KR960016538B1 - Fully distributed atm switching system and congestion by bandwidth allocation - Google Patents

Abstract

an access switching module(1) comprising an access switch network module(ASNM)(8), a call processor part(CPP)(2), an interface part(IP)(4), a subscriber call processor(SCP)(5), a trunk call processor(TCP)(6), a broadcasting call processor(BCP)(7), an access switch module(ASM)(8), a link interface module(LIM)(9) and a signalling handler processor(SHP)(3); and a central interconnection module(CIM)(10) comprising an operation and maintenance processor(OMP)(11), a network synchronization processor(NSP)(12) and an interconnection switch network module(ISNM)(13).

Description

Fully Distributed Asynchronous Transfer Mode Switching System and Congestion Control Method by Bandwidth Allocation

1 is a schematic diagram of a fully distributed ATM switching system according to the present invention;

2 is a block diagram of a call control block in a fully distributed ATM switching system according to the present invention;

3 is a flow chart of preventive congestion control at the entrance of the ATM switching system according to the present invention;

4 is a congestion control flow chart using band allocation of the ATM switching system according to the present invention,

5 is a flowchart of call establishment by path and band allocation in an ATM switching system according to the present invention.

* Explanation of symbols for main parts of the drawings

14: System Call Data Control (SCDC)

15: Number Translation Control Block (NTC)

16: Self-Routing Switch Control (SRSC)

17: Call Admission Control (CAC)

18: Subseriber Call Control (SCC)

19: Trunk Service Control Block (TSC)

20: Multi-Connection Service Control Block (MCSC)

21: Multi-Party Service Control (MPSC)

22: Global Announcement Service Control (GASC)

23: Global Multi-Connection Service Control Block (GMCSC)

24: Global Multi-Party Service Control (GMPSC)

25: Global Connectionless Service Control Block (GCLSC)

26: ConGestion Control (CGC)

The present invention is to control the congestion in order to continuously and stably provide the service required from the subscriber station of the traffic control function to be performed in each node of the asynchronous transfer mode (hereinafter referred to as 'ATM') switching system The present invention relates to a fully distributed ATM switching system and a congestion control method using bandwidth allocation.

ATM technology exchanges communication information with various speeds and various attributes by 53 octets of cell units, and is networked by virtual circuits based on logical channel numbers assigned to links between nodes during call setup through statistical multiplexing. It operates at line speed of more than 155.52Mb / s by exchanging data at high speed.

Therefore, it is desirable to design the system configuration to facilitate the structure and control while satisfying various communication requirements.

Services provided in ATM environment show various traffic characteristics such as maximum generation rate and average generation rate of information, burst index, average length of burst interval, etc., so ATM network can satisfy network service efficiency and service demand quality. There is a need for highly effective traffic control techniques such as predicting traffic changes and preventing congestion.

Therefore, the robust control of congestion control function to prevent the congestion and minimize the effect of congestion in case of congestion by performing traffic control functions such as call admission control and priority control for the flow of call and information continuously from the entrance of the network to the completion of the call. It needs to be done.

By the way, the existing call processing technology mainly used to control voice and low speed data in response to the time when congestion occurs in the network. In addition, there is a demand for a communication network in which high-speed communication services including dynamic video transmission services are provided. In order to satisfy the demands of such communication services, ATM exchange technology for processing various traffic types has been studied in Korea as well as in advanced countries. In the process.

The present invention devised to meet the above requirements is a complete and reliable service for ATM switching system under research / development to provide high-speed communication service including high-speed data communication, still picture, dynamic image and so on. By configuring the system in a decentralized manner, it is possible to prevent congestion and control effectively when congestion occurs and to prevent and control the congestion state of the switching system in order to satisfy the service demand of the user. The purpose is to provide a congestion control method.

In order to achieve the above object, a device of the present invention comprises an access switch network module (ASNM), a call processing unit (CPP) for user call processing, a subscriber matching module (SIM), and a trunk line matching module (TIM). Subscriber call processing unit (SCP) and relay line call processing unit (TCP) to control one-to-one service and broadcasting call processing unit (BCP) to control call connection for providing multi-connection related (broadcast) service Link matching module between access exchange module (ASM) and central internal connection module (CIM), dedicated link cell module (LIM) that provides IMI protocol, dedicated network matching, user network matching / network node matching (UNI / Terminating a signal information cell on the NNI protocol, and a signal processing unit (SHP) for processing signal information together with the subscriber call processor (SCP), the trunk line call processor (TCP), and the broadcasting call processor (BCP). Access exchanging modules (ASM) to perform; An operation maintenance processor (OMP) for performing fault detection and recovery of the system, a switch for transferring information to the access exchange module (ASM), a network synchronizer processor (NSP) for performing time management and system synchronization management of the system, A blockless, self-routing switch consisting of a single piece of information in a switch network and having a central internal connection module (CIM) including an internal network module (ISNM) used as an internal interconnector with an access switch network module (ASNM). do.

In addition, the system call data control unit (SCDC), number translation control unit (NTC), magnetic routing switch control unit (SRSC), rock lock control unit (CAC), subscriber call control unit (SCC), relay line service control unit (TSC), multiple connection service control unit (MCSC), multi-party service control unit (MPSC), global guidance service control unit (GASC), global multi-connection service control unit (GMCSC), global multi-party service control unit (GMPSC), global connectionless service control unit (GCLSC), congestion control unit ( In the congestion control method using bandwidth allocation applied to an asynchronous delivery mode switching system having a CGC), when the system call data controller (SCDC) receives a call control request, the received signal data is compared with the system call control data. A first step of determining whether the traffic characteristics match; After performing the first step, if the reported traffic characteristics match, the corresponding path connection procedure is performed. If the reported traffic characteristics are different, the transmission of the cell into the network is delayed to match the buffering at the network entrance. A second step of checking whether the traffic characteristics match; After performing the second step, if the characteristics do not match, discarding the cell to suppress transmission to the network, and if the matching traffic characteristics match the third step of performing a corresponding path connection procedure; When allocating the usage band for user service request in the asynchronous forwarding mode switching system for the routed call, the traffic characteristics are required by dividing the initial band, and the service priority is high or constant rate traffic band ( A fourth step of accepting a connection if possible at a threshold of a variable speed traffic band); A fifth step in which the service priority is not high or is a constant speed traffic band (variable speed traffic band), and if it is not within the threshold, rejects the call connection and reallocates or uses the virtual path or switches the congestion; When the user requests the call, the call processing processor performs the connection admission control, requests the route allocation to the lock control section, and the lock control section considers the status of the virtual path and the virtual channels to satisfy the service request. A sixth step of allocating a virtual path and a virtual channel to the calling party and the called party and notifying the call processor of the calling party and the called party; And attach routing information of the switch network through an internal signaling channel with an originating call side access exchange module (ASM) in the switching system, and the call processing processor of the originating call side notifies the virtual path and the virtual channel, and routes the switch network. After attaching the information, a seventh step of updating the input / output header conversion table and transmitting the information is included.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention;

1 is a schematic configuration diagram of a fully distributed ATM switching system according to the present invention, which basically performs a connection function between an Access Switching Module (ASM) and an Access Switching Module (ASM) that perform an aggregation function. It has a modular structure composed of a central interconnection module (CIM).

The access exchange module (ASM) can operate independently with a minimum system configuration, and in large systems, the multiple access exchange modules (ASM) operate as concentrators having a constant current collecting ratio, and the central interconnection module (CIM) has a large capacity. As a distribution device for the system, it operates for the maintenance of switching functions and system state management.

The access exchange module (ASM) 1 includes a single stage N × N magnetic routing switch module, and includes a call processor part (CPP) 2 and a signal processor for user call processing. (SHP: Signaling Handler Processor) (3) and a matching interface (IP: Interface Part) (4) consisting of a Subscriber Interface Module (SIM) and a Trunk Interface Module (TIM).

The access exchange module (ASM), which is a concentrator, is a non-blocking magnetic routing switch that performs a concentrator function with a constant concentrating ratio according to the system configuration, and the one-to-one service is SCP (5) / TCP (6). Under the control of ASM, Access Service Network Module (ASNM) (8) and User Network Interface (UNI) are connected under the control of the Access Switching Module (ASM). Terminating the signal information cell on the NNI (Network Node Interface) protocol and processing the signal information along with SCP, TCP, and BCP, and a general subscriber using the user network matching (UNI) protocol. As a processor that performs call processing of subscribers, subscriber call processing unit (SCP: Subscriber C) performs overall traffic control such as call admission control, user / network parameter control, priority control, and congestion control with subscriber matching circuit. all Processor), a processor that performs call processing with a network using the network node matching (NNI) protocol, and performs call connection control for incoming / outgoing calls with a trunk line matching circuit. Trunk Call Processor (TCP) that manages the functions required for network matching, Broadcasting Call Processor (BCP) that controls call access for providing multiple access-related (broadcast) services, and access exchange within the system Link matching module between ASM and central interconnection module (CIM). It is composed of link interface module (LIM) (9) that provides IMI (Inter Module Interface) protocol dedicated to internal cell format. do.

The central interconnection module (CIM) 10 includes an operation and maintenance processor (OMP) 11 for detecting and recovering a failure of a switch and a system for transmitting information to an access exchange module (ASM), Network Synchronization Processor (NSP) (12) that performs time management and system synchronization management of the system, a switch network of a central internal connection module (CIM). It consists of an Interconneciton Switch Network Module (ISNM) 13 which is used as an internal connector with the network module ASNM.

2 is a block diagram of a call control block in a fully distributed ATM switching system according to the present invention, which is completely distributed to an access exchange module (ASM) in order to maintain stable service with minimal impact on subscribers in case of system failure. A flowchart of a block for configured call control is shown.

When a call setup request is received from a user terminal, meta signaling is processed from a matching function, and a message is transmitted to a system call data control block (SCDC) 14 of call control.

The SCDC block 14 compares and analyzes the data of each user group with the data of the system, and then sends a message to the Number Translation Control (NTC) 15 for number translation. The NTC block 15 Calling / calling number translation is performed to notify the number translation result to the Self-Routing Switch Control (SRSC) 16.

A magnetic routing switch control block (SRSC) 16 analyzes the switch path between incoming / outgoing calls and the input / output header data in the subscriber matching module (SIM) or the trunk line matching module (TIM), thereby providing a lock lock control block (CAC) 17. In case of own station, it is notified of the processing by subscriber call control block (SCC) (18) in case of own country and trunk service control (TSC) (19) in case of other station, and in case of multiple or broadcast type Multi-Connecter Service Control Block (MCSC) 20, Multi-Party Service Control Block (MPSC) 21, Global Announcement Service Control (GASC) 22, Global Multi-Connection Service Control Block (GMCSC) (23), Global Multi-Party Service Control (GMPSC) (24), Global Connectionless Service Control Block (GMPSC) GCLS C: Notifies the Global ConnectionLess Service Control (25) to allocate the corresponding resources to perform the unique function of each block.

If it becomes difficult to perform call control function due to congestion of system or resource during function of each block, Magnetic Routing Switch Control Block (SRSC), Lake Lock Control Block (CAC), Congestion Control Block (CGC) (26) This interlocking function performs a reactive congestion control function such as blocking an incoming or outgoing call according to the class of the subscriber call.

3 shows a preventive congestion control flow chart at the inlet of the network according to the invention.

When the meta-signaling is processed from the matching function (27) and the call control request comes to the system call data control block (SCDC), the SCDC block compares the system call control data with (28) and monitors the traffic occurrence status from the call source.

If the SCDC block matches the declared traffic characteristic 29, the SCDC block performs the corresponding path connection procedure (30), and if it is different from the reported traffic characteristic (31), it transmits the cell into the network to be matched by buffering at the network entrance (31). Delay (32).

If the SCDC block does not match the characteristics even after delaying the transmission (33), the cell is discarded (34) to suppress transmission into the network.

In other words, the factor of congestion is suppressed by comparing and analyzing the service class and traffic characteristics indicating the allowable cell discard rate, average transmission delay, and dispersion, which are the required quality of the call, when the call is made.

4 shows a flowchart for band allocation for a path for preventive congestion control according to the present invention.

After the routing is completed, there are two methods for allocating a bandwidth used for a user's service request in an ATM switching system, by using a request peak value allocation method and a band allocation method using a statistical multiplexing method. Since the amount of bandwidth of the connection path between the called party and the state of the connection node must be accurately and promptly processed, it is necessary to manage the entire network resources and nodes. It is classified as speed traffic and applied whether to accept or reject connection when call establishment request is applied using statistical multiplexing technique.

First, call control is performed by dividing the total bands accommodated in the network into constant speed traffic bands (C _c ) and variable speed traffic bands (C _v ) according to the characteristics of each traffic by using statistics on ATM traffic. In order to prevent the allocation of a predetermined margin (C _p ) (35).

The B-ISDN terminals report the traffic characteristics to the network using the virtual path and the virtual channel through the meta-signaling procedure (36).

The network determines whether the type of traffic is constant speed traffic or variable speed traffic by using the reported traffic characteristic (37).

In the case of constant / variable speed traffic, it is determined whether there are enough resources in the network within the constant / variable speed band (C _c / C _v ) that the quality of communication required by the parameter declared by the QOS is guaranteed. The network determines (38), if there is a free band, accepts the connection (43), and if not, performs band extension analysis through the free band C _{p in the} network (39).

By analyzing the band scalability (40), if possible, accept the connection (43), otherwise, compare and analyze the priority (41), and if the service is high priority, allocate it within the free band (42). Accept (43).

In the case of a service having a low priority, the connection is refused to prevent congestion (44), but a cumulative number of times that a service having a high priority cannot be connected or a service having a low priority is rejected (45) is calculated. If the connection is rejected over a certain threshold (46), the responsive congestion control function is performed.

In a runaway state, service degradation occurs, so it is required to execute the service as soon as possible without causing service degradation to the user. As this control, measures to be taken in accordance with the relocation of resources are considered.

In order to reduce the effect on the user terminal, the virtual path (VP) capacity is expanded to analyze the link in which the corresponding virtual path (VP) is accommodated (47). If there is, reassign the virtual path (VP) and use it (48), if not, analyze the capacity of the virtual path (VP) (49) and check the virtual path where the congestion is detected (50). Several virtual channels VC in VP) are transferred to the virtual path VP in which no other congestion occurs.

In particular, instantaneous cut transitions such that the links are not cut should be made technically.

5 shows a flowchart for setting up a band-allocated call for preventive congestion control according to the present invention.

In an ATM switching system, a signal cell is converted into an internal cell form in which additional information for routing is added in order to switch it to a called party according to the signal information.

When the user requests a call, the call processing processor (CPP) performs the connection admission control, and requests the path allocation to the rock lock control block (CAC) (52).

The rock lock control block (CAC) allocates a virtual path and a virtual channel for the calling party and the called party capable of satisfying the service request in consideration of the status of the virtual path and the virtual channels (53). The call processing processor (CPP) is notified (54, 55).

The routing information of the switch network is attached through the internal signaling channel with the called party access exchange module (ASM) in the switching system (56), and the call processing processor (CPP) of the calling party sends an input virtual path identifier / input virtual channel identifier ( Update the input header translation table for VPI ₁ / VCI ₁ (57), and the output call processing processor (CPP) updates the output header translation table with the output virtual path identifier / output virtual channel identifier (VPI _o / VCI _o ). (58).

Cells having an input virtual channel identifier (VCI ₁ ) input from a user are converted into an internal cell structure having routing information by an input header conversion table, and then reach a destination side aggregation end according to routing information. The output header conversion table converts the foreign cell structure with the output VPI _o / VCI _o to transmit information to the network (59) to set up the call.

During the call setup process, the network will continue to perform the preventive congestion control of FIG. 3 to determine whether to accept / reject a newly generated call request or not to have enough idle bandwidth to service a new call 60 on the connection path for this call. The bandwidth required by the new call should be calculated by the Lake Lock Control Block (CAC), taking into account the current and ongoing control features.

In addition, to satisfy the service, data for each user group is statistically analyzed by the system call data control block (SCDC), which is a call processing block, and the establishment and destruction of links between stations, the number of relay lines on each link, the bandwidth of each relay line, etc. To efficiently control the runaway in the network, and control the runaway in conjunction with the runaway control block (CGC) when runaway occurs.

As described above, the present invention prevents the system failure due to the congestion from the entrance of the network, prevents the congestion caused by the effective use of the band for the improvement of service, the effective performance of the system performance and traffic control function Can contribute to

Claims (5)

Access Switch Network Module (ASNM) (8), Call Processor Part (CPP) (2) for User Call Processing, Subscriber Interface Module (SIM) and Relay Line Matching Module (SIM) Interface part (IP) (4) consisting of TIM (Trunk Interface Module), Subscriber Call Processor (SCP) (5) and Trunk Call Processor (TCP) which controls one-to-one service (TCP) 6) Broadcasting Call Processor (BCP) (7), Access Switch Module (ASM) and Central Interconnect Module in system to control call connection for providing multiple access related (broadcast) service Link interface module (CIM: Central Interconnection Modlue), which is dedicated to internal cell format, provides link interface module (LIM) (9) that provides IMI (Inter Module Interface) protocol, and user network matching / network Node Matching (UNI / NNI) A signal processing unit for terminating the signal information cell on the protocol, and processing signal information together with the subscriber call processing unit (SCP) 5, the trunk line call processing unit (TCP) 6, and the broadcasting call processing unit (BCP) 7; An access switching module (ASM) 1 having a signaling handler handler (SHP) 3 to perform an aggregation function; Operation and Maintenance Processor (OMP) 11 which performs fault detection and recovery of a switch and a system for information transfer with the Access Exchange Module (ASM) 1, system time management and system synchronization. Network Synchronization Processor (NSP) 12, which performs management, is a non-blocking, self-routing switch composed of a single piece of information on a switch network, and is used as an internal interconnector with an access switch network module (ASNM). A fully distributed asynchronous delivery mode switching system having a Central Interconnection Module (CIM) 10 including an Interconneciton Switch Network Module (ISNM) 13. The system of claim 1, wherein the access exchange module (ASM) 1 processes meta signaling from a matching function when a call establishment request is received from a user terminal, and inputs the data and system of each user group. A system call data control (SCDC) 14 for comparing and analyzing data of the data; A number translation control unit (NTC) 15 for number translation of the data analyzed by the system call data control unit (SCDC) 14; Self Routing Switch Controller (SRSC: Self), which is informed of the result of the number translation of the originating / receiving number from the number translation control unit (NTC) 15 and analyzes the input / receive switch path and the input / output header data in the subscriber matching module or the trunk line matching module. Routing Switch Control 16; Call Admission Control (CAC) 17, which is a band allocation block; A subscriber call control unit (SCC) 18 which is allocated a band from the rockacy control unit 17 and processes it when it is a local station; A trunk line service control unit (TSC: Trunk Service Control) (19), which is allocated a band from the lake control unit (CAC) 17 and processes when it is a foreign station; Multi-Connection Service Control (MCSC) 20 to perform a unique function of each block by allocating a band and a corresponding resource in the case of a multiple or broadcast type from the lake control unit (CAC) 17, a multi-party Multi-Party Service Control (MPSC) 21, Global Announcement Service Control (GASC) 22, Global Multi-Connection Service Control (GMCSC) 23, Global Multi-Party Service Control (GMPSC) 24, Global ConnectionLess Service Control (GCLSC) 25; And if the call is difficult to perform due to the congestion of the system or the resource during the execution of the function of each block, the self-routing switch controller (SRSC) 16 and the rock lock controller (CAC) 17 interlock with the subscriber call. Fully distributed asynchronous delivery mode switching system comprising a congestion control (CGC) (26) to perform a responsive congestion control function to block the incoming or outgoing call according to the grade. System call data control unit (SCDC) 14, number translation control unit (NTC) 15, magnetic routing switch control unit (SRSC) 16, rock lock control unit (CAC) 17, subscriber call control unit (SCC) 18 ), Trunk line service control unit (TSC) 19, multi-connection service control unit (MCSC) 20, multi-party service control unit (MPSC) 21, global guide service control unit (GASC) 22, global multi-connection service control unit (GMCSC) 23, global multi-party service control unit (GMPSC) 24, global connectionless service control unit (GCLSC) 25, congestion control unit (CGC) with a bandwidth allocation applied to the asynchronous delivery mode switching system (CGC) A congestion control method comprising: a first step of, when a system call data control unit (SCDC) 1 receives a call control request, compares the received signal data with system call control data to determine whether traffic characteristics match; After performing the first step, if it matches the reported traffic characteristics, the corresponding path connection procedure is performed, and if it is different from the reported traffic characteristics, the transmission of the cell to the network is delayed to match by buffering at the network entrance, and then again. A second step of checking whether the traffic characteristics match; After performing the second step, if the characteristics do not match, discarding the cell to suppress transmission to the network, and if the matching traffic characteristics match the third step of performing a corresponding path connection procedure; In case of allocating the use band for user service request in the asynchronous forwarding mode switching system for the routed call, the traffic characteristics are required by dividing the initial band, and the service priority is high or constant rate traffic band ( A fourth step of accepting a connection if possible at a threshold of a variable speed traffic band); A fifth step in which the service priority is not high or is a constant speed traffic band (variable speed traffic band), and if it is not within the threshold, rejects the call connection and reallocates or uses the virtual path or switches the congestion; When the user requests the call, the call processing processor performs the connection admission control, requests the route allocation to the lock control section, and the lock control section considers the status of the virtual path and the virtual channels to satisfy the service request. A sixth step of allocating a virtual path and a virtual channel to the calling party and the called party and notifying the call processor of the calling party and the called party; And attach routing information of the switch network through an internal signaling channel with an originating call side access exchange module (ASM) in the switching system, and the call processing processor of the originating call side notifies the virtual path and the virtual channel, and routes the switch network. And a seventh step of updating the I / O header conversion table after attaching the information and transmitting the information. 4. The method of claim 3, wherein the fourth step comprises: classifying whether the traffic type is a constant speed by requesting traffic characteristics after initial band division; Investigating whether quality can be guaranteed in a constant rate traffic band (variable rate traffic) at a constant rate; If possible, waits after accepting the connection, or extends and analyzes the spare band, and if possible, waits after accepting the connection and investigates whether the service priority is high; Congestion control method according to the band allocation characterized in that it comprises the step of accepting the connection, if possible at a threshold of a high service priority or a constant speed traffic band (variable speed traffic band). 4. The method of claim 3, wherein the fifth step is to refuse connection in order to prevent congestion when the service is not a high priority, but the service having a high priority cannot be connected or a service having a low priority is also rejected. Calculating a cumulative number of times and performing reactive congestion control when the connection is rejected above a certain threshold; Analyzing the link that contains the virtual path and reallocating the virtual path if there is free capacity on the link, or analyzing the capacity of the virtual path to switch to the free virtual path and waiting. Congestion control method by band allocation, characterized in that made.