KR20040058389A - Method for controling adaptive cell admission and routing according to traffic characteristic - Google Patents

Abstract

PURPOSE: An adaptive cell accept and routing control method by traffic characteristics is provided to newly schedule a transmission path of cells to prevent the cells from being excessively lost, and to selectively accept the cells as controlling the path, thereby magnifying QoS control according to traffic characteristics. CONSTITUTION: When cells are inputted to a traffic characteristic identifier(S501), a system analyzes CLP bit or priority level bit information of headers of the cells, classifies the cells, and transmits the cells to a QoS monitor(S502). The QoS monitor calculates a cell loss rate(S503), and decides whether the cells correspond to a sensitive traffic class(S504). If so, the system compares the current loss rate with a threshold value, and decides whether the current loss rate is less than the threshold value(S505). If so, the system transmits the cells to a traffic output device without changing a path of the cells(S513).

Description

Method for controling adaptive cell admission and routing according to traffic characteristic}

The present invention relates to a method for adaptive cell admission and routing control by traffic characteristics, and more particularly, by using a double threshold of a cell loss rate given according to the characteristics of traffic for which a service is required and the current network situation. The present invention relates to a technology capable of discriminating and flexibly determining whether to provide a service and reconfiguring a service path.

In order to secure customer satisfaction in providing all application services through a communication network, a certain level of quality of service (hereinafter, referred to as QoS) must be guaranteed.

However, the traffic in the high-speed communication network, which must provide various and rapidly changing communication services in total, is a mixture of voice, data, and video, and infers a certain pattern of traffic because the traffic forms are constantly changing. There is a significant difficulty in estimating the demand for.

In order to efficiently accommodate these various traffic sources, it is necessary to grasp the characteristics of each characteristic and important service quality and to apply differentiated service provision methods. In order to solve the problem of satisfying several types of QoS simultaneously, it is effective to divide the problem into a plurality of small problem areas and approach the combination of appropriate techniques to solve each.

Indicators associated with QoS include a loss rate that indicates the probability of cell loss, a cell delay that is the amount of time delay to reach a desired destination, or a variation in cell latency. Cell delays can include propagation delays, switching delays, and queueing delays, while cell loss can be due to transmission errors, insufficient buffer space, and cell delay limits. Causes such as excess of.

Various techniques and approaches have been proposed to maintain the level of QoS required by the user, and the types of the proposed schemes are classified into the following categories.

The initial technique is to transmit a cell (Resource Management Cell) that requires resource allocation before the cell is actually transmitted, and to use the RM cell to periodically monitor the delay time of the cell returning through the network. Method, a method in which a node transmits an RM cell to the source node that sent the cell to adjust traffic transmission rate when any cell is discarded due to congestion, and a group of all cells on a per-message basis, not per cell. And control the runaway by deleting the.

There are two types of modern methods. The first method is that the receiving node of each link monitors the state of the standby buffer, and then regulates the traffic flow of the network by setting a maximum cell transmission allowance called "credit". The second is an end-to-end control scheme in which each transit node marks its state during congestion and induces the final destination node to feed back the result to the source node to adjust the cell rate.

However, the end-to-end control method has a disadvantage of causing a delay time required for waiting for feedback and unnecessary overload required therein, and there is a problem in that it is not possible to perform differential control according to characteristics of various traffics.

Many attempts have been made to control by traffic characteristics, which classify traffic delay-sensitive traffic (video or voice) and loss-sensitive traffic (data, etc.) by type to meet the QoS requirements of each. Various schemes have been proposed, including wait policies and prioritization policies in buffers.

However, most of these schemes control or prioritize queues at their own nodes or exchanges to selectively service the cell to avoid excessive congestion or to preferentially satisfy each traffic. The disadvantage is that it does not maximize the quality characteristics that should be.

Accordingly, the present invention has been made to solve the above-mentioned conventional problems, and the transmission path of a given cell can be newly scheduled to prevent excessive loss of the cell which should not be lost, and at the same time, it can add speed to delay-sensitive traffic. The purpose of the present invention is to provide an adaptive cell admission and routing control method for each traffic characteristic that can maximize QoS control according to the mixed traffic-specific characteristics of the broadband communication network.

1 is a block diagram of a system to which the adaptive cell acceptance and routing control method according to traffic characteristics according to an embodiment of the present invention is applied.

2 is a diagram illustrating a relationship between traffic and a double threshold applied to the present invention.

3 is a node configuration to which the present invention is applied.

4 is a flowchart illustrating a method for adaptive cell acceptance and routing control according to traffic characteristics according to an embodiment of the present invention.

In order to achieve the above object, an adaptive cell admission and routing control method for each traffic characteristic according to a preferred embodiment of the present invention,

A first step of identifying a traffic characteristic of the cell when a cell is input;

Calculating a cell loss rate (current loss rate) for the current path of the cell;

A third step of determining a path of the cell using the current loss rate, the L threshold, and the D threshold if the traffic characteristic of the cell is loss sensitive traffic;

A fourth step of determining a path of the cell using the current loss rate and the D threshold if the traffic characteristic of the cell is delay sensitive traffic; And

And a fifth process of transmitting the cell to the next destination path.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention in more detail.

Service quality management apparatus according to an embodiment of the present invention is a traffic input device 101, traffic characteristic identification unit 102, cell acceptance control unit 103, routing control unit 104, service quality monitoring unit 105, traffic output It is comprised including the part 106.

Hereinafter, the structure of each part is demonstrated in detail.

The traffic input device 101 is provided with a multiplexer (MUX) 10a to multiplex a cell input through n channels and output the multiplexed cell to the traffic characteristic identification unit 102.

The traffic characteristic identification unit 102 identifies traffic characteristics of individual cells inputted from the traffic input device 101 in units of cells, and thus delays sensitive traffic that must be relayed in real time even if some information loss such as voice or video is reduced. (Class D) and loss-sensitive traffic (Class L), which requires complete transmission of error-free end-to-end information even with some delay, such as data information.

The traffic can be classified by extracting and analyzing the CLP BIT, that is, the priority level BIT information, of the cell header, and by appropriately utilizing this information, according to the four types or other criteria as in the present invention based on the protocol of each integer value. Multiple priority settings and information delivery can be performed. For reference, the CLP BIT is a bit in which the priority information of a cell which is set at a fixed position of the cell header and is configured as a policy is stored as an integer value.

The quality of service monitoring unit 105 monitors the average probability of any cell being lost for all nodes connected to each other through the network and the link between the nodes, and inputs the path configuration information to any destination. By calculating the total cell loss rate to be lost while passing through a complete path for a given cell, it is determined which part of a given reference range the current state of the network is.

In addition, in consideration of the current network situation and the traffic type identified by the traffic characteristic identification unit 102 and the criteria set by the policy comprehensively, if the routing of the cell needs to be changed, the routing control unit 104 transmits the input cell. If the service itself of the cell is not available, the inputted cell is outputted to the cell acceptance control unit 103, and if the network situation is determined to be good, the inputted cell is outputted to the traffic output device 106.

This is possible by exchanging the loss rate of the current cell between the neighboring nodes and the topology and routing information of the network, and the function of the quality of service monitoring unit 105 is the function of one or more software modules operating in connection with a processing unit such as, for example, a microprocessor. It can be configured in combination.

Hereinafter, a probability of losing a cell in which a cell is lost during the entire path will be described.

Each variable is defined as follows when the path from node a to node d is a → b → c → d.

R (a, d): current root path set from departure node a to destination node d

R (a, d) = (a, b, c, d)

r (a, d, i): i-th node in R (a, d)

P (a, d): probability of losing any cell on any link a ~ d

H (a, d): HOP number from a to d

Here, the value of P (a, d) representing the probability of finally losing the cell via the path may be obtained from the cell loss rate for each section. In other words, assuming that a cell is transmitted from node a to node d via node b and node c, successful service is achieved when any cell is transmitted without being lost in all intervals. The probability of not being lost is obtained, and the cell loss rate is obtained by subtracting the probability of not being lost from 1 as shown in <Equation 1>.

P (a, d) = 1- (1-p (a, b)) × (1-p (b, c)) × (1-p (c, d)) <Equation 1>

In general terms, this is expressed as <Equation 2>

<Equation 2>

In addition, the service quality monitoring unit 105 determines the policy of the service according to the type of traffic classified by the traffic characteristic identification unit 102. The service quality monitoring unit 105 determines the selective service of the cell according to the criteria shown in FIG. The cell is sent directly to the traffic output section 106 or to the cell acceptance control section 103 or the routing control section 104. The optimal bypass path information extracted by the quality of service monitoring unit 105 is output to the routing control unit 104 in the form of an address value of the next node, and the routing control unit 104 transmits the next destination in the cell header. By changing the value of the address field and changing the output port of the traffic output device through the rearrangement of the elements / logic blocks in the exchange, the routing of the cell is ultimately changed.

As shown in FIG. 2, the criteria for accepting a cell and changing a path are two threshold values D threshold P _D and an L threshold P _L.

A cell having a traffic characteristic of class D, which is sensitive to delay, determines that the risk level of congestion has been reached only when the cell loss rate (hereinafter, referred to as the current loss rate) for the current path exceeds the D threshold. The cell is output to the cell 103 and immediately discarded, or the cell is sent to the routing control unit 104 under the condition that the number of HOPs does not increase, and then the cell is transmitted through the transmittable bypass path. However, when the current loss rate is less than or equal to the D threshold, all of the cells are accepted and output to the traffic output device 106 to be serviced immediately. This means that the service is provided along the current optimal path until the cell loss rate is above the threshold level, and when the cell loss rate is exceeded, the transmission time increases and discards when there is a high possibility of delay.

Cells with L-class traffic characteristics that are sensitive to loss are judged to accept that the path is in good condition only when the current loss rate is less than or equal to the L threshold, and otherwise, select a suitable bypass path irrespective of the number of HOPs, and delay a little. Even if you take the risk, you can reduce the possibility of loss. In this way, data loss can be eliminated without adversely affecting the quality of service for the delay sensitive traffic currently provided by the node.

The calculation method of the bypass path when the number of HOPs is considered or not and the method of calculating the cell loss rate at that time will be described with reference to FIG. 3.

Assuming that the path currently optimally set is a → b → c → d, the HOP numbers H (a, d) are three. Possible bypass routes below this HOP number include a → f → c− → d in FIG. Paths that do not consider the number of HOPs are other than a → e → f → g → d, a → f → g → h → d.

As described above, the next set of next stops (n, a, d) of the detour route that can provide the service configuration path from a to d are defined as N (a, d), and go from a to d but pass through z. The loss probability (P (a, d, z)) at time is shown in <Equation 3>.

<Equation 3>

Among these bypass paths, the path where the cell loss rate is minimum and the neighboring node at that time must satisfy <Equation 4>.

<Equation 4>

The equation for calculating the minimum loss rate among the paths where the HOP number is less than or equal to the HOP of the current path is as shown in <Equation 5>.

<Equation 5>

Hereinafter, a service quality management method through adaptive cell admission and routing control for each traffic characteristic according to an embodiment of the present invention will be described in detail with reference to the flowchart shown in FIG. 4.

When the cell is input to the traffic characteristic identification unit 102 (S501), after analyzing the CLP BIT or priority level BIT information configured in the header of the cell, the cell is classified by traffic characteristics, and then the quality of service monitoring unit 105 is performed. Transfer to (S503).

After the service quality monitoring unit 105 calculates a cell loss rate (current loss rate) when passing through the current path using Equation 2 (S503), does the cell correspond to a loss-sensitive traffic class Class L? Determine (S504).

As a result of the determination, if the cell belongs to class L (Yes in S504), it is determined whether the current loss rate is less than the L threshold by comparing the current loss rate with the L threshold (S505). If the current loss rate of the cell is less than the L threshold (P (a, d) < PL) (Yes in S505), it is determined that the network is good and transmitted to the traffic output device 106 without rerouting the cell. (S513).

However, as a result of the determination in step S505, if the current loss rate of the cell is equal to or greater than the L threshold value (P (a, d) ?? PL) (No in S505), the minimum of all bypass paths The bypass loss rate is calculated, and the bypass path and the next stop node z ^* having the minimum bypass loss rate are extracted (506).

Then, the quality of service monitoring unit 105 compares the current loss rate of the cell with the D threshold, and determines whether the current loss rate is less than the D threshold (S507). If the current loss rate of the cell is less than the D threshold (P (a, d) <P _D ) (Yes in S507), the current loss rate of the cell and the minimum bypass loss rate are compared (S508). If the current loss rate of the cell is less than or equal to the minimum bypass loss rate, the process proceeds to step S513 to transmit the cell to the traffic output device 106 without rerouting the cell.

However, if it is determined in step S507 that the current loss rate of the cell is equal to or greater than the D threshold (No) or if the current loss rate is greater than or equal to the bypass cell loss rate (Yes), it is calculated in step S506. The address value of the next pass through node Z ^* having the minimum detour loss rate is output to the routing controller 104 so that the cell is transmitted to the detour path passing through the node Z ^* (S509). In this case, when the current loss rate of the cell is equal to or greater than the D threshold, the reason for not comparing the current loss rate with the minimum bypass loss rate is that the current loss rate is so large that the cell unconditionally bypasses the cell with the minimum bypass loss rate. To bypass.

On the other hand, if it is determined in step 504 that the input cell belongs to class D which is delay sensitive traffic (No), it is determined whether the current loss rate is less than or equal to the D threshold by comparing the current loss rate with the D threshold. It is determined (S510). If the current loss probability of the cell is equal to or less than the L threshold (P (a, d) < P _D ) (Yes in S510), the process proceeds to step S513, where the next destination recorded in the header of the cell without rerouting the cell. Send the cell to.

However, as a result of the determination in step S510, if the current loss ratio of the cell is equal to or greater than the D threshold (P (a, d)> = P _D ) (No in S510), the currently set path is heavy and causes a congestion state. Since the probability of delaying or delaying the latency is very high, <Equation 5> is used to determine the minimum cell loss rate of the bypass path whose HOP number is less than or equal to the current HOP number, and the path and the next node (Z ^** ). Extraction (S511).

Next, the service quality monitoring unit 105 compares the current address of the loss rate and said minimum bypass loss rate (S512), if the loss rate is above the minimum bypass flow proceeds to step S509 only if the loss rate is less than the current node (Z ^**) A value is output to the routing controller 104 to change the cell to the bypass path so that the cell passes through the node Z ^** .

However, unlike the case described above, if the minimum bypass loss rate is equal to or greater than the current loss rate (No), the cell is output to the cell acceptance controller 103 and discarded (S514).

Meanwhile, the traffic output device 106 analyzes headers of cells whose paths output from the routing controller 104 have changed and cells whose paths have not been changed from the quality of service monitoring unit 105 and analyzes the headers of the next destination addresses. Route to.

According to the present invention as described above, a method for selectively determining the service acceptance of the cell and the scheduling of the route to the destination in consideration of the characteristics of any input traffic in a comprehensive information communication network in which various types of traffic must be complexly accommodated By presenting this, it is possible to efficiently manage the quality of service (QoS) of network traffic and to improve the quality of service.

In addition, the present invention reduces the time required to feed back the network conditions, set two thresholds that are set in policy, and monitor the probability of cell loss when passing through a given service path, while monitoring the entire service interval according to individual traffic characteristics. By performing differentiated control management across the network, even if the characteristics of the traffic are ignored or considered, the efficiency of limited resource operation of the network can be improved compared to the conventional method based on the priority control of the node itself. have.

On the other hand, the present invention is not limited to the above-described embodiment, but can be modified and modified within the scope not departing from the gist of the present invention, such modifications and changes should be regarded as belonging to the following claims. will be.

Claims (7)

A first step of identifying a traffic characteristic of the cell when a cell is input; Calculating a cell loss rate (current loss rate) for the current path of the cell; A third step of determining a path of the cell using the current loss rate, the L threshold, and the D threshold if the traffic characteristic of the cell is loss sensitive traffic; A fourth step of determining a path of the cell using the current loss rate and the D threshold if the traffic characteristic of the cell is delay sensitive traffic; And And a fifth process of transmitting the cell to the next destination path. The method of claim 1, The D threshold is Adaptive cell acceptance and routing control method for each traffic characteristic, characterized in that it is higher than the L threshold. The method of claim 2, The third process, A first step of comparing the current loss rate with the L threshold; If the current loss rate is less than the L threshold, determining a current path of the cell as the path of the cell; Calculating a minimum bypass loss rate for all bypass paths if the current loss rate is equal to or greater than the L threshold value; And Following the third step, comparing the current loss rate with the D threshold value, if the current loss rate is greater than or equal to the D threshold value, changes the path of the cell to a bypass path having the minimum bypass loss rate, and if the current loss rate is And a fourth step of determining the current path of the cell as the path of the cell if the D threshold is less than or equal to the D threshold value. The method of claim 3, wherein The bypass loss rate calculated in the third step is Assuming you pass through node (Z ^* ), Adaptive cell acceptance and routing control method according to traffic characteristics, characterized in that calculated by. The method of claim 1, The fourth process, A first step of comparing the current loss rate with the D threshold; If the current loss rate is less than the D threshold, determining a current path of the cell as the path of the cell; Calculating a minimum bypass loss rate for a path having a HOP number less than or equal to the HOP number of a current path if the current loss rate is greater than or equal to the D threshold; And Following the third step, comparing the current loss rate with the minimum bypass loss rate, if the current loss rate is greater than the minimum bypass loss rate, changes the path of the cell to the bypass path with the minimum bypass loss rate, and if the string And a fourth step of discarding the cell if the loss rate is less than the minimum bypass loss rate. The method of claim 5, wherein The minimum bypass loss rate calculated in the third step is Suppose we pass through node (Z ^** ), Adaptive cell acceptance and routing control method according to traffic characteristics, characterized in that calculated by. The method of claim 1, The current loss rate is, Adaptive cell acceptance and routing control method according to traffic characteristics, characterized in that calculated by. Where r (a, d, i): i-th node in R (a, d) P (a, d): probability of losing any cell on any link a to d H (a, d): HOP number from a to d