JP5469136B2 - Call control method - Google Patents

Description

  The present invention relates to a technique for controlling a call.

  Currently, it is equivalent to the telephone service provided by the existing telephone exchange network by the IP communication network system composed of the transfer network consisting of call processing servers and transfer devices and the access network consisting of subscriber terminals and home gateways. Service can be provided.

  There are various types of call processing servers depending on the application. In addition to subscriber-type call processing servers that perform services using information on subscribers within a certain accommodation range, a plurality of subscriber-type call processing servers can be used. There is an AS (Application Server) that provides a common nationwide service. As a typical service example of AS, there are a number conversion system service such as toll-free number, and a guidance service that sends a specified message according to the state of an incoming user.

  Here, the processing operation until service provision will be described with reference to FIG. When a logical telephone number is inputted to the subscriber terminal 1a of the access network 10a and a call is made (for example, SIP call), the call signal (for example, SIP invite signal) is branched by the transfer device 3a. Then, it is transmitted to the call processing server 5a of the subscriber system, and a predetermined process using the subscriber information is executed.

  If the destination is a logical number, it is transmitted to the AS 100 and converted into a physical telephone number as a destination, and then returned to the sender's subscriber call processing server 5a or becomes a destination. It is transmitted to the subscriber system call processing server 5b that accommodates the physical number. At this time, depending on the configuration and number band of the IP communication network, the message is transmitted to the destination subscriber call processing server 5b via the relay call processing server 5c.

  Thereafter, acceptable conditions are exchanged via the above-described route between the calling subscriber terminal 1a and the called subscriber terminal 1b (SIP negotiation), and media communication is performed based on the established conditions. . The media communication here is generally a call.

  A signal used for media communication (for example, a packet in an IP communication network) is transmitted / received via the transfer device 3 without passing through the call processing server 5 or the AS 100. In addition, the guidance voice of guidance communication is transmitted via the guidance server 7.

  Here, since such AS is intended for processing nationwide calls, a large number of signals are simultaneously transmitted from a plurality of subscriber call processing servers, and therefore the processing frequency is extremely high.

  In addition, since high availability is required so that high-frequency call processing can be continuously provided without interruption, the service is switched from the ACT (active system) to the SBY (standby system) when a failure occurs. The redundant configuration of ACT-SBY is used.

  Specifically, cluster management technology that conforms to SAF (Service Availability Forum) specifications that define high availability, continuity, and portability, and a function called CKPT (checkpoint) can be used. 8).

  The CKPT function restores call information using data managed in the CKPT section area when the process is recovered by restart or failover, and restarts from the state before the failure to minimize the impact of the failure. Technology. On the program, by executing the CKPT process, the processing is executed and the function is realized.

  The maximum size of the management data area in the CKPT is the number of sections × the section size. If it is within the size, it can be used as appropriate by a use program for the CKPT function. The CKPT process and its management data area are synchronized between the systems between ACT and SBY.

Japanese Patent Laid-Open No. 2002-300272

  When calls are concentrated on a specific number on a system equipped with such an AS, the load on the subscriber / relay-system call processing server or PBX that concentrates the specific number behind the AS is concentrated. There was a problem. This problem will be described below with reference to FIG.

  When calling from the subscriber terminal # 1, the call signal is transmitted to the call processing server # 1 of the subscriber / relay system on the caller side via the access network or core network on the caller side, and based on the subscriber information Processing is executed.

  At that time, if the destination is a logical number, it is sent to the AS, converted into a physical number to be the destination, and then sent to the call processing server # 10 of the receiving side subscriber system / relay system, and PBX It is transmitted to # 1 and home gateway # 1.

  In PBX # 1 and home gateway # 1, since the number of lines that can be received is determined in advance, for example, when a representative number is set, calls equal to or more than the subscriber number associated therewith cannot be accepted.

  In that case, a busy response indicating that there is no free line is not received for a new call from the subscriber terminal # 2, and a busy response indicating that the call is congested from the PBX # 1 to the call processing server # 10 of the subscriber / relay system on the receiving side Will be replied to. Based on the busy response, the call processing server # 2 and the subscriber terminal # 2 of the AS or the originating side subscriber / relay system can determine that the incoming side is incapable of answering / not answering.

  As call concentration continues, the amount of busy responses also increases. However, in this example, calls are concentrated on PBX # 1, and PBX # 2 can accept new calls.

  On the other hand, the AS processing capacity is also limited, and if the processing capacity is allocated to the call load concentrated on PBX # 1 returning a busy response, calls to other call processing servers # 11 and PBX # 2 There is a possibility that the processing will be affected and the processing may become impossible or impossible. Therefore, it is necessary to suppress a call for a PBX (PBX # 1 in the above case) that has returned a busy response and has insufficient response capability.

  According to Patent Document 1, when a callee number in a congested state is received from a subscriber terminal, a pseudo call is periodically repeated on the called side, and the call is registered in a queue until connection is established by the pseudo call. The technology to keep is disclosed. However, since the pseudo signal is repeatedly transmitted, the above-described problem cannot be solved.

  The present invention has been made in view of the above problems, and a first problem is to suppress a signal to a subsequent apparatus that is concentrated in a load without affecting the main flow of call processing. A second problem is to generate list data used at the time of regulation with a size that fits in the size of the management data area of the synchronized CKPT function.

  The call control method according to claim 1 is a call control method performed by a call control device, wherein a response signal to a call signal is received from a called side and held in a totaling queue different from a queue used for call processing. A holding step, a measurement step of taking out a signal from the aggregation queue, generating a count list in which a predetermined measurement number for the signal is counted for each destination, and periodically copying the count list A generation step for generating a plurality of count lists having the number of measurements at the regular timing, a count step for calculating the number of variable measurements per unit time using two count lists, and the unit A first determination step for determining that a destination having a measured number of fluctuations per hour exceeds a predetermined threshold is a destination having a problem; A storage step of storing as a rejection number list in the management section area of the checkpoint function, a second determination step of determining whether or not a destination of a newly received call signal is included in the rejection number list, and a rejection number A restriction step of restricting a call signal determined to be included in the list is executed separately from the call processing.

  According to the present invention, a process for receiving a response signal to a call signal from the incoming side and holding it in a totaling queue different from the queue used for call processing and generating a rejection number list is called call processing. Since it is executed separately, determination and restriction processing can be performed independently of the main flow of call processing, so the signal to the subsequent device that is concentrated in the load without affecting the main flow of the call processing. Can be suppressed.

  A call control method according to a second aspect is the call control method according to the first aspect, wherein the counting step repeats the calculation of the variation measurement number at intervals shorter than the unit time.

  According to the present invention, since the calculation of the number of fluctuation measurements in the counting step is repeated at intervals shorter than the unit time, the determination target periods overlap, so that there is a concentration of call connections that span two determination target periods. Even in such a case, the data can be reliably counted and included in the determination target.

  The call control method according to claim 3 is the call control method according to claim 1 or 2, wherein after the number of fluctuation measurements per unit time is calculated, the call control method is generated first of the two aggregation lists used for the calculation. A deletion step of deleting the total list, and the first determination step calculates the determination using a data area of the deleted total list.

  According to the present invention, the previously generated aggregation list is deleted from the two aggregation lists used for the calculation of the variation measurement number, and the determination in the first determination step is performed using the data area of the deleted aggregation list. Since the calculation is performed, a small limited resource can be effectively used, and the apparatus itself can be reduced in size.

  A call control method according to a fourth aspect of the present invention is the call control method according to any one of the first to third aspects, wherein the totaling step includes two counts every 60 seconds out of a plurality of total lists generated every 30 seconds. It is characterized in that the number of fluctuation measurements per unit time is obtained by calculating the difference in the number of measurements included in the aggregation list.

  The call control method according to claim 5 is the call control method according to any one of claims 1 to 3, wherein the measurement step resets the measurement number of the count list every time the aggregation list is generated. The counting step obtains the number of fluctuation measurements per unit time by adding the number of measurements included in two total lists generated within 60 seconds among a plurality of total lists generated every 30 seconds. It is characterized by that.

  The call control method according to claim 6 is the call control method according to any one of claims 1 to 5, wherein the storing step includes only the destinations limited to a size that can be stored in the management section area. It is memorized in a rejection number list.

  According to the present invention, since only the destinations limited in size that can be stored in the management section area of the checkpoint function are stored in the reject number list, the size of the management data area of the CKPT function synchronized between the systems is increased. List data to be used at the time of regulation can be generated with a size that fits.

  The call control method according to claim 7, further comprising a synchronization step of transferring the reject number list stored in the management section area to the standby system in the call control method according to any one of claims 1 to 6. It is characterized by.

  The call control method according to claim 8 is the call control method according to any one of claims 1 to 7, wherein the predetermined number of measurements includes the number of simultaneous signal connections, the number of busy responses indicating congestion, and the opposite device in the subsequent stage. Any one or more of the number of errors from

  The call control method according to claim 9 is the call control method according to any one of claims 1 to 8, wherein the call control device is a number conversion device for converting a logical telephone number into a physical telephone number. It is characterized by being.

  According to the present invention, it is possible to suppress a signal to a subsequent apparatus that is concentrated on a load without affecting the main flow of call processing, and to reduce the size of the management data area of the CKPT function synchronized between systems. List data to be used at the time of regulation can be generated with a size that fits.

It is a figure which shows the functional block structure of a number converter. It is a figure which shows a refusal number list production | generation processing flow. It is a figure which shows typically a rejection number list production | generation processing flow. It is a figure which shows a totaling / determination process typically. It is a figure which shows typically the processing time required for a totaling / determination process. It is a figure which shows the control process flow of a calling signal. 1 is a diagram illustrating an overall configuration of an IP communication network system. It is a figure which shows a CKPT function typically. It is a figure which shows typically the simultaneous connection state of a call.

  Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings. However, the present invention can be implemented in many different modes and should not be construed as being limited to the description of the present embodiment.

  In the present embodiment, description will be made using the number conversion device described in the background art as an example of AS. Since AS provides a nation-wide service as described above, it targets all users. Therefore, unlike subscriber call processing servers such as SSC (Subscriber Session. Control server), AS One feature is that no information (subscriber DB) is provided. Of course, the present invention is also applicable to other call processing servers (for example, SSC having subscriber information) located on a path through which a call signal and a response signal are transmitted and received.

<First Embodiment>
[Function of number converter]
FIG. 1 is a configuration diagram showing a functional block configuration of a number conversion apparatus according to the present embodiment. The number conversion device 100 has a redundant configuration of ACT (active system) -SBY (standby system), and includes a signal transmission / reception unit 11, a worker thread unit 12, a signal holding unit 13, a signal counter unit 14, and the like. The signal determination unit 15, the CKPT function unit 16, and the intersystem synchronization unit 17 are mainly configured.

  The signal transmission / reception unit 11 and the worker thread unit 12 are basic processing function units conventionally provided in the number conversion device 100 and play a role of executing a main flow of call processing performed by the number conversion device 100. .

  That is, in the signal transmission / reception unit 11, a call signal from a preceding device (for example, a caller terminal that is a call originator, a call control server on the caller side), or a subsequent device (for example, a call on the callee side) A response signal from a control server, a PBX, a receiving terminal as a destination), and the worker thread unit 12 extracts various information contained in the signal from the signal received by the signal transmitting / receiving unit 11 Then, it is determined whether or not the contract conditions match or whether or not it is registered in the subscriber information database, and a transmission signal is generated based on the determination result.

  In addition to the above, the worker thread unit 12 performs various analysis / editing / determination processes (for example, SIP signal analysis, editing to service analysis, number conversion processing, inter-system transfer, etc.). Therefore, it is necessary to suppress a signal to a subsequent apparatus having a concentrated load so as not to affect the main flow of the call processing.

  Therefore, the number conversion device 100 includes various processing function units in addition to the signal transmission / reception unit 11 and the worker thread unit 12 as described above. The function will be described in detail below.

  In the main flow of the call processing, the signal transmission / reception unit 11 receives signals from the same device (the front device and the rear device are the same) or receives signals from the guidance server shown in FIG. I also add that there is.

  In addition to the main flow processing of the call processing, the worker thread unit 12 receives a response signal to the calling signal from the incoming side, and uses all the response signals as a queue used in the main flow of the call processing. Has a function of stacking in different signal holding units (totaling queue (hereinafter referred to as queue)) 13.

  In addition, the worker thread unit 12 is determined to have a function of determining whether or not a destination of a newly received call signal is included in a later-described rejection number list, or is included in the rejection number list. It has a function of executing a predetermined restriction before transmitting a calling signal to the receiving side.

  The signal counter unit 14 obtains the stacked signals from the queue 13, and the number of simultaneous connections currently connected to the incoming physical number of the signal or the incoming physical number of the signal from the subsequent device. A count list that counts each measured physical number (for each destination), such as the number of busy responses indicating congestion transmitted and the number of system errors on the opposite device transmitted from the opposite device for the signal. It has the function to generate.

  Further, the signal counter unit 14 has a function of generating a plurality of count lists (snapshots) having the above respective measurement numbers at the regular timing by periodically copying (duplicating) the count list. Have.

  The signal determination unit 15 calculates the difference measurement number (increase / decrease number) per unit time with respect to the destination by calculating the difference using two aggregation lists among the plurality of aggregation lists generated by the signal counter unit 14. After deleting the first total list generated from the two total lists, there is a function to determine the destination that has a predetermined number of fluctuations per unit time as the destination that has a problem. doing.

  The CKPT function unit 16 has a function of storing, as a rejection number list, a destination that is determined to have a problem by the signal determination unit 15 in a management section area managed by the CKPT function unit 16.

  The intersystem synchronization unit 17 has a function of synchronizing the rejection number list stored in the management section area of the CKPT function unit 16 with SBY.

  Each processing function unit described above is provided for the sake of convenience in understanding the features of the present invention, and each process performed by a plurality of processing function units is one in terms of implementation in a program. It can also be executed by the processing function unit. For example, the CKPT function unit 16 and the intersystem synchronization unit 17 may be executed as one function on the CKPT process.

  The function of each processing function unit included in the number conversion apparatus 100 has been described above. The queue 13 can be realized by a storage unit such as a memory, and the other processing function units can be realized by a control unit such as a CPU. Each process of all the processing function units is realized by a program that executes the process.

[Operation of number converter: Processing flow until rejected number list is generated]
Next, the operation of the number conversion device 100 will be described. Here, a processing flow until the rejection number list is generated will be described with reference to FIGS. 2 and 3.

  However, in the present embodiment, since the number conversion apparatus 100 that does not have subscriber information is used as an example of AS, the following processes are executed without using the subscriber information.

  Note that the main flow of call processing by the signal transmission / reception unit 11 and the worker thread unit 12 is described in S1 to S8 in FIG. 2 for reference.

  First, in S <b> 101, all response signals received from the succeeding apparatus are accumulated in the queue 13 by the worker thread unit 12.

  As described above, the queue 13 is a totaling queue provided so as to be handled independently, unlike the call processing queue used in the main flow of call processing. Further, the determination / restriction processing described below by stacking signals in the queue 13 is executed separately from the main flow of call processing. From these, it is possible to execute a counting process, which will be described later, by a process that is minimally stored in the queue 13.

  Further, since all the response signals are accumulated, it is possible to eliminate the process of determining which one of the response signals, and the influence of call processing on the main flow can be more reliably reduced.

  The call processing ends when the received signal is written into the totaling queue 13, and after the completion of the writing, it is accumulated in the call processing queue (not using the signal of the queue 13). Normal call processing is performed using the signal.

  In step S102, the signal counter unit 14 extracts a signal from the queue 13, counts the current number of simultaneous connections to the incoming physical number of the extracted signal, and counts the simultaneous number counted as the incoming physical number. A count list in which the number of connections is associated is generated.

  For example, as shown in FIG. 4C, the number of calls existing at the time of counting is the same as the number of simultaneous connections. However, if all the calls that have been added to the determination target time (D in the figure) are new calls, the amount of connections (specific load amount ÷ 3600 × number of connections increased in the determination target period (seconds) described later (same as above) There will be a number of unique call connections consisting of A) in the figure and the number of simultaneous connections (B in the figure).

  Note that the counting target can be arbitrarily selected according to a specified condition, and the number of busy responses and the number of system errors described above can be further included depending on the condition. In this processing flow, it is assumed that the number of simultaneous connections is selected.

  The signals accumulated in the queue 13 increase and decrease in real time, and the count list is updated whenever the number of simultaneous connections increases or decreases, when a busy response is received, or when a system error is received, and is updated for the updated incoming physical number Time is given. A counter whose update time is older than the start time of the determination target period, which will be described later, can be cleared.

  Next, in S103, the counter list updated at any time is copied every 30 seconds by the signal counter unit 14, and a temporary aggregation list (snapshot every 30 seconds) used for aggregation / determination is created. Generated each time.

  In the present embodiment, a maximum of three summary lists are generated. For example, as shown in FIG. 4, when the total list A is generated at a certain time, the total list B is generated after 30 seconds have elapsed, and the total list C is generated after 30 seconds have elapsed.

  Next, in S104, the signal determination unit 15 uses the two total lists generated every 60 seconds (in the above example, the total list A and the total list C) from among the plurality of total lists generated. By calculating the difference in the number of simultaneous connections for each number, the cumulative number of simultaneous connections (increase / decrease in the number of simultaneous connections per unit time) in the determination target period (60 seconds) is obtained.

  In the case of the above example, when the aggregation list C shown in FIG. 4 is generated, the determination target period is obtained by subtracting the number of simultaneous connections aggregated in the aggregation list A from the number of simultaneous connections aggregated in the aggregation list C. The increase / decrease number of the simultaneous connection number in # 0 (unit time) is calculated.

  Next, in S105, the signal determination unit 15 deletes the previously generated aggregation list (in the above example, the aggregation list A) from the two aggregation lists used in S104.

  Henceforth, each process to S109 is calculated using the data area | region of the deleted total list. In other words, in the present embodiment, only the data areas necessary for at most three aggregation lists are used for aggregation and determination, so that determination / regulation processing can be realized with limited resources.

  Next, in S106, the signal determination unit 15 determines whether or not the increase / decrease number of the simultaneous connections per unit time exceeds the call number determination threshold. That is, if it is determined that the number exceeds, calls for that physical number are determined to be in a concentrated state.

  In the case where the count target is the number of busy responses, the call concentration state is determined in the same manner. However, in the case of the system error number, it is determined that there is a malfunction such as a failure in the opposite device in the subsequent stage. Further, a different threshold may be used for each count target.

  Next, in S107, a signal determined to be in a concentrated state is extracted, and in S108, the CKPT function unit 16 writes it as a rejection number list in the management section area.

  Finally, in S109, the intersystem synchronization unit 17 reads the rejection number list stored in the management section area of the CKPT function unit 16, and synchronizes the system by transferring it to SBY in a timely manner.

  Each process of S104-S108 demonstrated above is repeatedly performed by overlapping each determination object period every 30 second, as shown in FIG. Since the determination process is performed by overlapping the determination target periods, even when there is a concentration of call connections extending over two determination target periods, the determination target periods can be reliably added up and included in the determination target.

  If the determination target periods are not overlapped (when the determination target period shown in FIG. 4 is changed from # 0 to # 2), the call concentration state of the incoming physical number xxx extending over the two periods cannot be accurately grasped. By detecting the determination target period overlapping with these periods (the determination target period shown in FIG. 4 is # 1), the state of the call for the incoming physical number is reliably and accurately detected.

  As shown in FIG. 5, the above-described series of processing flow realizes each process by the worker thread unit 12 pruning a signal accumulated in the reception queue from the reception queue.

  Here, for example, when the number of threads of the worker thread unit 12 is 8, processing can be executed in a maximum of 8 in parallel. When the processing time of the worker thread unit 12 is 2 ms, a signal of 1000 ms × 8 = 8000 can be processed per second.

  The time in which the length of the queue 13 needs to be considered, that is, the case where the queue 13 stays in the queue 13 is a case where the area used for post-processing is locked by another function. Are executed independently, so only copy processing at the time of list aggregation is required. That is, if about 200 us <2000/8 (processing time / number of threads in worker thread unit 12) = 250 us, there is almost no stagnation.

  In FIG. 3 and FIG. 4, in order to facilitate understanding, a list other than the total list such as an increase / decrease list and an extraction list is additionally used. However, these new lists are generated. Rather than reflecting the difference calculation result of two summary lists to one summary list and considering the summary list as an increase / decrease list, the total number of required lists is limited to a maximum of three as described above Is used to calculate the rejection number list.

[Counter list storage]
Next, the storage amount of the counter list will be described. When there is a simultaneous connection restriction, the stored amount of the counter list is the number of simultaneous connections and the call amount when the counter list exists.

  For example, if the number of counter lists is 20000 and the allowable call volume of the system is 1,200,000 BHCA (Busy Hour Call Attempts), 1,200,000 BHCA = 12,000,000 / 60 (calls / minute) = 20000 (calls / minute) can be saved A counter list of data areas is required. That is, a counter list of 20000 + 20000 = 40000 is required.

  Here, when the call count determination threshold is 500 (calls / minute), 20000/500 = 40 numbers are extracted in S107 of FIG. On the other hand, when 50 numbers instead of 40 numbers are targeted for extraction regardless of the number of simultaneous connections, 50 × 500 = 25000 (calls / minute) and 25000 × 60 = 1,500,000 BHCA is the allowable call volume. .

  In other words, only the destinations limited to the number of 40 numbers and 50 numbers to be extracted (that is, only the destinations limited to the size that can be stored in the management section area) are included in the reject number list. Since it only has to be stored and transferred to the SBY system, it can be transferred within the inter-system transfer area in the range fixed by the existing inter-system synchronization logic.

  That is, the allowable call volume in the present embodiment is not the original call volume determined by the relationship between the CPU usage rate and the call volume, but the resource storage volume. In the case of the SIP signal, even if the connection is not reached, an error response is returned and the reception buffer is consumed. This resource amount becomes the allowable call amount.

[Determination method of threshold value]
Although the threshold value used in S106 of FIG. 2 can be determined as appropriate, an example thereof will be described. For example, as provided by NTT FLET'S (Hikari Denwa Office Type: http://flets.com/hikaridenwa/office/service/option.html), it enables simultaneous calling of up to 8 channels and up to 32 numbers. May be used as a threshold value.

  Also, the distribution of calls existing in the determination target period is 27000 calls (111 (calls / second) × determination target time (for example, 60 seconds) = a value obtained by adding the number of simultaneous connections 20000 to 2660≈27000) in a normal distribution. A range that falls within σ2 may be set as a threshold value.

  In addition, the call volume corresponding to 400,000 BHCA-300,000 BHCA = remaining capacity 100,000 BHCA is 28 (calls / second), and the remaining power 28 (calls / second) × 60 seconds = 1667 (calls) may be set as a threshold value. Is possible.

[Operation of number conversion device: Processing flow for regulating call signal]
Next, a processing flow when the call signal is regulated will be described with reference to FIG.

  First, in S201, when a call signal is received from the calling terminal, in S202, the worker thread unit 12 refers to the rejection number list stored in the management section area of the CKPT function unit 16, and in S203, It is determined whether or not the incoming physical number (destination) of the received signal is included in the rejection number list.

  Next, in S <b> 204, a predetermined restriction is executed before the worker thread unit 12 transmits the call signal determined to be included in the reject number list to the receiving side.

  Finally, in S204, the regulated signal is transmitted to the receiving side by the worker thread unit 12. If it is determined in S203 that it is not included in the rejection number, it is transmitted as it is without executing the restriction process of S204.

  As the restriction process in S204, for example, a process for waiting for a certain period of time or a process for disabling reception for the same incoming physical number for a certain period of time is employed.

  For example, when it is impossible to accept for a certain period of time, a method of rejecting and responding without connecting to a subsequent stage is adopted. At this time, a method of stacking in a queue and delaying connection to a subsequent stage may be employed.

  For example, the time when the call cannot be accepted is (1) the time until the accepted call ends, and (2) the time after the retransmission time has passed in order to prevent an increase in network load due to the retransmission signal. Can do.

  In the case of (1) in particular, when the rejection number list is registered, the cancellation time is registered as a timer and restarted after the timer expires. It is also possible to detect that the threshold value is exceeded in the determination target period, and then disable acceptance for a certain time. In addition, when the reception is impossible, the counter may not be updated.

  When the counter exceeds the threshold value due to an error response to a connected call or continued call again while reception is not possible, it is determined that the threshold has been exceeded again at the summation timing, and the release time is updated each time to extend the reception unavailable time. You may make it do. It is also possible to count the number of connected calls for each destination and further use this number of connected calls.

<Second Embodiment>
In the first embodiment, the count list is updated at any time, and the unit time is calculated by calculating the difference between two count lists every 60 seconds among a plurality of count lists collected at the timing of every 30 seconds. The method for calculating the cumulative number of simultaneous connections in the determination target period has been described.

  The second embodiment proposes another calculation method for this cumulative number. Specifically, the count number (measurement number) in the count list is cleared (reset) every time a count list is generated by copying the count list, and the count is newly started from 0 every 30 seconds. Like that. As a result, the number of simultaneous connections in the 30 seconds is counted in the aggregation list generated every 30 seconds.

  And based on that, among the multiple total lists generated every 30 seconds, using two total lists for every 30 seconds within the determination target period, by adding the number of simultaneous connections for each incoming physical number, The cumulative number of simultaneous connections in the determination target period (60 seconds) is obtained.

  For example, in the case of FIG. 4, a total list B that collects the number of simultaneous connections that are totaled for 30 seconds from the total list A to the total list B, and a simultaneous connection that is totaled for 30 seconds from the total list B to the total list C The total number C of the number of simultaneous connections in the determination target period # 0 is calculated by adding the count list C that has collected the numbers.

  Although the method of calculating the cumulative number of simultaneous connections per unit time is different, even when the method described in this embodiment is adopted, the same effect as that described in the first embodiment is obtained. be able to.

  As described above, according to each embodiment, a response signal to a call signal is received from the incoming side, and is held in a totaling queue different from the queue used for call processing, and a reject number list is generated. Since the processing is executed separately from the call processing, determination / regulation processing can be performed independently of the main flow of call processing, so the load is concentrated without affecting the main flow of the call processing. Signals to the subsequent apparatus can be suppressed.

  According to each embodiment, since the calculation of the number of fluctuation measurements in the counting step is repeated at intervals shorter than the unit time, the determination target periods overlap, so that call connection concentration spans two determination target periods. Even if there is, it can be reliably counted and included in the determination target.

  According to each embodiment, the first determination step is performed by deleting the previously generated aggregation list from the two aggregation lists used for calculating the number of fluctuation measurements and using the data area of the deleted aggregation list. Therefore, the limited small resources can be effectively used, and the device itself can be downsized.

  According to each embodiment, since only the destinations limited in size that can be stored in the management section area of the checkpoint function are stored in the rejection number list, the management data area of the CKPT function synchronized between the systems is large. List data to be used at the time of regulation can be generated with a size that fits within this range.

DESCRIPTION OF SYMBOLS 1 ... Subscriber terminal 3 ... Transfer apparatus 5 ... Call processing server 7 ... Guidance server 10 ... Access network 100 ... Number conversion apparatus (AS)
DESCRIPTION OF SYMBOLS 11 ... Signal transmission / reception part 12 ... Worker thread part 13 ... Signal holding part (queue)
DESCRIPTION OF SYMBOLS 14 ... Signal etc. counter part 15 ... Signal determination part 16 ... CKPT function part 17 ... Inter-system synchronization part S1-S8, S101-S109, S201-S205 ... Processing step

Claims (9)

In a call control method performed by a call control device,
A response step for receiving a response signal to the calling signal from the called side, and holding in a totaling queue different from the queue used for call processing;
A measurement step of extracting a signal from the aggregation queue and generating a count list in which a predetermined measurement number for the signal is counted for each destination,
A step of generating a plurality of tabulation lists having the number of measurements at the periodic timing by periodically replicating the count list;
An aggregation step for calculating the number of fluctuation measurements per unit time for each destination using two aggregation lists;
A first determination step of determining, as a destination having a problem, a destination having a fluctuation measurement number per unit time exceeding a predetermined threshold;
A storage step of storing the destination determined to be troublesome as a reject number list in the management section area of the checkpoint function;
A second determination step of determining whether or not a destination of a newly received call signal is included in the rejection number list;
A restriction step for restricting a call signal determined to be included in the reject number list;
Is executed separately from the call processing. The counting step includes
The call control method according to claim 1, wherein the calculation of the variation measurement number is repeated at an interval shorter than the unit time. After calculating the number of fluctuation measurements per unit time, it further includes a deletion step of deleting the previously generated aggregation list from the two aggregation lists used for the calculation,
The first determination step includes
3. The call control method according to claim 1, wherein the determination is calculated using a data area of the deleted total list. The counting step includes
The variation measurement number per unit time is obtained by calculating a difference between the measurement numbers included in two aggregation lists every 60 seconds among a plurality of aggregation lists generated every 30 seconds. Item 4. The call control method according to any one of Items 1 to 3. The measurement step includes
Each time the count list is generated, the count of the count list is reset,
The counting step includes
The variation measurement number per unit time is obtained by adding the measurement numbers included in two aggregation lists generated within 60 seconds among a plurality of aggregation lists generated every 30 seconds. The call control method according to any one of claims 1 to 3. The storing step includes
6. The call control method according to claim 1, wherein only the destinations limited to a size that can be stored in the management section area are stored in the reject number list.   7. The call control method according to claim 1, further comprising a synchronization step of transferring a rejection number list stored in the management section area to a standby system. The predetermined number of measurements is
The call control method according to any one of claims 1 to 7, wherein the number of simultaneous connections of signals, the number of busy responses indicating congestion, and the number of errors from the opposite device at the subsequent stage are one or more. The call control device includes:
9. The call control method according to claim 1, wherein the call control method is a number conversion device that converts a logical telephone number into a physical telephone number.

Images