection request for a new call, it determines its accessibility, so as not to degrade the traffic generated by the new call and the quality of service of the previously connected call.

The network user presents the parameter of the user traffic and the service quality information to the corresponding access controller of the access node and waits for the decision of whether or not to make a call connection.

At this time, the call admission control should be capable of real-time control and be designed to have high link efficiency within a range that can guarantee the quality of service of user traffic.

Therefore, the call admission control scheme according to the present invention proposes a controller having a parallel structure, and proposes a call admission control scheme based on a maximum cell rate (PCR) and a probability distribution function of traffic.

The proposed parallel controller has M first to Mth sub-call admission control units (100 to 102).

The call requesting the connection is determined by the controller of the corresponding class among the M control units whether or not the call is accepted for connection. The cells generated in the corresponding glass are connected to the outputs of the first through Mth call admission control units 100 through 102 And are stored in the first to m-th buffer units 103 to 105 connected to the first to m-

In order to control call for uniform traffic, we propose a control scheme that can efficiently utilize the link for arbitrary traffic sources with high real-time controllability.

The proposed method uses only the maximum cell rate as a traffic parameter provided by the user.

The reason for this is that although the traffic sources can not accurately estimate the average cell rate in the case of specific traffic when using the jiggling cell rate, it is not suitable as a traffic parameter for the call admission control, but the maximum cell rate can be estimated relatively accurately to be.

However, when the call admission control is performed depending on only the maximum cell rate except for the average cell rate, information on traffic is insufficient. Therefore, a probability distribution function for the number of cells arriving in a certain measurement period is obtained through actual measurement of traffic

The equivalent bandwidth used by the traffic corresponding to each class is estimated from the probability distribution function obtained above.

Thus, the proposed scheme accepts a connection if the maximum number of cells of the call is less than the available extra bandwidth when a new call makes a connection request, and rejects the connection otherwise.

The extra bandwidth is determined by subtracting the sum of the equivalent bandwidth of each class from the total link capacity.

This will be described in detail with reference to FIG.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram illustrating an embodiment of a high-speed call admission control apparatus based on the traffic monitoring of the present invention.

According to the present embodiment, when there is a call request having the maximum cell rate from the first to Mth input terminals 108 to 102, the maximum cell rate in the class is compared with the available extra available bandwidth (100 to 102) for determining whether or not to accept a connection and outputting a link capacity value, and a link control unit (100 to 102) for receiving a link input from the first to Mth sub- A switching unit 106 for selecting a link capacity value input from the first to Mth buffer units 103 to 105, A server unit 107 for obtaining a monitoring value for traffic corresponding to each class in a predetermined period of time from the link capacity value selected by the server unit 106 and outputting the monitoring value through an output terminal 11, Traffic traffic Obtaining the cell distribution function by the gettering value is configured to calculate the bandwidth available to them on the basis of the first to capacity allocation control unit 112 that allocates the M sub-allowed call control section (100 to 102).

The first sub-call admission control unit 100 includes a comparison unit 100a for comparing the maximum cell rate under pressure with a size of an available bandwidth allocated and input by the capacity allocation control unit 112, And an available bandwidth operation unit 100b for calculating a new available bandwidth with an available bandwidth and a maximum cell rate input from the mobile station.

The capacity allocation control unit 112 includes a cell rate distribution determination unit 112a that obtains a cell distribution function based on the traffic monitoring value input from the server unit 107, And an effective usable bandwidth determiner 112b for determining the bandwidth limit and allocating the bandwidth to the first to Mth sub-call admission control units 100 to 102.

In this way, the configured high-speed call admission control apparatus based on the traffic monitoring can allocate bandwidth to the parallel controller when the number of incoming traffic sources from the first to Mth input terminals 108 to 110 is M for call admission control And a first to Mth buffer units 103 to 105, respectively.

When there is a connection acceptance request of the calls entering the access node, the first to Mth sub call permission control units 100 to 102 of the class determine whether to accept the connection.

The asynchronous transfer mode cells entering from the connected call are stored in the first to Mth buffer units 103 to 105 connected to the first to Mth sub-call admission control units 100 to 102 to wait for a service.

The traffic corresponding to each class is monitored and the equivalent bandwidth is updated using it.

The first to the M sub-called allow controller (100 to 102), when you are calling for the class to provide the maximum selyul R _p as the traffic parameters connection compared in value with a comparison unit (100a) of the available bandwidth available for this acceptance .

At this time, if the call is connected, the available bandwidth calculator 100b of the sub-call admission control units 100 to 102 calculates the new available bandwidth by subtracting the maximum cell rate Rp from the calculated available bandwidth, To be input to the units 103 to 105.

If the available bandwidth is larger than the maximum cell of the call, the comparison unit 100a permits connection of the call, and if the available bandwidth is smaller than the maximum cell rate of the call, the connection is refused.

The size of the available bandwidth is determined by subtracting the sum of the equivalent bandwidth of each class from the total link capacity.

The first to Mth buffer units 103 to 105 store the output values of the first to Mth sub call permission control units 100 to 102 and provide the output values to the server unit 107 through the switching unit 106 .

The server unit 107 obtains a monitoring value for traffic corresponding to each class at a predetermined period of time from the link capacity value input from the first to Mth buffer units 103 to 105 and outputs the monitoring value through the output terminal 111 To the cell rate distribution determination unit 112a of the capacity allocation control unit 112. [

The cell rate distribution determination unit 112a obtains a cell distribution function based on the traffic monitoring value input from the server unit 107 and provides the cell distribution function to the available bandwidth determination unit 112b.

The effective usable bandwidth determiner 112b calculates an available bandwidth based on the cell distribution function obtained by the cell rate distribution determiner 112a and assigns the available bandwidth to the first to Mth sub call permission controllers 100 to 102 .

The capacity allocation controller (N2) determines the service capacity for each class.

The equivalent bandwidth of each traffic class to the sum of the bandwidths of the entire traffic class is defined as the class capacity.

The server unit 107 scheduled by the class capacity transmits the cells stored in the buffer of each class to the physical channel by round robin.

In the above traffic measurement, important measurement parameters in the case of actually measuring the traffic multiplexed on the link are as follows.

a) Renewal period: N

One update period consists of N measurement periods. In other words, it updates after N measurements.

If the update period is long, the dynamic adaptability to the change of traffic drops, but the influence of the time delay for calculation of the used bandwidth is reduced.

On the other hand, if the update period is short, adaptation according to the traffic change is fast, but the number of samples decreases and the accuracy of the probability distribution becomes poor.

b) Measurement Reflection Ratio (MRR): a

This value is a weight that determines how much of the value obtained through actual measurement is reflected in the existing probability distribution function.

In the probability distribution function estimation, the probability that the number of cells arriving during the measurement period S in the nth update period is K is p ⁽ⁱ⁾ n (k ⁽ⁱ⁾ ; n).

This probability distribution function is the best but it is difficult to define it correctly. Therefore, this probability distribution function profit estimate is expressed as p ⁽ⁱ⁾ (k ⁽ⁱ⁾ ; n).

In Equation (7), R denotes the maximum number of cells that can be reached during the measurement period, and a probability distribution function when a call with a maximum cell rate of Rp is accepted as a connection is estimated as shown in Equation (8).

If an existing call that was multiplexed on the link is released, no action is taken to simplify control.

This will waste bandwidth for a while, but it will soon be calculated by actual bandwidth.

When the probability distribution function for the number of cells arriving during the measurement period s is given, the cell loss rate in the nth update period is obtained as shown in Equation (9).

The bandwidth satisfying the limited cell loss rate can be obtained using Equation (9).

The equivalent bandwidth calculation defines a service requirement condition for a user's cell loss rate for any i-th class traffic as ε ⁽ⁱ⁾ .

( ⁱ⁾ (k ⁽ⁱ⁾ ; n + 1) of the probability distribution function of the number of cells to be transmitted during the measurement period s of the ith class of traffic in the ⁽ n + 1) The equivalent bandwidth C ⁽ⁱ⁾ used by the traffic of the class can be calculated.

The proposed equivalent bandwidth calculation algorithm for any i-th class is shown in FIG.

The calculation of the equivalent bandwidth will be described in detail with reference to FIG.

First, the parameters necessary for the equivalent bandwidth calculation are inputted (ST100).

Here, the parameter P ⁽ⁱ⁾ represents the maximum value (Peak) for traffic of the i-th class and A ⁽ⁱ⁾ represents the average value of the traffic of the i-th class.

A variable bit rate (VBR: Variable Bit Rate), so the case of a service equivalent bandwidth is determined between the average selyul and up selyul Here, the initial value of the P ⁽ⁱ⁾ the upper limit value b ^(i), the lower limit of A ⁽ⁱ⁾ a ⁽ⁱ⁾ (Step: ST101).

Then, an intermediate value between the lower limit value a ⁽ⁱ⁾ and the upper limit value b ⁽ⁱ⁾ is set as the equivalent bandwidth (ST102).

Thereafter, a cell loss rate P ⁽ⁱ⁾ loss (n) for traffic of the ith class in the update period is calculated (step: ST103).

If the calculated cell loss rate is? ^(I) , the process proceeds to step ST105, and if the cell loss rate is not? ^(I ), the process proceeds to step ST106.

The step (ST105) and the cell loss rate, so this case is ε ^⒤ a bandwidth equivalent to the calculated critical value and outputting the C ^⒤ end.

In step ST106, if the cell loss rate is not? ^(I) , the next cell is added to the calculation.

If the cell loss is larger than this? ^(I ) in step ST107, the process proceeds to step ST108. In step ST108, if the cell loss rate is larger than? ^(I) , since the currently designated equivalent bandwidth is under-estimated compared to the real traffic, the current equivalent bandwidth is designated as the lower limit, and the flow advances to step ST110.

In step ST109, if the cell loss rate is smaller than? ^(I) , the current equivalent bandwidth is over estimated relative to the actual traffic, so the current equivalent bandwidth is designated as the upper limit value, and then the process proceeds to step ST110.

If the number of cells counted so far is smaller than the specified value N, the process after step ST101 is repeated. If the counted number of cells is greater than the specified value N, And proceeds to step ST111.

The step (ST111) and outputs a current of a specified upper limit value b ^⒤ since the equivalent bandwidth b ⁽ⁱ⁾ values in one update period of the end state and then exit.

In addition, the connection admission control scheme is considered to be performed in units of one update period.

This is given the probability that the bandwidth currently used by the connected traffic, the bandwidth available to the user, and the number of cells when the nth update period begins.

If a connection request for a new call corresponding to the i-th class is received, the network compares the available bandwidth available to the user with the maximum cell rate of the new call.

If the available bandwidth is greater than the maximum cell rate of the call, the call is allowed to connect and the call is refused if the available bandwidth is less than the maximum cell rate of the call.

This will be described in detail with reference to FIG.

That is, when a new call connection request corresponding to the i-th class is received, the communication network compares the available bandwidth (C _BW ) available to the user with the new maximum cell rate R _p ⁽ⁱ⁾ (ST112).

If it is determined in step ST112 that the available bandwidth is greater than the maximum cell rate, the process proceeds to step ST114.

If the available bandwidth is smaller than the maximum cell rate, the process proceeds to step ST113.

In step ST114, if the available bandwidth is greater than the maximum cell rate of the call, the call connection is allowed.

In step ST113, if the available bandwidth is smaller than the maximum cell rate of the call, the connection is refused and the call is terminated.

If the connection of the new call is allowed, the available bandwidth C _BW is updated (ST115).

As described above, in this embodiment, efficient link utilization is possible for a random traffic source while real-time control is performed using a call admission control scheme based on a maximum cell rate and a probability distribution function of traffic.

While specific embodiments of the invention have been illustrated and described, it will be obvious that the same may be varied in many ways by those skilled in the art.

Such modified embodiments should not be understood individually from the technical idea and viewpoint of the present invention, and such modified embodiments should be included in the appended claims of the present invention.

It is clear from the above description that efficient link utilization is possible for a random traffic source while real-time control is performed using a call admission control scheme based on a maximum cell rate and a probability distribution function of traffic through a relatively simple hardware framework, It is possible to prevent the overload of exchanger by preventing entry in advance.

In addition, the communication quality can be improved by reducing the control time required for entering the call through a simple comparison operation.

Claims (6)

At least one sub-call admission control means for comparing the maximum cell rate in the class with a size of available extra available bandwidth to determine connection acceptance and outputting a link capacity value at the time of a call request having a maximum cell rate, At least one buffer means for storing a link capacity value obtained by the sub-call permission control means of the class, switching means for selecting a link capacity value obtained by the buffer means, And a capacity allocation unit for calculating a cell distribution function based on the traffic monitoring value obtained by the server unit and calculating the available bandwidth on the basis of the cell distribution function, Characterized in that it comprises a control means High-speed traffic monitoring based call admission control device. The apparatus of claim 1, wherein the sub-call admission control means comprises: comparison means for comparing the maximum cell rate inputted and the size of the available bandwidth allocated by the capacity allocation control means; and means for comparing the available bandwidth and the maximum cell rate And a usable bandwidth calculation means for calculating a new usable bandwidth based on the available bandwidth. 3. The traffic monitoring based fast call admission control apparatus according to claim 2, wherein the available bandwidth in the available bandwidth calculation means is determined by subtracting the sum of the equivalent bandwidths of the respective classes from the total link capacity. 2. The apparatus of claim 1, wherein the capacity allocation control means comprises cell rate distribution determination means for obtaining a cell distribution function based on a traffic monitoring value obtained by the server means, and determining an available bandwidth based on the obtained cell distribution function And a usable bandwidth determining means for assigning the usable available bandwidth to any one of the sub-call admission control means. The method comprising the steps of: receiving parameters required for calculating an equivalent bandwidth for traffic of an arbitrary class; designating an upper limit value and a lower limit value of the equivalent bandwidth after inputting the parameter; setting an intermediate value between the upper limit value and the lower limit value of the equivalent bandwidth; Calculating a cell loss rate for traffic of a given class in a corresponding update period of the equivalent bandwidth, determining whether the calculated cell loss rate is less than or equal to a required condition threshold, And if it is equal to or less than the cell loss rate requirement limit value, adding the next cell to the calculation if the cell loss rate requirement limit is greater than or equal to the calculated cell loss rate requirement limit, , The added cell loss rate Designating the presently designated equivalent bandwidth as an upper limit value when the loss rate requirement is equal to or lower than the loss rate requirement, determining whether the counted number of cells is greater than a designated value after designating the lower limit value and the upper limit value, and repeating the steps after the upper limit value and lower limit value designation step And outputting an upper limit value of the currently designated equivalent bandwidth if the difference is larger. The method of claim 5, further comprising the steps of: comparing a usable bandwidth available to a user in a communication network with a maximum cell rate of a new call when a new call connection request corresponding to the arbitrary class exists; Rejecting a call connection if the available bandwidth is smaller than the maximum cell rate, and if the available bandwidth is greater than a maximum cell rate, updating the available bandwidth by allowing connection of the call.