KR20060032741A - Effective Bandwidth Allocation System and its Method According to Traffic Quality Requirements - Google Patents

Abstract

The present invention relates to a system and method for actively allocating or reallocating effective bandwidth in order to increase resource utilization by using minimum resources while satisfying QoS requirements in a mobile communication system. According to an aspect of the present invention, there is provided an effective bandwidth allocation system according to a quality of service requirement for each traffic, the effective bandwidth allocation system according to a quality of service (QoS) requirement of a mobile communication system, comprising: an application system for inputting an input packet for each traffic class; Measuring means for measuring traffic-specific QoS parameters resulting from the operation of the applied system; A comparator for comparing the measured traffic-specific QoS parameters with the required target traffic-specific QoS parameters, respectively; A token bucket that waits an effective bandwidth for each traffic class corresponding to the compared parameter at a token generation rate; And a packet conditioner for processing the input packet according to the effective bandwidth for each traffic class in the token bucket. According to the present invention, by using a token bucket model that is easier to implement and has better performance than the existing active effective bandwidth scheme, the system can be allocated or reallocated the effective bandwidth to use a minimum amount of resources while satisfying the quality of heterogeneous traffic. Can be optimized.

Quality requirements, active effective bandwidth, token buckets, resource utilization

Description

A method for allocating effective bandwidth according to QoS requirement of each traffic, and a system therefore}

1 is a diagram illustrating an effective bandwidth allocation system according to traffic quality of service requirements according to an embodiment of the present invention.

FIG. 2 is a view for explaining the concept of a token bucket method for implementing the effective bandwidth allocation system of FIG.

3 is a diagram illustrating a correlation between an effective bandwidth and a packet loss rate obtained by applying an effective bandwidth allocation method according to an embodiment of the present invention when real-time video streaming and voice over IP are input. A diagram showing a relationship.

FIG. 4 is a diagram illustrating a correlation between an effective bandwidth and a handover drop rate obtained by applying an effective bandwidth allocation method according to an embodiment of the present invention when real-time video streaming and VoIP traffic are input.

5 is a diagram illustrating a value of an effective bandwidth allocated to apply an effective bandwidth allocation method according to an embodiment of the present invention when given QoS requirements for real-time video streaming and VoIP traffic.

6 is a diagram illustrating a VoIP traffic model when applying the effective bandwidth allocation method according to an embodiment of the present invention.

7 is a diagram illustrating VoIP traffic model parameters when applying the effective bandwidth allocation method according to an embodiment of the present invention.

8 is a diagram illustrating a real-time video streaming traffic model when applying the effective bandwidth allocation method according to an embodiment of the present invention.

9 is a diagram illustrating a real-time video streaming traffic model parameter when applying the effective bandwidth allocation method according to an embodiment of the present invention.

The present invention relates to an effective bandwidth allocation system and method according to the quality of service requirements for each traffic, and more particularly, in a mobile communication system, resource utilization by using the minimum possible resources while satisfying the quality of service (QoS) requirements A system and method for actively allocating or reallocating effective bandwidth to increase resource utilization.

Recently, the demand for many real-time multimedia application services that require end-to-end QoS guarantees such as Internet phones is increasing in the Internet. On the other hand, best-effort packet forwarding services are still provided in the current Internet. In other words, not only existing data services but also real-time multimedia services should provide packet forwarding in such a way that all traffic utilizes the bandwidth as quickly as possible without discrimination. This packet forwarding scheme provides the strict QoS guarantees required by real-time application services. Could not be satisfied.

Accordingly, the IETF has formed an Integrated Services (Intserv) work group and a Differentiated Services (Diffserv) work group to conduct research on how to provide integrated QoS in an IP-based network.

On the other hand, in a mobile communication system, the effective bandwidth is defined as a value between the maximum required bandwidth and the average bandwidth of the traffic. In addition, the effective bandwidth used may be reallocated taking into account the per-traffic QoS requirements (e.g., maximum packet loss rate required, maximum delay required, minimum yield required, etc.).

Conventional effective bandwidth techniques may be calculated by applying traffic tracking or statistical models collected in real time, or may be calculated directly from real time traffic tracking.

The effective bandwidth calculated in the above manner may also be used as a threshold for satisfying the QoS requirements and used to satisfy the QoS requirements. Accordingly, the concept of effective bandwidth plays an important role in satisfying QoS requirements in networks with limited resources.

Meanwhile, as a prior art, Korean Patent Application No. 2003-7004297 (filed March 25, 2003) discloses an invention entitled "System and method for designing, tracking, measuring, predicting and optimizing a data communication network". Is disclosed.

The preceding invention involves designing, tracking, measuring, and predicting a data communication network including a site specific model of a physical environment and performing various different calculations based on the components used in the communication network and the physical environment in which the network is deployed. And a system and method for optimizing.

On the other hand, as a prior art, Korean Patent Application No. 2003-0007431 (filed February 6, 2003), the invention named "remote access service system and information transmission method for transmitting information using minimum resources and bandwidth" Is disclosed.

However, the traffic tracking method collected in real time as in the prior art has a problem that it is difficult to implement because the effective bandwidth is calculated through a complex calculation. In addition, the effective bandwidth calculation method using a statistical model has a problem that it is difficult to reflect the instantaneous change of traffic.

SUMMARY OF THE INVENTION An object of the present invention for solving the above problems is to provide a system and method for allocating an effective bandwidth according to a quality of service requirement for each traffic for increasing system optimization by using minimum resources while satisfying the quality of heterogeneous traffic. will be.

In addition, another object of the present invention is to provide an effective bandwidth allocation system and method according to the quality of service requirements for each traffic reflecting the instantaneous changes of traffic while being easy to implement using a simple device such as a token bucket model .

As a means for achieving the above object, the effective bandwidth allocation system according to the quality of service requirements for each traffic according to the present invention,

In the effective bandwidth allocation system according to the quality of service (QoS) requirements for each mobile communication system,

An application system to which an input packet for each traffic class is input;

Measuring means for measuring traffic-specific QoS parameters resulting from the operation of the application system;

A comparator for comparing the measured traffic-specific QoS parameters with the required target traffic-specific QoS parameters, respectively;

A token bucket configured to queue effective bandwidth for each traffic class corresponding to the compared parameter at a token generation rate; And

Packet conditioner for processing the input packet according to the effective bandwidth for each traffic class in the token bucket

There is a feature configured to include.

Here, the packet conditioner is characterized by actively reallocating the effective bandwidth to the application system to satisfy the QoS requirements for each traffic.

Here, the input packet for each traffic class may be buffered for each of the traffic classes to further include a buffer for inputting to the application system according to a token generated at a preset token generation rate.

Here, the input packet is characterized in that the average generated data rate different for each traffic.

The traffic-specific QoS parameter may be a packet loss rate or a handover failure rate.

Here, the required QoS parameters for each target traffic may be provided from a parameter table.

Here, the packet conditioner is characterized in that for adjusting the effective bandwidth so that the difference value of the compared parameter is reduced.

On the other hand, as another means for achieving the above object, the effective bandwidth allocation method according to the quality of service requirements for each traffic according to the present invention,

In the effective bandwidth allocation method according to the quality of service (QoS) requirements for each mobile communication system,

a) measuring a traffic-specific QoS parameter that occurs as a result of the operation of the applied system to which the input packet for each traffic class is input;

b) comparing the measured traffic-specific QoS parameters with required target traffic-specific QoS parameters;

c) queuing the effective bandwidth for each traffic class corresponding to the compared parameter with the token generation rate of the token bucket;

d) reprocessing the input packet according to the effective bandwidth of the token bucket; And

e) actively reallocating the effective bandwidth to the applied system to satisfy the traffic specific QoS requirements

There is a feature consisting of.

Here, the step a) may include: a-1) buffering the input packet for each traffic class for each traffic class; a-2) inputting the input packet to the application system according to a token generated at a preset token generation rate; And a-3) measuring a QoS parameter for each traffic generated as a result of the operation of the application system.

Here, the QoS parameters for the target traffic required in step b) may be provided from a parameter table as a packet loss rate or a handover failure rate.

Here, step d) is characterized in that the effective bandwidth is adjusted so that the difference value of the parameter compared in step b) is reduced.

Accordingly, according to the present invention, by using a simple model such as a token bucket that is easier to implement and has better performance than the existing active effective bandwidth scheme, by allocating or reallocating effective bandwidth, the quality of heterogeneous traffic can be satisfied. You can optimize your system using the least resources possible.

Hereinafter, with reference to the accompanying drawings will be described in detail the effective bandwidth allocation system and method according to the quality of service requirements for each traffic according to an embodiment of the present invention.

According to an embodiment of the present invention, a simple device such as a token bucket model is provided to provide a method of calculating an effective bandwidth that is easy to implement and reflects an instantaneous change in traffic. In this case, the calculated effective bandwidth operates while associating with the QoS requirements for each traffic in the system design and actively reallocates the effective bandwidth according to the QoS requirements.

On the other hand, Figure 1 is a diagram illustrating an effective bandwidth allocation system according to the quality of service requirements for each traffic according to an embodiment of the present invention, the effective bandwidth allocation system according to the quality of service requirements for each traffic according to an embodiment of the present invention Buffers 110-1, ..., 110-n, application system 140, comparators 150-1, ..., 150-n, token buckets 130-1, ..., 130-n and packet conditioners 120 -1, ..., 120-n).

Referring to FIG. 1, in the effective bandwidth allocation system according to the quality of service requirements for each traffic according to an embodiment of the present invention, the application system 140 receives an input packet for each traffic class, wherein the buffer 110- 1,..., 110-n respectively buffer the input packets for each traffic class for each traffic class and input them to the application system 140 according to a token generated at a preset token generation rate.

Here, the application system 140 may include measurement means for measuring the QoS parameters for each traffic generated as a result of its operation.

The comparators 150-1,..., And 150-n compare the measured traffic-specific QoS parameters with the required target traffic-specific QoS parameters, respectively.

The token buckets 130-1,..., 130-n queue the effective bandwidth for each traffic class corresponding to the compared parameter at a token generation rate.

The packet conditioner 120-1,..., 120-n processes the input packet according to the effective bandwidth for each traffic class in the token bucket 130-1,..., 130-n. The packet conditioners 120-1,..., 120-n actively reallocate the effective bandwidth to the application system 140 to satisfy the traffic-specific QoS requirements.

Specifically, the packets of the input traffic classes are input to the corresponding buffers 110-1, ..., 110-n. That is, the data is input to the corresponding buffers 110-1, ..., 110-n at different data rates according to the traffic type, and processed at a constant token generation rate in the token buckets 130-1, ..., 130-n. do. At this time, the token generation rate according to the embodiment of the present invention becomes an effective bandwidth.

For example, the first service will be described below. After the first service is buffered in the corresponding buffers 110-1,..., 110-n, the first service is input to the system 140 to receive and apply a token generated at an x token generation rate (token / sec).

Here, if the average generated data rate of the input traffic is less than the token generation rate (token / sec) x, no packet loss will occur, but vice versa, packet loss will occur. Therefore, there will be an inverse relationship between the effective bandwidth (token occurrence rate) and the packet loss rate used in the embodiment of the present invention.

Comparing the difference between the QoS parameters of the first service actually measured and the target QoS parameter value required for the first service after processing the input packet in the application system 140 to apply the effective bandwidth as described above ( 150-1) for comparison. At this time, the requested QoS parameter value is provided from the parameter table 160-1. The parameter may be a packet loss rate or a handover failure rate.

Thereafter, the value of the effective bandwidth (token generation rate) for the first service is adjusted through the packet conditioner 120 so that the difference between the parameter values becomes smaller.

Similarly, for the n-th service, the buffer 110-n, the comparator 150-n, the parameter table 160-n, the token bucket 130-n, and the packet conditioner 120-n are connected to the first service. In the same way the value of the effective bandwidth can be adjusted.

2 is a view for explaining the concept of the token bucket method for implementing the effective bandwidth allocation system of FIG.

Referring to FIG. 2, there is typically a concept of a token bucket 210 as a model for defining the characteristics of traffic. The token bucket 210 may have two main parameters, a token generation rate R and a bucket size B. R defines an average data rate for a relatively long time, and B defines a data rate that can exceed R for a short time. To be more precise, the amount of data transmitted during time T does not exceed RT + B.

Also, the token generation rate R is a requirement for continuously provided bandwidth (bytes / second) for a flow, which is the average data rate entering the bucket and the bucket at the target ( Data rate is output at 210)

In other words, the bucket 210 may be viewed as a counter indicating the amount of data in bytes that can be transmitted at any moment. The bucket 210 is filled by byte tokens up to the bucket's capacity (maximum counter value) at a rate of R as the counter increments in R bytes per second. At this time, the IP packet is waiting in the queue 220 for processing after arrival. At this time, if enough tokens for the size of the IP data exist in the bucket 210, the packet can be processed immediately. This will pull that number of tokens out of the bucket. If enough tokens are not secured in bucket 210 when a packet arrives, the packet has exceeded the traffic specification (Tspec) of this flow. Typically, a packet that violates Tspec uses a method of marking the packet so that it can receive best-effort service, be discarded or discarded later.

The transmission rate of the IP data allowed by the token bucket method becomes R in the long term. However, if there is an idle period or a time interval in which packets are introduced at a relatively slow rate, tokens will accumulate in the bucket 210, and consequently can accommodate more B bytes than the promised transmission rate.

Accordingly, an embodiment of the present invention can calculate the effective bandwidth, which is simple and excellent in performance, by using the token bucket model described above, and the QoS parameter values and the requirements measured after applying the calculated effective bandwidth value to the system 140 can be calculated. Effective bandwidth can be actively adjusted using the token bucket model to minimize the difference between the target QoS parameter values.

Meanwhile, FIG. 3 shows the effective bandwidth and packet loss rate obtained by applying the effective bandwidth allocation method according to an embodiment of the present invention when real-time video streaming and voice over IP are input. It is a figure which shows the mutual relationship of.

Referring to FIG. 3, calls with VoIP traffic and real-time video streaming used were generated in a Poisson process with an average of 0.1. In addition, the movement of the UE occurred in a random direction model and simulated. Were used. Here, the Poisson process refers to a mathematical modeling of the behavior of thought that occurs randomly.

From the results of the effective bandwidth and the packet loss rate shown in FIG. 3, it can be seen that the packet loss rate decreases as the effective bandwidth according to the embodiment of the present invention increases. If the token generation rate (ie, effective bandwidth) of the token bucket used by the token bucket for the incoming traffic is small, the packet loss rate will be large, and vice versa, the packet loss rate will be small. Eventually there will be an inverse relationship between effective bandwidth and packet loss rate according to an embodiment of the present invention.

Therefore, in order to maintain the required packet loss rate in the system while using minimum resources, the embodiment of the present invention actively controls the token generation rate (effective bandwidth) of the token bucket to minimize the difference between the measured packet loss rate and the required packet loss rate. Will be reassigned.

On the other hand, Figure 4 is a diagram showing the relationship between the effective bandwidth obtained by applying the effective bandwidth allocation method according to an embodiment of the present invention, the handover drop rate when the real-time video streaming and VoIP traffic is input .

Referring to FIG. 4, it can be seen from the above result that the handover failure rate decreases as the effective bandwidth increases, and vice versa, the handover failure rate increases. The reason can be interpreted as follows.

Since the total bandwidth that can be supported in a cell is limited, only a limited user is served. In such a situation, if a high effective bandwidth is allocated to each user, the number of users supported in a cell is small, and eventually the handover failure rate is increased. Therefore, in order to keep the handover failure rate required by the system while using minimum resources, the effective bandwidth should be allocated to minimize the difference between the current measured handover failure rate and the required handover failure rate.

On the other hand, Figure 5 is a diagram illustrating the value of the effective bandwidth allocated to apply the effective bandwidth allocation method according to an embodiment of the present invention, given the QoS requirements for real-time video streaming and VoIP traffic.

Referring to FIG. 5, when the packet loss rate requirement for VoIP traffic and real-time video traffic is set to 1%, an effective bandwidth value that satisfies the requirement is 31 kbps of VoIP traffic and 33 Kbps. In addition, when the handover drop rate requirement is set to 8%, the effective bandwidth for VoIP traffic and real-time video traffic satisfying the requirement is equal to 31 kbps.

Therefore, the minimum effective bandwidth required to simultaneously satisfy the packet loss rate and handover failure rate requirements for the VoIP traffic and real-time video traffic is 31 kbps and 33 kbps for real-time video traffic. If the QoS requirements for each traffic are different, the effective bandwidth value allocated may also be different.

From the results of FIG. 3 and FIG. 5, the effective bandwidth value is adjusted as described above in order to use the minimum possible resources while satisfying QoS parameters (e.g., handover failure rate, packet loss rate, etc.) required in designing a mobile communication system. You can find it useful.

On the other hand, the model and parameter values for the above-described VoIP traffic and real-time video streaming used values of, for example, FIGS. 6 to 9. 6 is a VoIP traffic model, FIG. 7 is a VoIP traffic model parameter, FIG. 8 is a real time video streaming traffic model, and FIG. 9 is a real time video streaming traffic model parameter when applying the effective bandwidth allocation method according to an embodiment of the present invention. It is a figure which illustrates each. 6 to 9 are models and parameters for effective bandwidth allocation according to QoS requirements for each traffic according to an embodiment of the present invention, and thus detailed description thereof will be omitted.

As a result, according to an embodiment of the present invention, it is possible to increase system optimization by using the minimum resources possible while satisfying the quality for heterogeneous traffic.

While the invention has been shown and described in connection with specific embodiments thereof, it will be appreciated that various modifications and changes can be made without departing from the spirit and scope of the invention as indicated by the claims. Anyone who owns it can easily find out.

According to the present invention, the QoS requirement is calculated by calculating the effective bandwidth by a simple method such as the token bucket model of the mobile communication system and by using the correlation between the calculated effective bandwidth and various QoS requirements required for system design. You can optimize your system using the least resources possible while still being satisfied.

Claims (11)

In the effective bandwidth allocation system according to the quality of service (QoS) requirements for each mobile communication system, An application system to which an input packet for each traffic class is input; Measuring means for measuring traffic-specific QoS parameters resulting from the operation of the application system; A comparator for comparing the measured traffic-specific QoS parameters with the required target traffic-specific QoS parameters, respectively; A token bucket configured to queue effective bandwidth for each traffic class corresponding to the compared parameter at a token generation rate; And Packet conditioner for processing the input packet according to the effective bandwidth for each traffic class in the token bucket Effective bandwidth allocation system according to the quality of service requirements for each traffic comprising a. The method of claim 1, And the packet conditioner actively reallocates the effective bandwidth to the applying system to satisfy the QoS requirement for each traffic. The method of claim 1, And buffering the input packets for each traffic class for each of the traffic classes, and further inputting a buffer to the applying system according to a token generated at a predetermined token generation rate. The method according to claim 2 or 3, The effective packet allocation system according to the quality of service requirements for each of the traffic, characterized in that the input packet is a different average generated data rate for each traffic. The method of claim 1, The QoS parameter for each traffic is a packet loss rate or handover failure rate, effective bandwidth allocation system according to the quality of service requirements for each traffic. The method of claim 1, The required target traffic-specific QoS parameters are provided from a parameter table, effective bandwidth allocation system according to the quality of service requirements for each traffic. The method of claim 1, And the packet conditioner adjusts the effective bandwidth so that the difference value of the compared parameter is reduced. In the effective bandwidth allocation method according to the quality of service (QoS) requirements for each mobile communication system, a) measuring a traffic-specific QoS parameter that occurs as a result of the operation of the applied system to which the input packet for each traffic class is input; b) comparing the measured traffic-specific QoS parameters with required target traffic-specific QoS parameters; c) queuing the effective bandwidth for each traffic class corresponding to the compared parameter with the token generation rate of the token bucket; d) reprocessing the input packet according to the effective bandwidth of the token bucket; And e) actively reallocating the effective bandwidth to the applied system to satisfy the traffic-specific QoS requirements Effective bandwidth allocation method according to the quality of service requirements for each traffic including. The method of claim 8, wherein step a) comprises: a-1) buffering the input packet for each traffic class for each traffic class; a-2) inputting the input packet to the application system according to a token generated at a preset token generation rate; And a-3) measuring QoS parameters for each traffic resulting from the operation of the applied system Effective bandwidth allocation method according to the quality of service requirements for each traffic including. The method of claim 8, The required target traffic QoS parameter of step b) is a packet loss rate or handover failure rate, and is provided from a parameter table. The method of claim 8, The effective bandwidth allocation method according to the quality of service requirements for each traffic, characterized in that step d) adjusts the effective bandwidth so that the difference value of the parameter compared in the step b) is reduced.

Images