JP2016052024A - Back-off control method and back-off control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate an impact of congestion in a short time after the congestion state of a network is eliminated.SOLUTION: A back-off control method of a network device is a method that, when the network device receives a communication request from a terminal side in a state in which the congestion of a network is detected, rejects the acceptance of the communication request and returns a back-off timer value which is time width during which retransmission of the communication request should be restrained. The back-off control method includes the steps of: predicting the time transition of a new request signal number which is the number of new communication requests from the terminal side on the basis of a prescribed prediction model; predicting the time transition of a re-request signal number which is the number of communication requests re-transmitted by the terminal to which the back-off timer value was returned on the basis of latest back-off control history; and predicting time when the total of the new request signal number and the re-request signal number will be below a request signal number which the network is able to process, determining a back-off timer value to be returned, and returning the value.SELECTED DRAWING: Figure 5

Description

  The present invention relates to a congestion countermeasure technique for a communication system, and more particularly to a back-off control method and a back-off control device for suppressing re-transmission of a communication terminal.

  With the development of information communication technology in recent years, provision of various new services has been started. On the other hand, the amount of communication from communication terminals is also increasing, and issues such as the seriousness of core network congestion due to the increase in communication amount have been clarified, and congestion control technology is being studied as a countermeasure against congestion .

  As an example of this, in mobile communication networks and fixed communication networks, congestion associated with the development of M2M (Machine-to-Machine) communication that connects not only between people but also between people and servers, and without human intervention. There is a problem. M2M communication has already been used in elevator remote monitoring, vending machine inventory management, point-of-sale (POS) point of sale (POS) systems, logistics distribution tracking management systems, and the like. In the future, the application target will be expanded to household appliances, meter devices, sensor devices, health-related devices, electric vehicles, etc. in each home, and the number of M2M terminals is expected to become enormous.

  If it is assumed that many M2M terminals and the network side are disconnected all at once due to a natural disaster or the like, there is a concern that each M2M terminal automatically performs a reconnection operation to cause congestion. On the network side, it is necessary to calm down the congestion at an early stage, and the development and standardization of technology relating to congestion control are being promoted by 3GPP (3rd Generation Partnership Project) or the like (for example, see Non-Patent Document 1).

3GPP TS 24.301 V12.5.0, "Non-Access-Stratum (NAS) protocol for Evolved Packet System (EPS); Stage 3", 2014-06

  As one of the conventional techniques for calming down congestion on the network side, back-off control that suppresses transmission of communication terminals at the time of congestion is used. FIG. 10 is an explanatory diagram of a conventional back-off control technique.

  When a call request is issued from the communication terminal 200A using a call signal ([1]), the call request ([1]) is accepted when the signal processing device 100B detects network congestion ([2]). I can't. Therefore, the signal processing device 100B returns an acceptance refusal signal to the communication terminal 200A ([3]). This acceptance refusal signal ([3]) is given a parameter indicating that the network side is in a “congestion state”, and a back-off timer value that is a time width in which the communication terminal 200A should suppress retransmission. This back-off timer value is set to a random value within a predetermined suppression time range (for example, 15 to 30 minutes) in order to avoid a large number of retransmission requests simultaneously from a plurality of communication terminals 200A. As a result, the timing of re-transmission requests varies.

  The communication terminal 200A starts by setting the back-off timer to the back-off timer value notified by the reception rejection signal ([3]) ([4]), and when the back-off timer times out ([6]) again. A call request is made ([7]). At this time, if the congestion state of the network has already been released ([5]), the signal processing device 100B accepts this call request ([7]) and returns a call acceptance signal to the communication terminal 200A ( [8]).

  In the conventional method in which a random back-off timer value is indicated by such an acceptance refusal signal ([3]), the communication terminal 200A continues until the back-off timer times out even if the network congestion state has already been released. The possibility of not being able to send time increases. Therefore, there is a problem that the influence of congestion cannot be eliminated in a short time.

  The present invention has been made to solve the above-described problems, and a back-off control method and a back-off control apparatus that can eliminate the influence of congestion in a short time after the congestion state of the network is released. The purpose is to provide.

  In order to achieve the above object, the present invention rejects acceptance of a communication request and retransmits the communication request when a communication request is received from the terminal side while detecting network congestion. A back-off control method for a network device that returns a back-off timer value that is a time width to be suppressed, the number of new request signals being the number of new communication requests from the terminal side based on a predetermined prediction model And predicting the time transition of the number of re-request signals that is the number of communication requests retransmitted from the terminal to which the back-off timer value is returned based on the latest back-off control history Predicting the time at which the total value of the step and the number of new request signals and the number of re-request signals will be less than the number of request signals that can be processed by the network, and Determining a timer value, and shall include the steps of returning the value, the.

  By doing so, an appropriate back-off timer value is determined based on the prediction of the time when the network congestion state is released, and the influence of congestion can be eliminated in a short time.

  According to another aspect of the present invention, in the back-off control method, the back-off control history is the number of terminals for each return back-off timer value, and the network device includes the back-off timer value for each of the terminals. When the device is returned, the number of terminals for each return back-off timer value is updated.

  By doing so, it is possible to improve the prediction accuracy of the number of rerequest signals, which is the number of communication requests retransmitted from the terminal to which the backoff timer value is returned.

According to another aspect of the present invention, in the backoff control method, the network device includes:
Each step is executed individually each time a communication request is received from the terminal side, and the back-off timer value to be returned can be processed by the network with the total value of the number of new request signals and the number of re-request signals It was determined to be equal to the elapsed time up to the time when it would be expected to be less than the number of required signals.

  In this way, for each communication request, a minimum backoff timer value equal to the elapsed time until the time when the network congestion state is expected to be released can be determined.

  According to another aspect of the present invention, in the back-off control method, the network device performs the steps in the first communication request among n communication requests received from the terminal side during a predetermined unit time. The backoff timer value to be returned for the first communication request is calculated so that the total value of the number of new request signals and the number of re-request signals is less than the number of request signals that can be processed by the network. A back-off timer value determined to be equal to the elapsed time up to the predicted time and returned for the remaining n-1 communication requests is returned to the determined first communication request. In addition, a value obtained by adding a predetermined fixed value once to n-1 times or a value obtained by adding n-1 pseudorandom numbers having a predetermined range is sequentially determined.

  By doing so, the processing for predicting the number of request signals need only be executed once for a plurality of communication requests, so that the processing load on the network device can be reduced.

  In another aspect of the present invention, when a communication request is received from the terminal side in a state where network congestion is detected, the acceptance of the communication request is rejected, and the time width in which retransmission of the communication request should be suppressed Is a back-off control device that returns a back-off timer value, and based on a predetermined prediction model, a new request that predicts a time transition of the number of new request signals that is the number of new communication requests from the terminal side A re-request signal for predicting a time transition of the number of re-request signals, which is the number of communication requests re-transmitted from the terminal to which the back-off timer value is returned, based on a signal number prediction unit and the latest back-off control history Predicting the time when the total value of the number of new request signals and the number of re-request signals will be less than the number of request signals that can be processed by the network, and determining the back-off timer value to be returned. And it was intended to comprise a back-off timer control unit to return the value, the.

  By doing so, an appropriate back-off timer value is determined based on the prediction of the time when the network congestion state is released, and the influence of congestion can be eliminated in a short time.

  According to another aspect of the present invention, in the back-off control device, the back-off control history is a number of terminals for each return back-off timer value, and a return timer for storing the number of terminals for each return back-off timer value. A terminal number storage unit by value is provided, and the back-off timer control unit updates the number of terminals by return back-off timer value when the back-off timer value is returned to each of the terminals.

  By doing so, it is possible to improve the prediction accuracy of the number of rerequest signals, which is the number of communication requests retransmitted from the terminal to which the backoff timer value is returned.

  According to another aspect of the present invention, in the back-off control device, the back-off timer control unit individually determines and returns the back-off timer value to be returned each time a communication request is received from the terminal side. Determining a back-off timer value equal to the elapsed time until the time when the total value of the number of new request signals and the number of re-request signals is expected to be less than the number of request signals that can be processed in the network; did.

  In this way, for each communication request, a minimum backoff timer value equal to the elapsed time until the time when the network congestion state is expected to be released can be determined.

  According to another aspect of the present invention, in the back-off control device, the back-off timer control unit responds to a first communication request among n communication requests received from the terminal side during a predetermined unit time. The back-off timer value to be returned is determined to be equal to the elapsed time until the time when the total value of the number of new request signals and the number of re-request signals is predicted to be less than the number of request signals that can be processed by the network. Then, the back-off timer value to be returned for the remaining n-1 communication requests is changed from a predetermined fixed value once to n- to the back-off timer value to be returned for the determined first communication request. A value obtained by adding once or a value obtained by adding n−1 pseudorandom numbers having a predetermined range is sequentially determined.

  By doing so, the processing for predicting the number of request signals need only be executed once for a plurality of communication requests, so that the processing load on the network device can be reduced.

  According to the present invention, the minimum back-off timer value can be set based on the processing capacity on the network device side and the predicted value of the requested traffic amount from the terminal at the time of network congestion, so after the congestion state is released, The influence of congestion can be eliminated in a short time.

It is a functional block diagram which shows the structural example of the network apparatus which concerns on this invention. It is explanatory drawing about the determination method of the back-off timer value which concerns on this invention. It is a figure which shows the structure and example of data of a terminal number memory | storage part according to return timer value, (a) is a data example before the back-off timer value return to a certain terminal, (b) is a back-off timer value return to the said terminal. It is a later data example. It is explanatory drawing about the method of estimating the re-request signal number from the data of the terminal number memory | storage part according to return timer value, (a) is a data example of the terminal number memory | storage part according to return timer value in the prediction time T, (b) is. The predicted value of the number of re-request signals when each terminal makes a re-request immediately after timeout, and (c) is the predicted value of the number of re-request signals when the safety factor is 1.5. It is a flowchart of the processing method which estimates the number of request signals for every request signal, and determines a return back-off timer value. A return back-off timer value is determined by predicting the number of request signals for the first request signal among n request signals received in a predetermined unit time, and a return back for the remaining n-1 request signals. The off-timer value is a flowchart of a processing method for sequentially determining the difference between the timer values to be a fixed value. A return back-off timer value is determined by predicting the number of request signals for the first request signal among n request signals received in a predetermined unit time, and a return back for the remaining n-1 request signals. The off-timer value is a flowchart of a processing method for determining using a pseudo-random number. It is a communication sequence figure at the time of applying the processing method which estimates the number of request signals for every request signal, and determines a return back-off timer value to MM-Backoff. It is a communication sequence diagram at the time of applying to MM-Backoff the processing method which determines a backoff timer value collectively with respect to n request signals received in unit time. It is explanatory drawing about the conventional back-off control technique. It is explanatory drawing of the conventional back-off control method in the mobile communication network prescribed | regulated by 3GPP.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings as appropriate. here,
An embodiment of the present invention will be described by comparison with a backoff control method in a mobile communication network defined by 3GPP.

  FIG. 11 is an explanatory diagram of a conventional back-off control method in a mobile communication network defined by 3GPP. As shown in FIG. 11, 3GPP defines three types of back-off control functions: SM-Backoff (Session Management Backoff), MM-Backoff (Mobility Management Backoff), and RRC-Backoff (Radio Resource Control Backoff). .

<SM-Backoff> (Refer to the upper communication sequence in FIG. 11)
A mobility control node (MME: Mobility Management Entity) 100A requests a PDN (Packet Data Network) gateway (P-GW: PDN Gateway) 100C to establish a communication bearer by calling ([1]) from the communication terminal 200A. ([2]). At this time, if the P-GW 100C detects congestion ([3]), a Reject signal ([4]) in which a congestion flag is set is returned to the MME 100A to notify connection rejection. When the MME 100A receives this Reject signal ([4]), the Reject signal in which a random back-off timer value is set together with the congestion flag to the communication terminal 200A in order to notify the connection refusal and suppress retransmission. ([5]) is returned. As a result, the retransmission from the communication terminal 200A is suppressed and the timing of the retransmission is distributed until the time of the back-off timer value elapses and the timeout occurs (hereinafter also referred to as “timer value expires”). Let

<MM-Backoff> (Refer to the middle communication sequence in FIG. 11)
When the MME 100B that has received the attach / location registration request ([1]) from the communication terminal 200A detects congestion ([2]), the communication terminal 200A is given a random backoff timer value together with the congestion flag. The set Reject signal ([3]) is returned. Thereby, until the timer value expires, the re-request of the attach / position registration from the communication terminal 200A is suppressed and the transmission timing of the re-request signal is distributed.

<RRC-Backoff> (Refer to the lower communication sequence in FIG. 11)
When the base station 300 that has received the wireless setting request ([1]) from the communication terminal 200A detects congestion ([2]), a reject in which a random back-off timer value is set for the communication terminal 200A. Return the signal ([3]). Thereby, until the timer value expires, the re-request of the wireless setting from the communication terminal 200A is suppressed and the transmission timing of the re-request signal is distributed.

  The range of the back-off timer value set in these three types of back-off control functions is defined by 3GPP TS24.008. The default value is randomly set for each communication terminal 200A between 15 minutes and 30 minutes. It is recommended to set For this reason, even if the congestion state of the network or device is released early, transmission and transmission from the communication terminal 200A cannot be performed until the timer value expires (up to 30 minutes). There is a problem that the influence cannot be eliminated.

  On the other hand, the present invention predicts the time at which the congestion state of the network or device will be released based on the traffic transition prediction, and sets an appropriate backoff timer value. Eliminate the effects of congestion in a short time.

  FIG. 1 shows a configuration example of a network apparatus (back-off control apparatus) that executes a back-off control method according to the present invention, and FIG. 2 shows an image of a method for determining a back-off timer value based on traffic transition prediction. The basic back-off control operation in the present invention will be described below with reference to FIGS.

  FIG. 1 is a functional block diagram showing a configuration example of a network device according to the present invention. As shown in FIG. 2, the network device 100 is connected to one or more terminal-side communication partners 200 (direct connection with a terminal or indirect connection with a terminal via a network device such as an access device).

  The network device 100 is configured by, for example, a commercially available computer, and includes a transmission / reception unit 101, a transmission / reception buffer 102, a data processing unit 103, and a signal control unit 104 as functional units in which respective functions are realized by program control. The transmission / reception unit 101 transmits / receives a signal between the data processing unit 103 or the signal control unit 104 and a terminal-side communication partner (hereinafter abbreviated as “terminal”) 200 via the transmission / reception buffer 102.

  The signal control unit 104 includes a back-off timer control unit 105, a new request signal number prediction unit 106, a re-request signal number prediction unit 107, and a terminal number storage unit 108 by return timer value. When the data processing unit 103 detects a congestion state of the network or its own device, the signal control unit 104 starts a predetermined back-off control operation. When the signal control unit 104 receives a communication request from the terminal 200, the signal control unit 104 rejects acceptance of the communication request and sends a signal for notifying a back-off timer value that is a time width in which reissuance of the communication request should be suppressed. Return to 200.

At this time, the number of new request signals prediction unit 106 predicts the number of new request signals r _T (t) per unit time transmitted from the terminal 200 that is not subject to backoff control based on a predetermined prediction model. . Several prediction models for the number of new request signals r _T (t) have been proposed.For example, the number of request signals over time can be obtained by acquiring the number of past request signals over a long period of several months or one year. There is a method of creating a prediction model representing the tendency of the above and removing noise by a Kalman filter or the like.

Further, the re-request signal number prediction unit 107 refers to the notification history data of the back-off timer value (the latest back-off control history) stored in the return timer value-specific terminal number storage unit 108, thereby performing back-off control. The number of re-request signals B _T (t) per unit time transmitted from the target terminal 200 is predicted. Details of this prediction method will be described later. The back-off timer control unit 105 determines that the total request signal number R _T (t) obtained by adding r _T (t) and B _T (t) is less than the number of processable signals C (= equipment capacity) per unit time. A time at which the congestion state will be released is predicted, and a back-off timer value to be returned to the terminal 200 is determined from that time. Further, the back-off timer control unit 105 updates the notification history data in the return timer value-specific terminal number storage unit 108 corresponding to the back-off timer value notified to the terminal 200 (counts up the number of terminals).

  When the back-off timer value determined here is notified to the terminal 200 that has transmitted the communication request, the terminal 200 waits until the timer value expires and then issues the communication request again. Therefore, if the prediction accuracy of the congestion state release time is high, a communication request reissued from the terminal 200 after the timer value expires can be accepted immediately, so that the influence of the congestion state can be eliminated in a short time.

FIG. 2 is an explanatory diagram of a method for determining a back-off timer value according to the present invention. This graph conceptually shows the temporal change in the predicted traffic value from the present time (t = T) when the back-off control operation is performed. In the drawing, a horizontal broken line represents the number C of signals that can be processed per unit time of the network or network device 100 (= equipment capacity). r _T (t) is a predicted value of the number of new request signals per unit time transmitted from the terminal 200 that is not subject to back-off control, and B _T (t) that is hatched is back-off This is a predicted value of the number of re-request signals per unit time transmitted from the terminal 200 being controlled. When the total required signal number R _T (t), which is the sum of both, exceeds the equipment capacity C, it becomes congested and the back-off control operation is started, and the total required signal number R _T (t) becomes less than the equipment capacity C. Thus, back-off control is performed.

In FIG. 2, since the actual total number of required signals exceeds the equipment capacity C at the present time (t = T), it is necessary to apply back-off control to the request signals that exceed the equipment capacity C. Here, the number of newly requested signals r _T (t) continues to decrease from the present time (t = T), and the total number of requested signals R _T (t) falls below the installed capacity C after time T ₀ (t = T +). Then, the request signal to be backed off at the present time (t = T) can be accepted after time T ₀ (t = T + T ₀ ). Therefore, this time T ₀ when the total required signal number R _T (t) is less than the equipment capacity C is set as a back-off timer value, and is returned to the terminal 200 that is newly subject to back-off control. As a result, it is possible to perform back-off control while maximally utilizing the processing capability of the network or network device at the time of congestion, and the influence of congestion can be eliminated in a short time.

In determining the timer value T ₀ , ωR _T (t) and the equipment capacity C are calculated by applying a predetermined safety factor ω (value of 1 or more) from the viewpoint of preventing underestimation of traffic prediction. It may be compared. There are several methods for determining the safety factor ω. For example, when the predicted result R _T (t) for the last n seconds is compared with the actual number of required signals, and the predicted result R _T (t) is exceeded, The value of the safety factor ω can be calculated based on the average value or the maximum value. Specifically, when the actual number of requested signals is an average X increase of the prediction result R _T (t), ω = 1.X may be used, and the maximum value of the actual number of requested signals is predicted. If the result R _T (t) is Y times, ω = Y may be set. Further, the safety factor ω may be a fixed value determined based on a long-term tendency. Alternatively, instead of the safety factor ω, the average value m of the difference between the actual number of required signals and the prediction result R _T (t) can be obtained, and R _T (t) + m can be compared with the installed capacity C. Good.

  FIG. 3 is a diagram illustrating a configuration and data example of a terminal number storage unit by return timer value, where (a) is a data example before returning a back-off timer value to a certain terminal, and (b) is a back to the terminal. It is an example of data after an off timer value is returned.

  As shown in FIG. 3, the number-of-terminals-by-return timer value storage unit 108 stores the number of scheduled signal retransmission terminals for each timer value to be returned (after X seconds) as the backoff timer value notification history. Each time the backoff timer value determined by the backoff timer control unit 105 is returned to the terminal 200, the data corresponding to the timer value is incremented by one and updated. . For example, if the back-off timer value N is returned to a certain terminal 200, the number of terminals scheduled to retransmit the signal after N seconds, which was “0” before the return, as shown in the data example of FIG. As shown in the data example of FIG. 3B, the value is updated to “1” after the back-off timer value is returned.

  FIG. 4 is an explanatory diagram of a method for predicting the number of re-request signals from the data in the terminal number storage unit by return timer value, (a) is a data example of the terminal number storage unit by return timer value at the prediction time point T, (B) is a predicted value of the number of re-requested signals when each terminal makes a re-request immediately after a timeout, and (c) is a predicted value of the number of re-requested signals when the safety factor is 1.5.

As shown in FIG. 4A, the number of signal retransmission scheduled terminals stored in the return timer value-specific terminal number storage unit 108 at the prediction time T is “8” after 1 second, “12” after 2 seconds, Suppose that ... At this time, by reading the data stored in the terminal number storage unit 108 by return timer value, as shown in FIG. 4B, the re-request signal number B _T (t) is changed to B _T (T + 1 ) = 8, B _T (T + 2) = 12,. For example, when the safety factor ω _B for the re-request signal number is 1.5, the re-request signal number prediction unit 107 multiplies each value by 1.5, as shown in FIG. In addition, the number of re-request signals B _T (t) is predicted as B _T (T + 1) = 12, B _T (T + 2) = 18,.

  Subsequently, in addition to the basic processing method in which the method for determining the back-off timer value described above with reference to FIG. 2 is applied to each request signal, as a modification thereof, n request signals received in a predetermined unit time. A method of applying the above determination method only to the first request signal will be described. The latter method further includes a method of sequentially setting back-off timer values for the remaining n-1 request signals with a fixed time interval, and a method of sequentially adding pseudo-random numbers having a predetermined range. There is.

  FIG. 5 is a flowchart of a processing method for determining the return back-off timer value by predicting the number of request signals for each request signal.

  When the network device 100 detecting the congestion state receives a request signal from the terminal 200 at the current time T (step S51), the network device 100 determines the back-off timer value I by the processing of steps S52 to S56. Then, the network device 100 suppresses retransmission from the terminal 200 by setting the timer value I as an acceptance refusal signal and returning it to the terminal 200 in step S58.

In step S52, the back-off timer control unit 105 sets a counter i for updating the predicted time to “0”. In step S53, the back-off timer control unit 105 calculates a predicted value R _T (T + i) of the total number of requested signals. This R _T (T + i) is predicted by the new request signal number prediction unit 106 and the new request signal number r _T (T + i) from the back-off unapplied terminal and the re-request signal number prediction unit 107 It is calculated by adding the number of re-requested signals B _T (T + i) from the terminal that is applying the back-off.

In subsequent step S54, the back-off timer control unit 105 determines whether or not the value of R _T (T + i) calculated by taking the safety factor ω into consideration is larger than the equipment capacity C of the congestion device. The process proceeds to step S55 (Yes side), and if not, the process proceeds to step S56 (No side). In step S55, the back-off timer control unit 105 increments the value of the counter i, returns the process to step S53, and repeats the same process until the value of R _T (T + i) becomes C or less.

  In step S56, the back-off timer control unit 105 determines the value of the counter i at that time as the back-off timer value I. In step S57, the back-off timer control unit 105 updates the data corresponding to the timer value I in the return timer value-specific terminal number storage unit 108 (increments the number of terminals), and then in step S58, the request source The timer value I is returned to the terminal 200 and the process is terminated.

  FIG. 6 shows the number of request signals for the first request signal among n request signals received in a predetermined unit time to determine the return back-off timer value, and the remaining n-1 requests. The return back-off timer value for the signal is a flowchart of a processing method for sequentially determining the difference between the timer values to be a fixed value.

  When the network device 100 detecting the congestion state receives n request signals in a predetermined unit time (between T and T + 1) starting from the current time T (step S61), the network device 100 performs the step The representative timer value I is determined by the processing of S62 to S66. The representative timer value is a back-off timer value that is returned for the first request signal received by the network device 100 during the unit time. Since the process of determining the representative timer value I is the same as the process of steps S52 to S56 in FIG. 5, detailed description thereof is omitted.

  After determining the representative timer value I in step S66, the back-off timer control unit 105 returns n-1 back-off timer values to be returned for each of the remaining n-1 request signals in step S67. Confirm. At this time, the back-off timer value returned for the second and subsequent request signals is obtained by sequentially adding a predetermined fixed value A to the representative timer value I in order to avoid simultaneous re-requests from the terminal 200. Go. Therefore, the back-off timer value returned for the nth received request signal is I + (n−1) A. Note that the value of the fixed value A can be changed as appropriate before or after the start of system operation, and it is preferable to determine the optimum value from the equipment capacity C of the network or network device 100, the number of accommodated terminals, the number of rerequested signals, and the like. .

  In subsequent step S68, the back-off timer control unit 105 updates (increments the number of terminals) data corresponding to all timer values determined in the return timer value-specific terminal number storage unit 108, and then requests each request in step S69. The timer value determined for each signal is returned to the requesting terminal 200 and the process is terminated.

  As described above, in this processing method, the processing load on the network apparatus 100 can be reduced because the process of predicting the number of request signals need only be executed once for a plurality of communication requests.

  FIG. 7 shows the number of request signals for the first request signal among n request signals received in a predetermined unit time, determines the return back-off timer value, and the remaining n-1 requests. The return back-off timer value for a signal is a flowchart of a processing method for determining using a pseudo-random number.

Since the processing in steps S71 to S79 in FIG. 7 is the same as the processing method shown in steps S61 to S69 in FIG. 6 except for step S77, detailed description thereof is omitted. In step S77, n-1 back-off timer values to be returned for each of the remaining n-1 request signals are determined. At this time, the back-off timer value to be returned for the nth received request signal is a value obtained by adding the pseudo-random number α _n−1 having a predetermined range to the representative timer value I. Note that the range of the pseudo random number can be changed as appropriate before or after the system operation is started, and it is preferable to determine the optimum value from the equipment capacity C of the network or network device 100, the number of accommodated terminals, the number of re-requested signals, and the like.

  8 and 9 show examples of communication sequences when the present invention is applied to MM-Backoff, which is one of the conventional back-off control techniques described above with reference to FIG. Among these, FIG. 8 is an example of a communication sequence when a processing method for predicting the number of request signals for each request signal and determining a return back-off timer value is applied. FIG. 9 is an example of a communication sequence when a processing method for collectively determining a back-off timer value for n request signals received per unit time is applied.

  In FIG. 8, the mobility control node (MME) 100A having the function as the network device of the present invention changes the number of request signals individually for each received request signal (attach / location registration), that is, for each communication terminal 200A. Predict. Thereby, each back-off timer value (denoted as transition predicted back-off timer value) is determined, and the value is returned to the requesting communication terminal 200A by the Reject signal. FIG. 8 shows a communication sequence when back-off control is performed on m communication terminals 200A.

In FIG. 9, a mobility control node (MME) 100A having a function as a network device according to the present invention has 1 1 for n request signals (attach / location registration) transmitted from a communication terminal 200A for a predetermined unit time. Predict the transition of the number of requested signals only once. Thereby, n back-off timer values (denoted as transition predicted back-off timer values) are determined collectively, and these values are notified to each communication terminal 200A by a Reject signal. At this time, the representative timer value I is determined based on the transition prediction of the number of request signals for the first received request signal, and the fixed value A for the representative timer value I is set to n for the nth received request signal. -1 times or by adding a pseudo-random number α _n-1 . FIG. 9 shows a communication sequence in the case where backoff control is collectively performed for n communication terminals 200A.

  Here, an example of a communication sequence when the present invention is applied to MM-Backoff is shown, but the present invention is based on SM-Backoff and RRC-Backoff which are the conventional back-off control technologies described above with reference to FIG. It can be similarly applied to.

  Although the description of the mode for carrying out the present invention has been described above, the embodiment of the present invention is not limited to this, and it goes without saying that various modifications can be made without departing from the spirit of the present invention. Yes.

100 Network device (back-off control device)
100A Mobility control node (MME)
DESCRIPTION OF SYMBOLS 101 Transmission / reception part 102 Transmission / reception buffer 103 Data processing part 104 Signal control part 105 Back-off timer control part 106 New request signal number prediction part 107 Rerequest signal number prediction part 108 Terminal number memory | storage part according to return timer value 200 Terminal side communication partner (terminal) )
200A communication terminal (terminal)
300 base station

Claims (8)

When a communication request is received from the terminal while network congestion is being detected, acceptance of the communication request is rejected and a back-off timer value that is a time width for suppressing retransmission of the communication request is returned. A network device back-off control method comprising:
Predicting a time transition of the number of new request signals, which is the number of new communication requests from the terminal side, based on a predetermined prediction model;
Based on the latest back-off control history, predicting a time transition of the number of re-request signals, which is the number of communication requests re-transmitted from the terminal to which the back-off timer value is returned;
Predicting the time when the total value of the number of new request signals and the number of re-request signals will be less than the number of request signals that can be processed in the network, determine the backoff timer value to be returned, Step to return,
A back-off control method. The back-off control method according to claim 1,
The back-off control history is the number of terminals by return back-off timer value,
When the network device returns the back-off timer value to each of the terminals, the number of terminals for each return back-off timer value is updated.
A back-off control method. In the back-off control method according to claim 1 or 2,
The network device is:
Each of the above steps is executed individually each time a communication request is received from the terminal side,
The returned back-off timer value is determined to be equal to the elapsed time until the time when the total value of the number of new request signals and the number of re-request signals is predicted to be lower than the number of request signals that can be processed in the network. To
A back-off control method. In the back-off control method according to claim 1 or 2,
The network device is:
Performing each of the steps for the first communication request among n communication requests received from the terminal side during a predetermined unit time;
The time at which the total value of the number of new request signals and the number of re-request signals is expected to be less than the number of request signals that can be processed by the network, as a back-off timer value to be returned for the first communication request Determined to be equal to the elapsed time until
The back-off timer value to be returned for the remaining n-1 communication requests is set to a predetermined fixed value once to n-1 times as the back-off timer value to be returned for the determined first communication request. Sequentially determined as an added value or a value obtained by adding n-1 pseudorandom numbers having a predetermined range.
A back-off control method. When a communication request is received from the terminal while network congestion is being detected, acceptance of the communication request is rejected and a back-off timer value that is a time width for suppressing retransmission of the communication request is returned. A back-off control device,
Based on a predetermined prediction model, a new request signal number prediction unit that predicts a time transition of the number of new request signals, which is the number of new communication requests from the terminal side,
Based on the latest back-off control history, a re-request signal number prediction unit that predicts a time transition of a re-request signal number that is the number of communication requests retransmitted from the terminal to which the back-off timer value is returned;
Predicting the time when the total value of the number of new request signals and the number of re-request signals will be less than the number of request signals that can be processed in the network, determine the backoff timer value to be returned, A back-off timer control unit to be returned;
A back-off control device comprising: The back-off control device according to claim 5,
The back-off control history is the number of terminals by return back-off timer value,
A return timer value-specific terminal number storage unit for storing the number of terminals by return back-off timer value;
The back-off timer control unit updates the number of terminals for each return back-off timer value when the back-off timer value is returned to each of the terminals.
A back-off control device. The back-off control device according to claim 5 or 6,
The back-off timer control unit
Each time a communication request is received from the terminal side, the backoff timer value to be returned is individually determined,
The returned back-off timer value is determined to be equal to the elapsed time until the time when the total value of the number of new request signals and the number of re-request signals is predicted to be lower than the number of request signals that can be processed in the network. To
A back-off control device. The back-off control device according to claim 5 or 6,
The back-off timer control unit
The back-off timer value returned for the first communication request among the n communication requests received from the terminal side during a predetermined unit time is the total value of the number of new request signals and the number of re-request signals Is determined to be equal to the time elapsed until the time at which it is expected to be below the number of required signals that can be processed by the network,
The back-off timer value to be returned for the remaining n-1 communication requests is set to a predetermined fixed value once to n-1 times as the back-off timer value to be returned for the determined first communication request. Sequentially determined as an added value or a value obtained by adding n-1 pseudorandom numbers having a predetermined range.
A back-off control device.

Images