JPH01143439A - Packet exchange type lan - Google Patents

Abstract

PURPOSE:To decrease the time for one frame and a repetitive cycle by providing a means sending a deputy packet of a station in system-down when one station is in system-down and a means moving up the packet transmission timing of stations after said station in system-down. CONSTITUTION:With a station (n) sending a packet, a packet transmission monitoring section 7 monitors the packet transmission state and when the station (n-1) sends a packet, the station (n) sends the packet after a time till said packet reaches the entire network is awaited. If the station (n-1) does not send any packet till a time to send the packet is arrived, the station (n) sends the packet informing the system-down of the station (n-1) and decrements the logical number by one after the packet is sent. A station receiving a deputy packet sends the packet in the present packet transmission timing and only when the logic number of the station in system-down is smaller than its own logic number, the logic number of its own station is decremented. Then the station whose logic number is 0 decrements the packet transmission monitoring logic number by one.

Description

[Detailed Description of the Invention] The present invention is based on a packet switching type (C5MA/CD method)% type% In such a LAN, a plurality of stations connected to the network each contain information. The predetermined packets are sent sequentially. However, not all stations connected to the network always transmit packets, and some stations do not participate in the network due to failure or other reasons. As a process when such non-participation occurs, Japanese Patent Laid-Open No. 10342/1983 discloses that the master station sends out a proxy packet in place of the non-participating station. That is, in this conventional example, the entire transmission time (time for one frame) is fixed regardless of the presence or absence of non-participation, and the master station fills in the gaps.

As mentioned above, in the case of the conventional example, even if there is a non-participating station (hereinafter referred to as "down"), the packet sending time of this down station is determined by the parent station. Because of this, the l-frame time is always fixed, which has the disadvantage of being inefficient.

In addition, in the case of voice data transmission where no delay is allowed in -i, it is desirable to minimize the delay time, but in the conventional example above, the transmission time of the l frame is constant, so the delay time is shorter than that. It won't happen.

The present invention has been made in view of these points, and an object of the present invention is to provide a packet-switched LAN that can shorten the overall transmission time (one frame time) in the event that a station goes down. With the goal.

Means for Solving the Problems In order to achieve the above object, the present invention provides a packet-switched LAN in which a plurality of stations are connected to a network and packets are sent sequentially from the plurality of stations.
The present invention is configured to include means for transmitting a proxy packet for a station that has gone down when one station goes down, and means for advancing the packet transmission timing of a station other than the failed station.

According to this configuration, when a station goes down, the downtime is transmitted to the network by a proxy packet. Accordingly, the means for advancing the sending timing is operated, and the packet sending timings of the stations after the down station are sequentially advanced, so that the time of one frame and the repetition cycle are shortened by the amount of the down station. If other down stations occur, the one frame time and repetition cycle will be further shortened by similar operations.

Execution - Outside of construction As shown in Figure 2, the network (NT) has multiple stations (STo), (STI), (Si2),
(Si3).

... are connected, but sequential numbers are assigned in advance to these stations, and packets are sent out in the order of the numbers. Although FIG. 1 specifically shows one of the stations, it may be assumed that the other stations have almost the same configuration.

In FIG. 1, (1) is a transmitting/receiving section that directly exchanges signals with the network (NT), and (2) is an encoder/decoder section. (3) is a link management unit that mainly sends and receives data from adjacent stations, (4) is a data frame assembly unit, and (5) is a packet generator consisting of a logical station number management unit (6) and a packet transmission monitoring unit (7). An interval control section, (8) a terminal interface, and (9) a terminal device.

As shown in Figure 3, the data frame consists of a group of packets (P) from stations participating in the network.
repeated many times. Each packet contains signals (al) to (a
, ), of which (al) is a preamble for synchronization, (a2) is a destination address, (a,) is a source address, and (a4) is a type indicating the type of packet.

There are three types of packets: a proxy packet, a new packet, and a normal packet. (a,) is 1 blue report data, which is data such as audio data, image data, or computer files. (a6) is a check bit as an error correction code.

Each station connected to the network (NT) has an absolute number as shown in FIG. 2 and a logical number managed by a logical station number management section (6). The absolute number is determined when the station is connected to the network (NT).

Furthermore, the logical number changes depending on the operating status of the stations connected to the network (NT). Normally, the absolute number and the logical number are the same, and the packet transmission order of a station is determined by the logical number. In other words, when station n sends a packet, the packet sending monitoring unit (7) monitors the packet sending status, and when station n-1 sends a packet, station n waits for the packet to reach the entire network. Send a packet. If the n-1 station does not send a packet when it is time to send a packet, the n-1 station immediately sends a packet informing that the n-1 station is down. Then, after station n transmits the packet at the timing at which it transmits the packet, it decrements the logical number by one. The station that received the proxy packet informing that station n-1 is down sends a packet at the current packet sending timing, and then determines that the logical number (n-1) of the down station is smaller than its own logical number. Only when you freeze one of your logical numbers. Then, the station whose logical number is O decreases its packet sending monitoring logical number by one.

The packet transmission monitoring logical number of a station other than logical number 0 is (
own logical number - 1), and the logical station number management unit (
6).

The above operation will be explained according to the flowchart of FIG. 4. First, in step (#l), the n-station
One station determines whether it is time to send a packet. next,
In (#2), it is determined whether the n-1 station has issued a packet at that time. The packet transmission monitoring unit (7) of station n monitors and determines the packet transmission of this n-1 station.
is mainly carried out.

Here, if the n-1 station is sending a packet, the determination result is YES and the operation flow returns to (I1).

If station n-1 has not sent a packet at the above time, (+
Proceeding to step 13), station n sends out a proxy packet within the above-mentioned time period. In this case, the destination address (I2) of the sent packet is all stations, and the source address (I3) is n
It is a station. Type (I4) is a proxy packet. After the time period ends in (I4), station n sends its own packet at its own packet sending time. After that, in (115), all stations determine whether their own logical number is O, and if it is O, they skip to (118) and manage it in their own logical station number management section (6). Decrement the packet monitoring logical number by one.

The station whose answer is No in (I5) judges whether the logical number of the downed station is smaller than its own logical number in the next step (I6). If the answer is NO in (I6), the process skips to (I9) and changes its own packet sending timing. If YES, in (I7) the logical number managed by its own logical number management unit (6) is decreased by one, and then the process proceeds to (118), where the transmission monitoring logical number is also decreased by one. Thereafter, in (I9), the packet sending timing of itself is changed. Here, changing the packet sending timing is done because the time from the previous sending timing to the current sending timing is known.
This can be achieved by starting packet transmission at a time obtained by subtracting the transmission time of one station from that time.

In order to make the operation of the flowchart in Fig. 4 easier to understand, we will explain the above by assuming that there are 100 operating stations connected to the network (NT) from 0 to 99 and the 60th station goes down. The n-1 station is the 60th station,
Station n becomes the 61st station. In (I5), the transmission monitoring logical number of station 0 was 99 because the previous station was 99, but it is changed to 98 in (I8).

On the other hand, stations other than the 0 station and stations 1 to 59 change only their own packet sending timing at (I9) without changing their own logical number and sending monitoring logical number.

Stations 61-99 reduce their own logical numbers by one in (I7) to become 60-98. At (I8), the transmission monitoring logic number is decreased by one from 60-98 to 59-97. For example, the 65th station sets its own logical number to 64 and the transmission monitoring logical number to 63. Stations 0 and 61 to 99 also advance their own packet sending timings by one in step (I9).

In the above embodiment, all stations connected to the network (NT) have a packet sending monitoring unit (7), and each station sends a substitute packet and monitors logical numbers when the previous station goes down. However, the packet transmission monitoring unit (7) of each station is not used or abolished, and a specific one or the master station has the monitoring function and proxy packet transmission function. You can also leave it up to the station. However, even in this case, the packet transmission timings of stations after the downed station are sequentially advanced.

According to the present invention, if a station goes down, the packet transmission timing of subsequent stations is sequentially advanced, so that the total packet transmission time is shortened and the repetition cycle becomes faster.

Accordingly, when the information incorporated into the packet is audio data, the effect of reducing the delay of the audio signal can also be expected.

[Brief explanation of the drawing]

The figures all relate to a packet-switched LAN in which the present invention is implemented; FIG. 1 is a block diagram of one station, and FIG. 2 is a schematic diagram showing a plurality of stations connected to the network. FIG. 3 is a diagram showing the format of the packet, etc., and FIG. 4 is a flowchart of the operation. (NT)--Network. (Sr1) (STri) (ST3) (ST-) --Station. (P) -Packet.

Claims (1)

[Claims] (1) In a packet-switched LAN in which a plurality of stations are connected to a network and packets are sequentially sent out from the plurality of stations, when one station goes down, a means for sending out a proxy packet for the downed station; 1. A packet-switched LAN characterized by having means for advancing the packet transmission timing of stations after the downed station.