KR100503417B1 - QoS guaranteed scheduling system in ethernet passive optical networks and method thereof - Google Patents

Abstract

A QoS guaranteed scheduling system and method in an Ethernet passive optical network is disclosed. The first buffer stores a first traffic that requires a guarantee of quality of service (QoS). In the second buffer, a second traffic to which a priority transmission order is given is stored. The scheduler detects queue length information of the first buffer and the second buffer according to the inflow of the first traffic and the second traffic, and transmits the queue length information to the bandwidth allocation device, and according to the bandwidth allocated from the bandwidth allocation device, the first buffer and the second buffer. Allocates bandwidth for each traffic by calculating the expected service time based on the inflow and length of incoming traffic. The mux multiplexes the traffic input through the scheduler and sends it to the bandwidth allocation device. Accordingly, by using dual DEB-GPS scheduling to guarantee the limited delay time for traffic that requires quality assurance and no loss of traffic, it is possible to guarantee the quality of service while using the bandwidth for the best service as efficiently as possible. have.

Description

QoS guaranteed scheduling system in ethernet passive optical networks and method

The present invention relates to a QoS-guaranteed scheduling system and method in an Ethernet passive optical network. More particularly, the present invention relates to an Ethernet passive optical network, while satisfying a delay time constraint when transmitting data from an optical network unit to an optical fiber terminator. The present invention relates to a system and method for guaranteeing QoS through slot assignment capable of secure data transmission.

Passive Optical Network (PON) is an optical access subscription that directly connects optical fiber to subscribers in order to provide broadband services such as various data services and high-definition video services, which are rapidly increasing in the information society. It is self-reliance The PON is a network that connects a central office (CO), a provider of services, and subscribers (subscribers), using only passive optical devices. Services such as multiplexed voice, data, and video on the PON are carried in optical signals and passively transmitted to subscribers through optical fibers and optical splitters shared by the subscribers. Currently, PON is implemented in various forms, and among them, Ethernet Passive Optical Network (PON) using Ethernet is most prominent.

In the conventional E-PON, the same baseband is divided in time, and a central base station allocates a divided time slot to each subscriber station and transmits data. However, in an E-PON using Time Division Multiplexing (TDM), the optical network unit (ONU) of each subscriber station can transmit data only in the assigned time slot, so QoS should be guaranteed. In case of traffic, there is a problem that it is difficult to guarantee a satisfactory QoS.

TDM schemes for transmitting data by dividing the bandwidth in the conventional E-PON by time units include a polling scheme and an interleaved polling with adjusted cycle time (IPACT). The polling method is a method of allocating a fixed slot when allocating an uplink bandwidth of an ONU in an optical line terminator (OLT). The adaptive polling method is a method of dynamically allocating bandwidth in the conventional TDM technique in which a time period is applied to the polling method.

However, the conventional TDM scheme has a problem in that it is not possible to accommodate the demand of the traffic that changes in the fixed slot allocation. In addition, it is difficult to efficiently use the bandwidth, and the delay time is long depending on the number of ONUs, which makes it difficult to guarantee QoS due to the low scalability of the ONUs. In order to overcome this problem, a scheme of dynamically allocating slots has been used. However, in the conventional TDM technique, schemes of dynamically allocating bandwidth (IPACT, etc.) do not guarantee QoS by focusing only on efficient bandwidth usage. have. In addition, the IPACT method is difficult to make fair bandwidth allocation according to the traffic characteristics on the Ethernet and is limited in the scalability of ONU.

SUMMARY OF THE INVENTION The present invention provides a scheduling system that guarantees QoS by controlling delay time and delay variation that are restricted for a transmission flow requiring QoS by using a new slot allocation scheme for uplink data transmission in an Ethernet passive optical network. To provide a way.

Another technical problem to be solved by the present invention is a scheduling method for guaranteeing QoS by controlling delay time and delay variation that are restricted for a transmission flow requiring QoS by using a new slot allocation scheme for uplink data transmission in an Ethernet passive optical network. The present invention provides a recording medium that records a program for executing the program on a computer.

In order to achieve the above technical problem, an embodiment of the QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention includes storing a first traffic required for guarantee of quality of service (QoS). A first buffer; A second buffer in which a second traffic to which a priority transmission order is given is stored; The queue length information of the first buffer and the second buffer is detected and transmitted to the bandwidth allocation apparatus according to the inflow of the first traffic and the second traffic, and the first buffer and the first buffer according to the bandwidth allocated from the bandwidth allocation apparatus. A scheduler for allocating bandwidth for each traffic by calculating an expected service time based on an inflow amount and a length of traffic flowing into a second buffer; And a mux for multiplexing the traffic input through the scheduler and transmitting the multiplexed traffic to the bandwidth allocation apparatus.

Another embodiment of the QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention for achieving the above technical problem, is received from each subscriber station having a plurality of memory areas, each connected to the memory area; A first storage unit for storing the first traffic for which QoS is required, a transmission rate required for each of the first traffic, and queue length information; Each of the memory areas has a plurality of memory areas, and each of the memory areas has a second traffic given priority transmission received from each of the subscriber apparatuses connected thereto, a transmission rate required for each of the second traffics, and queue length information. The second storage unit is stored; And preferentially allocating a bandwidth for the first traffic based on queue length and traffic information for traffic requests received from at least one connected subscriber device, and for the first traffic at allocable bandwidth. And a scheduler which allocates bandwidth to each of the second traffics by dividing the allocated bandwidth by the number of the second traffics.

In order to achieve the above another technical problem, an embodiment of the QoS guaranteed scheduling method in an Ethernet passive optical network according to the present invention comprises: (a) a queue length for traffic requests from at least one connected subscriber device; And receiving traffic information; (b) allocating bandwidth to a first traffic requiring QoS based on the traffic information; (c) allocating bandwidth for each second traffic by dividing the bandwidth subtracted from the bandwidth allocated for the first traffic from the total allocable bandwidth by the number of second traffic to which priority transmission is given; And (d) informing the subscriber device of the bandwidth allocation for the first traffic and the second traffic.

According to the present invention, by using dual DEB-GPS scheduling to guarantee the limited delay time for traffic that requires quality assurance and no loss of traffic, to ensure the quality of service while using the bandwidth for the best service as efficiently as possible can do.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the QoS guaranteed scheduling system and method in an Ethernet passive optical network according to the present invention.

1 is a diagram illustrating a configuration of a QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention.

Referring to FIG. 1, the QoS guaranteed scheduling system 100 in an Ethernet passive optical network according to the present invention includes a subscriber station 110 and a network termination device 150.

The subscriber device 110 includes a switch fabric 112, a first buffer 114, a second buffer 116, a subscheduler 118, and a mux 120. The switch fabric 112 selectively outputs the packet flowing into the subscriber device 110 to the first buffer 114 and the second buffer 116 according to the characteristics of the packet. The switch fabric 112 outputs the packet to the first buffer 114 when the incoming packet is a packet requiring QoS guarantee, and outputs the packet to the second buffer 116 when the packet is the best packet. The first buffer 114 stores packets that are required to guarantee QoS received from the switch fabric 112, and the best packets are stored in the second buffer 116.

The sub-scheduler 118 is a main scheduler 136 provided in the network terminator 130 to queue length information according to the flow flowing into the first buffer 114 and the second buffer 116 from the switch fabric 112. Send monitored monitoring information. In addition, the sub-scheduler 118 is based on the inflow and length of the packets input to the first buffer 114 and the second buffer 116 according to the bandwidth allocated by the main scheduler 136 of the network termination device 130. Calculate the expected service time and allocate bandwidth for each packet. The mux 120 multiplexes a packet input through the subscheduler 118 and transmits the packet to the network terminator 130.

The network termination device 130 includes a first storage unit 132, a second storage unit 134, a main scheduler 135, and a weight buffer 138. The first storage unit 132 has a plurality of memory areas, each of which stores QoS required traffic, transmission rate required for each traffic, and queue length information received from each of the connected subscriber stations. do. The second storage unit 134 has a plurality of memory areas, each of which stores the best traffic received from each of the connected subscriber stations, the transmission rate required for each traffic, and the queue length information.

The main scheduler 136 updates and calculates the required information according to the QoS class of the traffic request collected from each subscriber device connected to the network termination device 130. At this time, the main scheduler 136 assigns a high weight to the traffic requiring QoS guarantee, and allocates bandwidth first. In addition, the main scheduler 136 allocates the bandwidth for each best traffic by the weight given by dividing the bandwidth subtracted from the bandwidth allocated for the traffic requiring QoS guarantee from the allocable total bandwidth by the total number of best traffics. .

The weight buffer 138 stores weights according to QoS classes and weights according to characteristics of best traffic.

In addition, the scheduler provided in the subscriber device 110 and the network termination device 130 is preferably a DEB-GPS scheduler to perform the schedule operation by the DEB-GPS (Deterministic Effective Bandwidth-Generalized Process Sharing) scheme.

2 is a flowchart illustrating a process of performing a QoS guaranteed scheduling method in an Ethernet passive optical network according to the present invention. In the following description, for convenience of description, one subscriber device is connected to the network termination device as an example, but it will be easily understood by those skilled in the art that a plurality of subscriber devices may be connected to the network termination device.

Referring to FIG. 2, on the other hand, the network termination device 130 transmits and receives a message for operating and managing the physical layer periodically with the connected subscriber device 110 (S200). In addition, the network termination device 130 receives the slot allocation information and the buffer status information from the subscriber device 110 and allocates information and a permission message of the slot to be allocated to each subscriber device 110 (S210).

The subscheduler 118 of the subscriber device 110 reports the queue length and the traffic information for each traffic flowing into the first buffer 114 and the second buffer 116 to the network terminating device (S220). The main scheduler 136 of the network termination device 130 checks whether the traffic request satisfies the limited delay time (S230). If the limited delay time is not satisfied, the subscriber station 110 requests the subscriber device 110 to reset and adjust the QoS requirements (S240). If the limited delay time is satisfied, the subscriber station 110 requests the DEB of the traffic requested by the subscriber device 110. Accordingly, by assigning a predetermined weight, the bandwidth to be allocated to the subscriber station 110 is calculated and then the calculated bandwidth is allocated (S250).

In operation S250, the main scheduler 136 of the network termination apparatus 130 updates and calculates information required according to the QoS class of the traffic request collected by the subscriber apparatus 110. At this time, the main scheduler 136 allocates bandwidth by first giving a fixed fixed weight to the traffic for which quality of service is required. In addition, the main scheduler 136 divides the bandwidth subtracted from the bandwidth allocated for the QoS traffic from the allocable total bandwidth by the total number of the best traffics and allocates the bandwidth to the best traffic by the weight given to each best traffic. In this bandwidth allocation process, the main scheduler 136 refers to the buffer length information calculated by referring to the time intervals in which the buffers 114 and 116 of the subscriber station 110 are emptied.

Steps S240 and S250 correspond to GPS functions performed by the network termination device 130.

The subscheduler 118 of the subscriber device 110 allocated the bandwidth compares and calculates the delay variation between the input traffic and the delay variation allowable error to determine whether the delay variation limited to the allocated bandwidth is satisfied (S260). ). To this end, the subscheduler 118 of the subscriber station 110 calculates the service time expected by the arrival and length of each traffic based on the type of the incoming traffic and the allocated bandwidth, and then allocates the fair bandwidth between the traffics. For this purpose, the variation in delay time between traffics is calculated by the following equation.

Where A _i (t): incoming traffic of the i th stream,

φ _i : weight of the i-th stream,

W _i (t): amount of incoming i-th stream traffic,

Q _i (t): queue length of the i th stream,

r _i : allocated ratio of incoming i-th stream,

D _i and δ _i : delay time and delay variation of the i th stream,

d _i and σ _i : The experienced delay and delay variation of the i th stream.

If the allocated bandwidth does not satisfy the constrained delay variation, the subscriber station 110 resets and adjusts QoS requirements (S270). If the constrained delay variation is satisfied, the subscriber station 110 allocates a bandwidth to each traffic to provide a service. Provided (S280).

3 is a diagram illustrating uplink transmission from a subscriber device configuring a QoS guaranteed scheduling system to an Ethernet termination device in an Ethernet passive optical network according to the present invention. Hereinafter, a concept of uplink transmission from a subscriber device to a network termination device constituting a QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention will be described with reference to FIG. 3. 3 illustrates a system in which a plurality of subscriber devices 320, 330, and 340 are connected to one network termination device 300.

The technique employed in the present invention is a bandwidth allocation technique of a dual DEB-GPS scheduler based on a time division multiple access (TDMA) scheme. The bandwidth allocation concept of the scheduler is related to the use of the upstream bandwidth from each subscriber station (320, 330, 340) to the network termination device (300). The overall configuration of the system consists of the requests 325, 335, and 345 for each user and the subscriber devices 320, 330, and 340 for accommodating the users, the network terminator 300 for allocating bandwidth, and the respective subscriber devices 320. , 330, 340 and a splitter 310 for branching traffic at the branch point of the network termination device 300.

Requests 325, 335, and 345 requested from respective users are communicated to the network termination device 400 through respective subscriber devices 320, 330, and 340. Each subscriber station (320, 330, 340) transmits its own transmission data to the slot assigned from the network termination device 300 and the data 350 is transmitted to the network termination device (300).

4 is a diagram illustrating a fair band allocation model in a QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention.

Referring to FIG. 4, the fair bandwidth is allocated by the GPS scheduler 410 provided in the subscriber device 400. When the request 450 having different requirements flows into the subscriber device 400, a request that is limited by delay according to the request (for example, traffic with reference number 460) and an request that is not restricted (eg For example, traffic having a reference number of 470). Each request divided according to the limitation of the delay time is stored in the buffers 420 and 430 provided in the subscriber device 400, respectively. In addition, each request is collected by the queues 420-1 through 420-n and 430-1 through 430-n in the buffers 420 and 430, and information about delay constraints is collected. Requests that are not delay constrained until terminated remain in a waiting state. At this time, the DEB-GPS scheduler 410 performs a schedule operation according to the queue lengths and weights of the queues 420-1 to 420-n and 430-1 to 430-n. The DEB-GPS scheduler 410 fairly allocates bandwidth for each stream by the EB-GPS scheduling technique. In this case, the allocated bandwidth satisfies the constraints of the delay time and the delay variation according to the guarantee of the bandwidth for the request 460 for which the QoS satisfying the constraints of the delay constraint and the delay variation is not required. Fairly allocated bandwidth to service by maximizing bandwidth usage of requests 470.

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

Although the preferred embodiments of the present invention have been shown and described above, the present invention is not limited to the specific preferred embodiments described above, and the present invention belongs to the present invention without departing from the gist of the present invention as claimed in the claims. Various modifications can be made by those skilled in the art, and such changes are within the scope of the claims.

According to the QoS guaranteed scheduling system and method in an Ethernet passive optical network according to the present invention, by using dual DEB-GPS scheduling to ensure the limited delay time for traffic requiring quality assurance, there is no loss of traffic, the best service It is possible to guarantee the quality of service while making the most efficient use of the bandwidth for.

1 is a diagram illustrating a configuration of a QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention;

2 is a flowchart illustrating a process of performing a QoS guaranteed scheduling method in an Ethernet passive optical network according to the present invention;

3 is a diagram illustrating uplink transmission from a subscriber device configuring a QoS guaranteed scheduling system to an Ethernet termination device in an Ethernet passive optical network according to the present invention;

4 is a diagram illustrating a fair band allocation model in a QoS guaranteed scheduling system in an Ethernet passive optical network according to the present invention.

Claims (21)

A first buffer in which a first traffic requiring a guarantee of a quality of service (QoS) is stored; A second buffer in which a second traffic to which a priority transmission order is given is stored; The queue length information of the first buffer and the second buffer is detected and transmitted to the bandwidth allocation apparatus according to the inflow of the first traffic and the second traffic, and the first buffer and the first buffer according to the bandwidth allocated from the bandwidth allocation apparatus. A scheduler for allocating bandwidth for each traffic by calculating an expected service time based on an inflow amount and a length of traffic flowing into a second buffer; And And a mux for multiplexing the traffic input through the scheduler and transmitting the multiplexed traffic to the bandwidth allocating device. The method of claim 1, And a switch fabric for selectively outputting the traffic to the first buffer and the second buffer according to the characteristics of the incoming traffic. The method of claim 1, The scheduler, A calculation unit configured to calculate a service time estimated by arrival and length of each traffic based on the type of incoming traffic and the allocated bandwidth, and then calculate a variation in delay time between traffic for fair bandwidth allocation between traffics; A verification unit for comparing the calculated delay time variation range and the delay time variation tolerance to determine whether the delay variation that is limited to the bandwidth allocated from the bandwidth allocation apparatus is satisfied; And An allocation unit for allocating bandwidth for each traffic by calculating an expected service time based on the inflow and length of traffic flowing into the first and second buffers when the allocated bandwidth satisfies the limited delay variation QoS-guaranteed scheduling system in an Ethernet passive optical network, comprising: a. delete The method of claim 3, wherein And the scheduler resets or adjusts QoS requirements for the traffic if the allocated bandwidth does not satisfy the constrained delay variation. The method of claim 1, The scheduler is a QoS-guaranteed scheduling system in an Ethernet passive optical network, characterized in that the DEB-GPS scheduler performs a scheduling operation by a DEB-GPS (Deterministic Effective Bandwidth-Generalized Process Sharing) scheme. A first traffic requiring QoS received from each of the connected subscriber apparatuses, a transmission rate required for each of the first traffics, and a queue length information, each of which has a plurality of memory areas; 1 storage unit; Each of the memory areas has a plurality of memory areas, and each of the memory areas has a second traffic given priority transmission received from each of the subscriber apparatuses connected thereto, a transmission rate required for each of the second traffics, and queue length information. The second storage unit is stored; And Allocate bandwidth first for the first traffic based on queue length and traffic information for traffic requests received from at least one connected subscriber device, and allocate for the first traffic at allocable bandwidth And a scheduler for allocating bandwidth for each of the second traffics by dividing the divided bandwidth by the number of the second traffics. The method of claim 7, wherein A synchronization unit for transmitting and receiving a message for operating and managing a physical layer periodically with the subscriber device to synchronize time with each other; And And a device manager for receiving slot allocation request and buffer status information from the subscriber device and assigning information of the slot to be allocated and a permission message to each of the subscriber devices. Scheduling System. The method of claim 7, wherein And a weight buffer for storing a first weight value according to the QoS class for the first traffic and a second weight value according to the characteristics of the second traffic. The scheduler assigns the first weight to the first traffic and the second weight to the second traffic to allocate bandwidth for the first traffic and the second traffic. QoS Guaranteed Scheduling System in Network. The method of claim 7, wherein The scheduler, A confirmation unit for confirming whether a request for traffic input from each of the subscriber devices satisfies a limited delay time; And And an allocating unit for allocating bandwidth to each of the subscriber stations if the traffic request satisfies the limited delay time. The method of claim 10, And if the request of the traffic does not satisfy the limited delay time, request the subscriber station to reset or adjust the requirement. (a) receiving queue length and traffic information for traffic requests from at least one connected subscriber device; (b) allocating bandwidth to a first traffic requiring QoS based on the traffic information; (c) allocating bandwidth for each second traffic by dividing the bandwidth subtracted from the bandwidth allocated for the first traffic from the total allocable bandwidth by the number of second traffic to which priority transmission is given; And (d) informing the subscriber station of the bandwidth allocation for the first and second traffic; QoS guaranteed scheduling in an Ethernet passive optical network. The method of claim 12, Before step (a), Synchronizing time with each other by periodically transmitting and receiving messages for operation and management of a physical layer with the subscriber device; And Receiving slot allocation request and buffer status information from the subscriber device, and assigning the information and the allow message of the slot to be allocated to each of the subscriber device; QoS-guaranteed scheduling in the Ethernet passive optical network further comprising Way. The method of claim 12, (e) calculating, by the subscriber device, an expected service time based on the inflow and the length of the first traffic and the second traffic, and allocating a bandwidth for each traffic; QoS Guaranteed Scheduling in Passive Optical Networks. The method of claim 14, In step (e), (e1) After calculating the service time expected by the arrival and length of each traffic based on the type of traffic flowing into the subscriber device and the bandwidth allocated in the step (d), Calculating a variation in delay time between traffics; (e2) comparing the calculated delay time variation with the delay time tolerance to determine whether the delay variation limited to the bandwidth allocated in step (e) is satisfied; And (e3) When the bandwidth allocated in the step (d) satisfies the limited delay variation, the inflow amount of the first traffic requiring the guarantee of the QoS flowing into the subscriber device and the second traffic given the priority transmission order And allocating bandwidth for each traffic by calculating an expected service time based on a length and a length. delete The method of claim 15, In the step (e3), if the allocated bandwidth does not satisfy the constrained delay variation, the QoS guaranteed scheduling method of the Ethernet passive optical network, characterized in that to reset or adjust the QoS requirements for traffic. The method of claim 12, The step (a) includes a step of checking whether a request for traffic input from each of the subscriber apparatuses satisfies a limited delay time. Step (b) to step (d) is performed when the traffic request satisfies the constrained delay time. The method of claim 18, If the traffic request does not satisfy the constrained delay time, requesting the subscriber device to reset or adjust the requirement. The method of claim 12, In the step (c), the first traffic is assigned to the first traffic according to the QoS class and the second traffic is assigned to the second traffic according to the characteristics of the traffic, thereby providing bandwidth for the first traffic and the second traffic. QoS-guaranteed scheduling method in an Ethernet passive optical network characterized by allocating. (a) receiving queue length and traffic information for traffic requests from at least one connected subscriber device; (b) allocating bandwidth to a first traffic requiring QoS based on the traffic information; (c) allocating bandwidth for each second traffic by dividing the bandwidth subtracted from the bandwidth allocated for the first traffic from the total allocable bandwidth by the number of second traffic to which priority transmission is given; And (d) informing the subscriber device of bandwidth allocation for the first and second traffic; and executing a program for executing the QoS guaranteed scheduling method in an Ethernet passive optical network in a computer. Recordable computer-readable recording media.