JP2965006B2 - CDV reduction algorithm circuit self-monitoring method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a self-monitoring method for a CDV (Cell Delay Variation) reducing algorithm circuit in a shaping function, and more particularly to a self-monitoring method using non-data cells in a cell flow.

[0002]

2. Description of the Related Art Conventionally, this type of CDV reduction algorithm requires a shaping function in an ATM (Asynchronous Transfer Mode) network in order to reduce burstiness which causes a reduction in network utilization efficiency. It is used at the time. More specifically, it is possible to guarantee the ideal transmission interval of cells to some extent using a CDV reduction algorithm implemented by a hardware logic circuit.

FIG. 1 is a processing flow of a conventional CDV reduction algorithm. The conventional CDV reduction algorithm includes a minimum cell interval Ts, a CDV residual allowable value τs, a cell arrival time ta, a theoretical transmission time TET of the next cell when shaping is assumed, and an ideal cell as a result of shaping. Assuming the transmission time to, V of the input cell
The transmission time of the input cell is determined from the arrival time of the cell immediately before P, and is written to the time position of the shaping buffer.

The operation patterns of the CDV reduction algorithm are shown in FIG. That is, when the arrival cell is a user cell, the TET with respect to the cell arrival time ta is:
If TET> ta + τs, perform shaping,
The transmission time to of the cell is set as to = TET−τs, and the theoretical transmission time TET of the next cell is set as TET + T.
s (pattern).

On the other hand, when TET ≦ ta + τs, T
If ET> ta, the transmission time to of the cell is to =
ta, and the theoretical transmission time TET of the next cell is TET + Ts (pattern). If TET <ta, the transmission time to of the cell is to = ta, and ta + Ts is the theoretical transmission time of the next cell. A new TET is set (pattern).

When the arrival cell is a non-user cell, ta <t
In the case of o, bypass is performed as it is (pattern) and ta
If> to, it is updated to to = ta and bypassed (pattern).

[0007]

By employing the above-mentioned CDV reduction algorithm, in an ATM network,
Burst characteristics that cause a decrease in network utilization efficiency are reduced,
Although it is possible to guarantee an ideal cell transmission interval to some extent, if an abnormality occurs in a calculation result due to an abnormality in a hardware logic circuit that realizes the CDV reduction algorithm,
Precise shaping is not possible.

As a means for avoiding the above-mentioned abnormal state, a method of preventing a malfunction due to an abnormality by making a hardware logic circuit redundant and comparing each operation result can be considered, but a problem that the circuit scale is increased. There is.

The present invention provides a mechanism capable of immediately detecting the occurrence of an abnormality in the above-described CDV reduction algorithm circuit and promptly restoring the abnormal state by notifying an alarm or the like. It is an object of the present invention to provide a mechanism capable of detecting an abnormal state in-service without interrupting the operation.

[0010]

A case is considered where non-data cells whose arrival intervals are periodic or whose arrival times can be predicted based on a certain function are included in a cell stream, and this is temporarily referred to as a periodic cell. The present invention is characterized in that the self-monitoring of the CDV reduction algorithm is performed using the arrival time of the periodic cell.

Since the arrival time of a periodic cell can be reliably predicted, a possible state can be artificially created by manipulating the theoretical transmission time or the ideal transmission time. Utilizing the characteristics, a self-monitoring pattern is defined and parameters for self-monitoring are set.

After detecting the periodic cell, a CDV reduction algorithm operation is performed on the basis of various parameters for self-monitoring, and the result of each determination as to which path of the algorithm the output cell has passed has been obtained. And output
If the values match with the expected value, the CDV reduction algorithm is determined to be normal. If the values do not match, an alarm is notified to the device management function as abnormal. In this case, the self-monitoring operation is continued.

The CDV reduction algorithm circuit self-monitoring method of the present invention is a monitoring function using a method of constantly monitoring the algorithm operation result. The CDV reduction algorithm circuit self-monitoring method uses a non-data cell (periodic cell) for control. By performing self-monitoring, shaping operation can be performed in-service without stopping. The calculation by monitoring the five patterns periodically result, the circuit of the probability is low portion of the operation execution abnormality can be detected immediately.

[0014]

FIG. 3 is a block diagram showing an embodiment of the present invention. In FIG. 3, the ATM cell is input to the CDV reduction algorithm circuit 31 having the processing flow function of FIG. 1, subjected to shaping, and output after the cell transmission interval is adjusted.

When detecting the periodic cell, the algorithm monitoring unit 32 performs self-monitoring of the CDV reduction algorithm 31 and outputs an alarm when the algorithm is abnormal. The self-monitoring of the algorithm is not performed on the arriving cells other than the periodic cell, and the normal CDV reduction algorithm operation is performed. The parameter memory 33 stores self-monitoring parameters and self-monitoring pattern expected values.

FIG. 4 is a diagram showing a processing flow of the CDV reduction algorithm self-monitoring method of the present invention. In FIG. 4, a parameter initial setting unit 41 performs initial setting of self-monitoring parameters. Cell arrival management unit 42
Manages the cell arrival time. Periodic cell determination unit 4
3 is for determining whether or not the arrival cell is a periodic cell. The self-monitoring parameter setting unit 44 sets a self-monitoring parameter as a parameter of an algorithm operation for a cell determined to be a periodic cell by the periodic cell determination unit 43.

CDV reduction type algorithm units 45-1, 4
5-2 is the same in hardware, and the CDV reduction algorithm unit 45-2 performs an operation using the self-monitoring parameter set in the self-monitoring parameter setting unit 44 when the periodic cell arrives. In addition, it stores the processing result of each determination unit for each cell, and outputs the self-monitoring pattern according to the table of FIG. The CDV reduction algorithm unit 45-1 performs an operation on the arriving cells other than the periodic cells using the parameters for normal processing.

The operation pattern comparing section 46 compares the self-monitoring pattern with the expected value of the self-monitoring pattern. The algorithm normality judging section 47 judges whether the algorithm is normal by receiving the result of the match in the comparison processing in the operation pattern comparing section 46, and the algorithm abnormality judging section 48 receives the result of the mismatch in the comparison processing in the operation pattern comparing section 46. This is to judge whether an algorithm is abnormal and to notify an alarm. modulo5
The unit 49 updates the self-monitoring parameter with modulo 5.

Next, the operation of the embodiment of the present invention will be described with reference to FIGS. In FIG. 4, the normal policing operation described with reference to FIGS. 1 and 2 is performed in the CDV reduction algorithm unit 45-1 for an arrival cell that is not a periodic cell. When the periodic cell arrives, the self-monitoring of the CDV reduction algorithm according to the present invention is performed.

The calculation pattern of the CDV reduction algorithm includes shaping (to = TET-τ) shown in FIG.
s), TET remains as it is, to = ta, new-T
ET: to = ta, to> ta: bypass, to <ta
At ta, to = ta is updated and there are five bypass patterns. Therefore, a self-monitoring pattern k (k = 1 to 5) corresponding to each pattern is defined.

After the periodic cell is detected, a preset k = i
A self-monitoring theoretical transmission time TET (i), a self-monitoring residual CDV
Allowable value τs (i) and minimum cell interval Ts for self-monitoring
Read (i). The CDV reduction algorithm operation is performed using TET (i) instead of the theoretical transmission time TET of the periodic cell.

In the CDV reduction algorithm, a self-monitoring pattern k, which is a processing result of each judgment unit in each of the above-mentioned calculation patterns, is output and compared with a self-monitoring pattern expected value k (i). As a result of comparison, if they match, C
The DV reduction algorithm determines that the algorithm is normal, and if it does not match, notifies the apparatus management function of an alarm as abnormal.

As a result of the self-monitoring, when an abnormality is detected, shaping is stopped, and a through (bypass) process is performed for all input cells for which shaping is set.

[0024]

An embodiment using the set values of various monitoring parameters will be described below. For periodic cells, k = 1-5
Is used as an absolute time management method. FIG.
Is an embodiment of an absolute time management method in which ta = 0 to 720 as one cycle. The monitoring parameter values are set as shown in the table of FIG. Further, a self-monitoring pattern m (m = 1 to 9) is defined according to the table of FIG. In this embodiment, an arbitrary protection pattern is inserted between expected operations in order to easily calculate the self-monitoring parameters. As a result, m self-monitoring patterns are obtained.

The self-monitoring parameters TET (i), τs (i), T
Read s (i). Then, a CDV reduction algorithm operation is performed using TET (i) instead of the theoretical transmission time TET of the periodic cell. In this CDV reduction algorithm, when the self-monitoring pattern m, which is the processing result of each of the above-mentioned patterns to each of the determination units, is output according to FIG. 5, the table shown in FIG. 6 is obtained. However, if the next arriving cell is a non-user cell of the pattern, the "t
It is necessary to perform update processing of to so that a <to ”. By comparing the self-monitoring pattern m with the expected self-monitoring pattern m (i), the normal / abnormal of the CDV reduction algorithm circuit is determined. Make a decision.

[0026]

According to the present invention, the self-monitoring of the CDV reduction algorithm is performed using the arrival time of the periodic cell, so that it is not necessary to make the circuit redundant, and it is possible to suppress an increase in the circuit scale. Moreover, the five patterns so periodically monitored, it is possible to execution of probability to circuit abnormality detected immediately in the lower part.

Further, since non-data cells in the cell stream are used as cells for monitoring, self-monitoring of the CDV reduction algorithm can be performed while performing normal shaping operation in service.

[0028]

[Brief description of the drawings]

FIG. 1 is a diagram showing a processing flow of a CDV reduction algorithm.

FIG. 2 is a diagram for explaining an operation pattern of a CDV reduction type algorithm in FIG. 1;

FIG. 3 is a block diagram showing an embodiment of a CDV reduction algorithm circuit self-monitoring method according to the present invention.

FIG. 4 is a diagram showing a processing flow of a CDV reduction algorithm circuit self-monitoring method of the present invention.

FIG. 5 is a diagram showing a self-monitoring pattern.

FIG. 6 is a diagram illustrating an example of an operation of obtaining a self-monitoring pattern.

[Explanation of symbols]

 Reference Signs List 31 CDV reduction algorithm circuit 32 Algorithm monitoring unit 33 Parameter memory 41 Parameter initial setting unit 42 Cell arrival management unit 43 Periodic cell determination unit 44 Self-monitoring parameter setting unit 45-1, 45-2 CDV reduction algorithm unit 46 Operation pattern Comparison section 47 Algorithm normality determination section 48 Algorithm abnormality determination section 49 modulo 5 section

Claims (5)

(57) [Claims] 1. An operation pattern for a user cell
When TET> ta + τs, shaping (to
= TET−τs), and when TET ≦ ta + τs, TE
If T> ta, TET remains unchanged, and to = ta, TE
If T <ta, set ta = TET and to = ta, and
For the user cell, if to> ta, bypass, t
When o <ta, to = ta is updated and bypassed (however,
Where TET is the theoretical transmission time, ta is the cell arrival time,
τs is the allowable residual CDV value, and to is the ideal transmission time.
In a CDV reduction algorithm circuit having five types of operation patterns, a self-monitoring pattern k (k = 1 to 5 ) corresponding to the five types of operation patterns is defined, and a periodic cell whose arrival time can be predicted is determined. On the other hand, after the detection of the periodic cell, the CD
A V-reduction algorithm operation is performed, and the implemented self-monitoring pattern is compared with a self-monitoring pattern expected value corresponding to the set self-monitoring parameter. A self-monitoring method for a CDV reduction algorithm circuit, wherein an alarm is notified to a device management function as an error if the values do not match. 2. The self-monitoring method according to claim 1, wherein said five types of operation patterns are periodically monitored. 3. The self-monitoring method according to claim 1, wherein said periodic cells are non-data cells for control. 4. Policing (monitoring) by an ordinary CDV reduction algorithm for an arrival cell that is not a periodic cell
2. The self-monitoring method according to claim 1, wherein the operation is performed. 5. The self-monitoring parameter includes a self-monitoring parameter.
Theoretical sending time, CDV allowable value for self-monitoring
2. The CDV reduction algorithm circuit self-monitoring method according to claim 1, wherein the self-monitoring minimum cell interval is used.