JPH08214029A - Data redundancy broadcast method and device - Google Patents

Abstract

PURPOSE: To make immunity against a burst error compatible with a leveling of a flow with excellent balance when a data is broadcasted in a redundant way. CONSTITUTION: This device is provided with a means 14 to set a parameter (redundancy, unit block length, the number deviation blocks) specifying duplicate transmission order of a packet. A redundancy processing section 15 decides a transmission sequence of a packet based on the set parameter and a packet broadcast section 16 broadcasts the packet with an error code in the sequence decided by the redundancy processing section 15.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention enables a system that redundantly broadcasts data from a center to an unspecified number of terminals and that is robust against burst errors and has a leveled data flow. Data redundancy broadcasting method and device.

[0002]

2. Description of the Related Art FIG. 2 shows a configuration (1) of a conventional data redundant broadcasting apparatus. 21 is a data input unit, 22
Is a data storage unit, 23 is a packetization processing unit, 24 is a redundancy processing unit, and 25 is a packet broadcasting unit.

The data input to the data input section 21 is
It is stored in the data storage unit 22. Packetization processing unit 23
Divides the data in the data storage unit 22 into small units and adds an error detection code to each unit to form a packet, and sends the packet to the redundancy processing unit 24.
Upon receiving one packet, the redundancy processing section 24 duplicates the packet and sends the duplicated plurality of packets to the packet broadcasting section 25. The packet broadcasting unit 25 sequentially broadcasts the received packets. As a result, packets having the same content are collected and sequentially broadcast.

FIG. 3 shows the configuration (No. 2) of another conventional data redundant broadcasting apparatus. Reference numeral 31 is a data input unit, 32 is a data storage unit, 33 is a redundancy processing unit, 34 is a packetization processing unit, and 35 is a packet broadcasting unit.

The data input to the data input section 31 is
It is stored in the data storage unit 32. The redundancy processing unit 33
The data stored in the data storage unit 32 is read several times from the beginning to the end and sent to the packetization processing unit 34. The packetization processing unit 34 divides the received data into small units and adds an error detection code to packetize the data, and sends the packet to the packet broadcasting unit 35. The packet broadcasting unit 35 sequentially broadcasts the received packets. As a result, a series of data packets is circulated several times and broadcast.

[0006]

When transmitting data by broadcasting, it is not possible to make a request to retransmit the data when a data transfer error as used in one-to-one communication occurs. In addition, it is not possible to control the flow by a request to suspend or restart transmission when the receive buffer is full.

If all the duplicates of a particular packet are lost due to a burst error, the data cannot be restored on the receiving side. Therefore, from the viewpoint of resistance to burst errors, it is desirable that the plurality of duplicated packets be broadcast as temporally as possible.

On the other hand, if the time until the receiving buffer overflows due to the intensive arrival of the packet series is too short, the number of packets that can be received per one round of the packet series becomes small. Depending on the number of laps, the data cannot be restored. Therefore, from the viewpoint of giving the processing on the receiving side a margin, it is desirable that the packet sequence of packetized data is broadcast so that the arrival of the packet sequence is scattered as temporally as possible.

The conventional data redundant broadcasting apparatus (No. 1) shown in FIG. 2 has a problem in resistance to burst errors because a plurality of duplicates are broadcast together. On the other hand, in the conventional data redundant broadcasting device (No. 2) shown in FIG. 3, the packet sequence of packetized data is broadcasted in a consolidated manner, so that the flow is not leveled and the processing on the receiving side can be relaxed. could not. That is, in the conventional data redundant broadcasting device, it was not possible to achieve a good balance between resistance to burst errors and flow leveling.

It is an object of the present invention to solve the above-mentioned problems and to provide a broadcasting means capable of achieving a good balance between resistance to burst errors and flow leveling.

[0011]

According to the present invention, a means for registering a parameter for defining a duplicate transmission order of packets is provided, and further, a redundancy for determining how many packets having the same contents are transmitted as a parameter. Degree, the unit block length that indicates how many consecutively-numbered packets are sent, and the number of deviation blocks that indicate how many blocks in the series of packets to be transmitted are shifted before transmission. Then, the packet transmission order is determined based on these parameters, and the packets are broadcast in the determined order.

[0012]

According to the present invention, it becomes possible to optimally adjust the transmission order of redundant data according to the communication environment by the set parameters for defining the redundant transmission order. In particular, the parameter redundancy, the unit block length, and the number of deviation blocks introduced in the present invention act as follows.

Since the redundancy is a parameter for setting how many packets having the same contents are to be sent, increasing this value increases the resistance to burst errors. Since the unit block length indicates how many packet sequences are transferred in succession, packets having the same content are distributed and transferred by the unit block length. Therefore, the larger the unit block length is, the packets having the same contents are transferred in a distributed manner, and the resistance to burst error increases.

However, since the sequence of packets is packed and sent for each block, if the unit block length becomes too large, the uniformity of the flow may be impaired, and the processing on the receiving side may not be able to catch up. Therefore, the unit block length cannot take a very large value and must take a small value. However, when the unit block length is shortened, packets having the same content are broadcast closer in time, and the tolerance against burst errors is impaired. Therefore, sufficient resistance to burst errors cannot be secured simply by introducing the parameter of unit block length.

This can be supplemented by introducing the number of deviation blocks. If the number of deviation blocks is increased, packets with the same content will be distributed and transferred by the deviation. On the other hand, since the length of the packet sequence within a block does not change, flow leveling is not impaired.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 1 is a block diagram of a center for explaining an embodiment of the present invention. In the figure, 11 is a data input unit, 12 is a data storage unit, 13 is a packetization processing unit, 14 is a parameter setting unit, 15 is a redundancy processing unit, and 16 is a packet broadcasting unit. The redundancy processing unit 15 includes a packet string storage unit 151.
And a transmission packet determination unit 152.

The data to be broadcast is first of all the data input section 1
It is input from 1 and stored in the data storage unit 12. The stored data is decomposed by the packetization processing unit 13 into smaller units, an error detection code is added to each unit to be packetized, and stored in the packet string storage unit 151 of the redundancy processing unit 15. It

In the parameter setting section 14, three parameters of redundancy, unit block length and deviation block number are set. When the transmission packet determination unit 152 determines whether to broadcast a packet in the next step or to wait without broadcasting in the next step, based on the redundancy, the unit block length, and the number of deviation blocks set by the parameter setting unit 14, , The packet number to be broadcast is determined, and the packet is sent to the packet broadcasting unit 16. When waiting, decide the waiting time and wait for that amount. The packet broadcasting unit 16 broadcasts the packet received from the transmission packet determining unit 152.

FIG. 4 shows the redundant transfer of packets depending on the settings of the three parameters used in the embodiment of the present invention: (a) redundancy, (b) unit block length, and (c) deviation block count. It shows how the order changes.

When redundancy = 3, three packets having the same content are transmitted as shown in FIG. 4 (a). However, the same packet is not always transmitted continuously. For example,
When the unit block length = 5, as shown in FIG. 4 (b), five consecutively numbered packets are collected and transmitted three times.

Further, when the number of deviation blocks = 2, as shown in FIG. 4 (c), the transfer of each redundant block is shifted by two blocks and transmitted. At the timing corresponding to the empty block, the center does not broadcast the packet and simply waits.

If the redundancy is increased, more packets having the same content are transmitted, so that the resistance to burst errors is increased. However, if the time for broadcasting one packet from the center is fixed, the time until the redundant broadcasting of data is completed becomes a redundant number times the transfer time of the entire data. Therefore, the redundancy cannot be increased without limit.

When the redundancy is fixed, the resistance to burst errors increases as the unit block length increases. However, if the unit block length is made too large, flow leveling will be impaired. Therefore, it is necessary to set the unit block length that is balanced with the reception speed on the terminal side. On the other hand, even if the number of deviation blocks is increased, the unit block length remains the same, so the leveling degree of the flow is not impaired. Then, the packets having the same content are temporally dispersed and broadcast by the number of deviation blocks, so that the tolerance against burst errors can be enhanced.

FIG. 5 shows a specific setting example of parameters. In the case of FIG. 5 (a), the parameter is redundancy = 3.
The unit block length = 1 and the deviation block number = 1 are set. The packet sequence of 0123345678 (each numerical value is a packet number) is 0XX10X21 by the redundancy processing.
032143254365654765876X87XX8
Will be sent in order. The part of the symbol X is a part which waits without sending anything.

In the case of FIG. 5B, the parameter is redundancy =
3, unit block length = 1, deviation block number = 0,
This is a conventional data redundant broadcasting device (No. 1) shown in FIG.
It is the same packet sequence as. In the case of FIG. 5 (c), the parameters are redundancy = 3, unit block length = 9, deviation block count = 0, which is the same packet sequence as the conventional data redundant broadcasting device (part 2) shown in FIG. Become.

Depending on the parameter settings, as shown in FIG.
As shown in (c), since the packet sequence in the same order as that of the conventional method can be obtained, it can be understood that the redundant broadcasting system according to the present invention is a system including the conventional system. The redundant packet sequence obtained by the setting in Fig. 5 (a) is compared with Fig. 5 (b) and (c), and both the packets with the same content and the continuous sequence of packets are well-balanced and appropriately distributed in time. I understand that it will be done. Therefore, by appropriately setting the parameters according to the communication environment, it is possible to achieve a good balance between resistance to burst errors and leveling of the data flow.

FIG. 6 is a flow chart showing the procedure of the process for determining the packet number of the packet to be broadcast in each step based on the redundancy, the unit block length, and the deviation block number in this embodiment. This flow chart also includes the procedure of standby processing for waiting an appropriate time when no packet is sent.

The variable N stores the total number of packets when all the data stored in the data storage unit 12 is packetized. Nb represents the number of blocks when the packet series is grouped by the number of unit block lengths. We also consider the case where the last block consists of fractional packets. i is a variable for tracing the block row in the horizontal direction in FIG. j is a variable for tracing the block sequence in the vertical direction in FIG. 4 (c).
k is a variable for tracing the sequence of packets in one block.

First, in step 601, variables N, B,
After initializing the values in Z, H, and Nb, the transmission process is started from the block of i = 0 and j = 0. i is Nb + (Z-
1) * H, and when the transmission of all blocks is completed, the transmission process is terminated. When j becomes the redundancy Z, 1 is added to i, and the same processing is continued from j = 0 to (Z-1) (steps 602 to 606).

It is judged whether the block is transmitted or waited (step 607). If the block is transmitted, the steps 608- are executed.
At 613, a packet for the unit block length is transmitted. After the transmission, 1 is added to j in step 617, and the process proceeds to the transmission process of the next packet series. Step 61
1, 614 and 615 are processes when the last block is composed of fractional packets.

If it is determined to be in standby in step 607, nothing is transmitted for the time corresponding to the B packet transmission time (step 616), and j is set to 1 in step 617.
Is added to proceed to the transmission processing of the next packet series.

FIG. 7 is a block diagram of the receiving terminal side for explaining the embodiment of the present invention. 71 is a packet receiving unit, 72 is an error detecting unit, 73 is a packet aligning unit, 74 is a data restoring unit,
Reference numeral 75 represents a data storage unit.

Upon receiving the broadcast packet, the packet receiving section 71 sends the packet to the error detecting section 72. The error detection unit 72 checks whether or not the received packet has an error, and discards the packet if an error is detected. If no error is detected, the packet is sent to the packet alignment unit 73. The packet sorting unit 73 sorts the received packets in the order of the packet numbers so that there is no overlap and there is no omission, and sends the sequence of the sorted packets to the data restoration unit 74. Data restoration unit 7
4 restores the data from the received packet and stores it in the data storage unit 75. The data stored in the data storage unit 75 is delivered to the outside as needed after all the data are collected.

FIG. 8 is a diagram for explaining how a packet that arrives intricately is stored in an array buf of size S and is output. The packet number p is given to the arrived packet. indm ← p mod
Let S, indd ← p div S. The variables bm and bd are
The values of indm and indd of the packet to be output next. Before the next packet to be output does not arrive, the buf is circulated and the range of packets that can be stored is stored.

That is, ((indm ≧ bm) and (i
ndd = bd)) or ((indm <bm) and (in
When a packet satisfying the condition of (dd = bd + 1)) arrives, the packet is stored in buf [indm]. Also, (indm = bm) and (indd = b
When the packet of d) arrives, at that time, bu
All the packets stored in f that are continuous are output. After output, the values of bm and bd are updated.

FIG. 9 is a flow chart for explaining the above procedure, that is, the procedure on the terminal side for arranging and outputting packets that arrive in a complicated manner. The array flag [i] is a flag that is set to "1" when the content of the unoutput received packet is stored in the array buf [i], and is set to "0" in other cases. After the necessary variables and arrays are initialized in step 901, the reception of a packet or the detection of a timeout is awaited in step 902. If no packet is received for a predetermined time or more and a time-out occurs, the process ends (step 903).

When a packet is received, indm and indd are calculated from the packet number p and the array size S (step 904). If the packet is the first received packet, and if the packet can be stored by circulating buf. , The content of the received packet is buf
Write to [indm] and set flag [indm] to "1" (steps 905 to 907). If indm is the value bm of the packet to be output next, buf
All the packets stored after the bm-th position of are consecutively output. If the flag is "0" and the packet is output up to the area before being stored,
bm is updated to a value indicating the position of the unstored area, and the arrival of the next packet is waited for. In order to circulate and use buf, bd is updated after the end of buf is output (steps 908 to 916).

In the packet receiving procedure of the terminal shown in FIG. 9, it is not necessary for the terminal side to know the value of the parameter set at the center. Therefore, it is possible to flexibly set the parameters on the center side without making any changes on the terminal side. Furthermore, by sending information about the quality of the broadcasting line from some terminals via the public line,
It is also possible to adaptively change the parameters on the center side. This can be easily inferred from the embodiment described above.

[0039]

According to the present invention, the redundant transmission order of packets can be defined by adjusting the redundancy, the block length, and the number of deviation blocks as parameters on the center side. Therefore, it is possible to achieve a good balance between resistance to burst errors, which requires packets of the same content to be dispersed in time as much as possible, and leveling of data flows, which requires that packet sequences be dispersed in time as much as possible. ， Can perform redundant broadcasting.

[Brief description of drawings]

FIG. 1 is a configuration diagram on a center side for explaining an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration (No. 1) of a conventional data redundant broadcasting device.

FIG. 3 is a diagram showing a configuration (No. 2) of a conventional data redundant broadcasting device.

FIG. 4 is an explanatory diagram of parameters defining a redundant transfer order of packets in the embodiment of the present invention.

FIG. 5 is a diagram showing a specific setting example of parameters.

FIG. 6 is a center side flowchart in the embodiment of the present invention.

FIG. 7 is a configuration diagram on the receiving terminal side for explaining an embodiment of the present invention.

FIG. 8 is a diagram illustrating a packet sorting process on the terminal side.

FIG. 9 is a flowchart illustrating a procedure on the terminal side.

[Explanation of symbols]

 11 data input unit 12 data storage unit 13 packetization processing unit 14 parameter setting unit 15 redundancy processing unit 151 packet sequence storage unit 152 transmission packet determination unit 16 packet broadcasting unit

Front Page Continuation (72) Inventor Harushi Tsukada 1-1-6 Uchisaiwaicho, Chiyoda-ku, Tokyo Nihon Telegraph and Telephone Corporation

Claims (4)

[Claims] 1. A data redundant broadcasting method for redundantly broadcasting packetized data from a center, registering parameters for defining redundant transmission order of packets, and storing data to be transmitted. Decomposing the data into several small chunks, adding an error detection code to each chunk and packetizing, and determining the packet transmission order based on the parameters defining the redundant transmission order. And a step of broadcasting a packet with an error detection code based on the determined transmission order, a data redundant broadcasting method. 2. A unit block representing redundancy for determining how many packets of the same content are to be transmitted, and how many consecutively-numbered packets are to be transmitted, as parameters for defining the duplicate transmission order of the packets. 2. The length and the number of deviation blocks indicating how many blocks in a sequence of packets to be transmitted are shifted before transmission are used.
Redundant data broadcasting method described. 3. Data storage means for storing data to be sent, and data stored in the data storage means are decomposed into several small chunks and an error detection code is added to each chunk to packetize. Packetizing processing means, packet string storing means for storing packetized data, parameter setting means for setting parameters that define redundant transmission order of designated packets, and parameters set by the parameter setting means Transmission packet deciding means for deciding the transmission order of packets based on the above, and packet broadcasting means for broadcasting packets with error detection code based on the transmission order decided by the transmission packet deciding means. Redundant data broadcasting equipment. 4. The parameter for defining the redundant transmission order of packets set by the parameter setting means is redundancy for determining how many packets having the same content are to be transmitted, and how many consecutive packets are provided. 4. The data redundancy broadcasting apparatus according to claim 3, wherein the unit redundant block length indicates the number of blocks to be transmitted, and the number of deviation blocks indicates how many blocks in a series of packets to be transmitted are shifted.