JP3734606B2 - Ring network system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to frequency deviation compensation in a ring network.
[0002]
[Prior art]
As one of the network shapes, there is a ring (loop) network represented by a token ring network defined in IEEE 802.5.
[0003]
In a ring network, each station is physically arranged on a ring-shaped transmission line, and the station receives a frame flowing in a certain direction from the upstream station. If the address of the received frame is not its own, the receiving station regenerates the frame and transmits it to the next downstream station. In this way, in a ring network, each station sequentially relays frames and circulates in the transmission path to perform data communication.
[0004]
In this ring network, when receiving a data frame sent from an upstream station, the receiving station samples and captures data using the sampling clock generated by its own clock, and after shaping the waveform, sends its own transmission clock. Use to send to downstream station.
[0005]
In this manner, in a ring network, each station extracts data from a data frame having the clock timing of the upstream station at the timing according to the clock of the receiving station. Therefore, since the transmission speed of the network is constant, the clocks of the stations must all have the same frequency.
[0006]
However, the clock of each station generally has a frequency variation (frequency deviation) of about 50 to 100 ppm (parts per million: 1/1 million) with respect to a specified frequency due to its temperature environment or the like.
[0007]
Therefore, even if all stations use the same specified frequency clock for data communication, the clock used by one station in the network is actually different from the clock used by another station. It is possible. Therefore, even when the first bit of the transmission frame is transferred at the same timing when the frame is relayed, the frequency gradually shifts between transmission and reception due to the frequency deviation. If the magnitude of the deviation exceeds the 1-bit length of the data frame, unnecessary addition (overrun) or omission (underrun) occurs in the frame data.
[0008]
As a method for solving this, there is generally used a method in which a buffer for deviation compensation by a plurality of 1-bit registers is provided and a data frame is buffered to compensate for a clock frequency deviation between transmission and reception.
[0009]
FIG. 5 is a block diagram showing a data transfer process in each station using a deviation compensation buffer. In FIG. 5, a data receiving unit 101 receives data from an upstream station of the ring network. The received data includes a clock component such as a Manchester code, and the data receiving unit 101 samples the received data with a clock having a frequency several times the transmission clock, extracts the data and the transmission clock, and stores them in the deviation compensation buffer 102. Store. The deviation compensation buffer 102 is a multi-stage 1-bit buffer composed of a plurality of 1-bit registers. The data transmission unit 103 extracts data from the buffer 102, uses this to perform Manchester encoding with a transmission clock to form a frame, and transmits the frame to a downstream station of the network.
[0010]
FIG. 6 is a diagram for explaining a frequency deviation compensation method by the deviation compensation buffer 102.
In FIG. 6, in order to cope with overrun / underrun of 2 bits at maximum, a buffer having a capacity of 5 bits is used.
[0011]
The frame data received from the network is stored in the buffer 102 at the speed of the transmission clock from the upstream station, and the received data is stored in stages 2, 3,... Next, it will be cyclically stored in the first stage again.
[0012]
On the other hand, when the data storage side stores the data in the third-stage buffer, the data is extracted from the buffer 102 by the data transmitting unit 103 by the FIFO (First In First Out) method and operating with its own clock. The data from the first stage to the second stage, the third stage,... Are taken out in the order of storage, and then the fifth stage is taken out from the first stage again.
[0013]
In this way, if the clock shift becomes 1 bit length or more normally, the loss of bits or the like may occur. By providing a buffer as shown in FIG. The frequency deviation can be compensated for up to overrun / underrun, and long data can be sent continuously.
[0014]
[Problems to be solved by the invention]
The deviation due to the frequency deviation during the data transmission is larger than the size of the transmission frame. Therefore, in the method using the buffer for compensating the frequency deviation of the clock, the number of stages of the buffer is determined so that the frame having the maximum data length can be transmitted among the frames flowing through the network. That is, the number of stages of the buffer is determined based on the data length (maximum data length) of the longest frame transmitted over the network, assuming that the maximum frequency deviation is constant. This maximum data length is the largest of the frames used by various systems using the network.
[0015]
By the way, the transmitted frame causes a transmission delay every time it passes through the buffer for compensating the clock frequency deviation of each station, and this delay increases as the number of stages of buffers in each station increases. For example, when the number of buffer stages is determined corresponding to a maximum shift of 2 bits as shown in FIG. 6, the transmission frame is delayed by 2 clocks for each station by this buffer. Thus, the transmission delay increases as the number of buffer stages increases.
[0016]
As the maximum data length is increased, the number of buffer stages in each station is increased. As the number of buffer stages is increased, the delay generated in each station is increased, and the transmission efficiency of the network is deteriorated. In particular, even in a system that does not require a data length that is much longer than the maximum data length of the network, the number of buffer stages becomes the number of stages defined by the network, and transmission efficiency is reduced more than necessary.
[0017]
The present invention has been made in view of the above points, and in a ring network used as one shape of a small-scale network among connection networks of process input / output devices, etc., a system that uses a network for the number of buffer stages It is an object of the present invention to provide a network system and a method thereof that can improve the transmission efficiency of the network by making it variable.
[0018]
[Means for Solving the Problems]
The network system according to the present invention relates to a ring network system having one master station and one or more slave stations. The master station has maximum length notification means, and the slave station has buffer means and stage number setting means.
[0019]
The maximum length notification means generates and transmits a frame including information based on the maximum length of the frame used by the system using the ring network.
The buffer means buffers received data from the network.
[0020]
The stage number setting means sets the number of stages of the buffer means based on the value of the maximum frame length of the notification frame.
In another embodiment of the present invention, a station in the ring network has a frame length measuring unit.
[0021]
The frame length measuring means measures the length of the frame passing through the own station. The stage number setting means sets the stage number of the buffer means based on the longest frame length measured by the frame length measuring means within a certain period.
[0022]
According to the present invention, each station determines the number of stages of its own buffer means based on information based on the maximum length of the frame used by the system using the ring network from the master station or on the measurement result by the frame length measuring means. Therefore, the delay caused by the buffer means of the transmission frame flowing through the network can be reduced.
[0023]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, as an embodiment of the present invention, several stations to several hundreds of stations used in FA (Factory Automation) and the like, a network length of several tens to several hundreds of meters, and a transmission frequency of several M to several tens of MHz are used. A case where it is applied to a LAN will be described.
[0024]
FIG. 1 is a diagram for explaining a schematic diagram of a ring network system in the first embodiment.
In the configuration of FIGS. 1 and 2, an arbitrary number of stations a, b,... M are connected on the ring network 10, and the station a is a master station that manages the network, and the stations b to m are slaves. Bureau. The arrow on the ring-coupled network 10 indicates the direction in which the frame data flows. The slave station closest to the master station a on the network 10 is the station b, the next station is the station c,. . . , The station that is most downstream is the station m. Since the network 10 is a ring network, the master station a receives a frame from the slave station m at the most downstream side.
[0025]
The stations a to m are each provided with a frequency deviation compensating buffer whose number of stages to be described later can be changed. When the network is started up, each station sets the number of stages of this buffer to a value based on the data length (maximum data length) of the longest frame transmitted over the network as in the conventional case.
[0026]
When the network system is started up, the master station a performs initialization processing of the network system, for example, setting of network status such as station number setting / collection and information collection.
[0027]
As one of the initialization processes, the master station uses the network 11 as information to set the stage values of the frequency compensation buffer of the stations b to m based on the performance and state of the system using the network. Send to.
[0028]
The information of the frame 11 includes, for example, information indicating the maximum length of a frame used by a system that uses a network, or the type and state of the system. In this case, the stage value is obtained for each station using the information in the frame 11. Alternatively, the stage value obtained by the station a can be transmitted as the frame 11, and each station can set it as the number of stages of its own buffer.
[0029]
Upon receiving this frame 11, the stations b to m take out the above information, set the number of stages of their frequency deviation compensation buffer based on the information, and transmit the frame 11 to the downstream station.
[0030]
The frame 11 makes a round in the network 10 and is received by the station a when the setting of the number of buffer stages in the stations b to m is completed. As a result, the station a recognizes that the setting of the number of buffer stages is completed in the stations b to m in the network.
[0031]
FIG. 2 is a block configuration diagram centering on a buffer for deviation compensation in each station in the first embodiment.
In FIG. 2, the data receiving unit 1 receives data from an upstream station of the ring network, samples the received data with a clock having a frequency several times the transmission clock of the network 10, and extracts the data and the transmission clock. And stored in the deviation compensation buffer 2. Further, the data transmission unit 3 takes out the data from the frequency deviation compensation buffer 2, regenerates the frame with its own clock, and sends it to the downstream station. The data receiving unit 1 and the data transmitting unit 3 are obtained by adding functions to the buffer stage number setting unit 4 described later to the data receiving unit 101 and the data transmitting unit 103 of FIG.
[0032]
The frequency deviation compensation buffer 2 is a multi-stage 1-bit buffer composed of a plurality of 1-bit registers, and is large enough to handle a frequency deviation in a frame having the longest data length (maximum data length) transmitted over the network. If the number of necessary buffer stages obtained based on the maximum data length is n, the capacity is 2n + 1 or more. The data storage format in the buffer 2 is a cyclic format in which the data is stored in order from the first level, the next level is the first level, and the data extraction is in the FIFO format.
[0033]
The buffer stage number setting unit 4 determines the stage value of the frequency deviation compensation buffer 2 based on the information from the station a, and controls the buffer 2 and the data transmission unit 3. For example, when the information given in the frame 11 from the station a is the stage value itself set in the buffer, the buffer stage number setting unit 4 uses the value of the frame used in the system in which the information from the station a uses the network. In the case of information indicating the maximum length or system type, status, etc., the number of stages to be set is calculated using the information, or the information obtained from the value obtained in advance and stored in the buffer stage number setting unit 4 The buffer 2 and the data transmission unit 3 are controlled based on the selection.
[0034]
When the data receiving unit 1 receives the frame 11 from the station a from the network 10, it notifies the buffer stage number setting unit 4 of the information in the frame 11. In response to this, the buffer stage number setting unit 4 determines the value of the stage number of the frequency deviation compensation buffer 2 from the values calculated by using this information or by decoding the information. Thereafter, the buffer stage number setting unit 4 controls the buffer 2 and the data transmission unit 3 so that the frequency deviation compensation buffer 2 functions as a buffer having this stage number.
[0035]
FIG. 3 is a diagram for explaining setting of the number of stages in the frequency deviation compensation buffer 2.
The number of stages of the frequency deviation compensation buffer 2 of each station is the number of stages corresponding to the data length (maximum data length) of the longest frame transmitted over the network as an initial setting. When the maximum data length is transmitted, a frequency deviation of a maximum of n bits may occur, and the number of stages of the buffer 2 is n.
[0036]
FIG. 3A is a diagram showing the setting of the number of stages in the initial state of the buffer 2.
Since the buffer 2 is used to compensate for frequency deviation, the number of stages of the buffer 2 here does not indicate the capacity of the buffer 2 itself, but the number of stages of data stored in the buffer 2 by the data receiving unit 1. It indicates whether the data transmission unit 3 takes out the data from the buffer 2 from the time when the data is sent. As shown in FIG. 3A, when data is stored up to the nth stage of the buffer 2 by the data receiving unit 1, the data transmitting unit 3 starts to extract the stored data, so that the buffer 2 has a maximum of n bits. Can handle underflow / overflow.
[0037]
Next, as for the number of stages of the frequency deviation compensation buffer 2, the frame 11 is transmitted from the station a, and in response to this, the buffer stage number setting unit 4 determines the value of the number of stages of the frequency deviation compensation buffer 2. When this value is k, k ≦ n.
[0038]
FIG. 3B is a diagram illustrating a state after setting the number of stages of the buffer 2 by the buffer stage number setting unit 4.
When the buffer stage number setting unit 4 receives the information in the frame 11 from the data receiving unit 1, the buffer stage number setting unit 4 determines the stage value k of the frequency deviation compensation buffer 2 based on the information. Thereafter, the data buffer stage number setting unit 4 controls the buffer 2 and the data transmission unit 3, and when the data transmission unit 3 takes out the stored data from the buffer 2, the data reception unit 1 transfers the data to the n stages of the buffer 2. It changes from the time of storing up to the eye to the time of storing up to the k-th stage.
[0039]
As a result, the buffer 2 operates as a k-stage frequency deviation compensation buffer that can cope with underflow / overflow of up to k bits. In addition, the transmission frame has a delay of n clocks in the case of FIG. 3A due to the buffer 2, but is a delay of k clocks in the case of FIG. Is minimized.
[0040]
As described above, in the first embodiment, based on the information from the master station a, the buffer stage number setting unit 4 of the stations b to m controls the frequency deviation compensating buffer 2 and the data transmission unit 3 to control the data transmission unit. 3 controls the timing at which the stored data starts to be fetched, thereby changing the number of stages of the buffer 2 to a value optimized for the system using the network. As a result, it is possible to prevent a decrease in transmission efficiency due to an unnecessary transmission delay in the buffer 2.
[0041]
Next, a second embodiment will be described.
In the second embodiment, the number of stages of the buffer is not determined by the information from the master station as in the first embodiment, but the transmission frame that passes through the own station to each station, that is, the transmission frame that flows on the ring network. A circuit for measuring the data length is provided, and each station sets the number of stages of its own frequency deviation compensation buffer using the result of measurement over a certain period.
[0042]
In the second embodiment, each station first sets the number of buffer stages for frequency deviation compensation based on the data length (maximum data length) of the longest frame transmitted over the network as in the first embodiment as an initial setting. n.
[0043]
After the network system is started up, each station measures the length of the transmission frame passing through its own station for a certain period, and stores the maximum value of the length of the frame passing through that station during that period.
[0044]
FIG. 4 is a schematic block diagram for realizing the second embodiment.
As a configuration for determining the maximum value of the length of the passing frame and storing the maximum value, the maximum value of the previous measured value is stored in a register or the like, and when the new measurement is completed, the value up to the previous value is stored. Compared with the measured value of this time, it can be realized by a general method such as holding a large value in a register as a maximum value.
[0045]
In FIG. 4, first, the data receiving unit 21 stores a frame received from the ring network 24 as received data in the frequency deviation compensation buffer 22 at the timing of the received frame storage signal 30 after sampling. Further, the data receiving unit 26 raises the frame head / end notification signal 29 and notifies the frame data length counter circuit 25 of the head of the frame. In response to this, the frame data length counter circuit 25 starts counting the frame storage signal 30.
[0046]
When the reception of one frame is completed, the data receiving unit 23 causes the frame head length / end notification signal 29 to fall and notifies the frame data length counter circuit 25 of it. In response to this, the frame data length counter circuit 25 finishes counting the frame length by the frame storage signal 30 and outputs the count value to the frame comparison circuit 27 as a measurement result. The frame comparison circuit 27 compares the frame length 33 of the frame received this time with the maximum frame length 34 of the previous measurement stored in the frame maximum data length storage register 26, and determines the larger value 35. Store in the register 26.
[0047]
When the period for repeatedly measuring these operations for a certain period is over, the frame length stored in the frame maximum data length storage register 26 is set as the maximum frame length value 36 of the frame transmitted on the network 24, and the buffer stage number conversion circuit 28 is notified.
[0048]
Based on the maximum frame length value 36, the buffer stage number conversion circuit 28 obtains the stage number k of the buffer 22 necessary for performing frequency deviation compensation of the frame transmitted through the network 24, and uses this as the buffer stage number signal 32 as the buffer 22. Notify
[0049]
Thereafter, when the data of the number k of stages notified by the buffer stage conversion circuit 28 is stored in the buffer 22 in the data receiving unit 21, the buffer 22 notifies the data transmitting unit 23 by the frame extraction signal 31. The data transmission unit 23 extracts the data from the buffer 22 by the frame extraction signal 31, starts regenerating the frame, and transmits the frame to the network 24.
[0050]
In this way, the configuration for realizing the second embodiment is such that the circuit itself for measuring the frame length recognizes the beginning of the frame and counts the data frame bit length from the timing until the end of the frame is recognized, It is a general circuit such as a comparison circuit for comparing whether or not the frame length is maximum every time the measurement of the frame length is completed, and a storage register for storing the maximum frame length, and can be easily realized.
[0051]
As described above, in the second embodiment, each station is provided with a configuration for measuring the data length of a frame passing through its own station within a certain period, and the data transmission unit is based on the measurement result as in the first embodiment. By setting the number of buffer stages by determining the timing for extracting data from the buffer, it is possible to prevent a decrease in transmission efficiency due to an unnecessary station delay in the buffer.
[0052]
In the first and second actual method methods, when the maximum frame length of the frame used by the system using the network does not change, the buffer stage number is set to an optimum value once at the time of starting the system. If there is no problem. However, in the case of a system in which the maximum frame length may vary during the process, the number of buffer stages cannot be fixed at the start of the network system. The third embodiment corresponds to such a case.
[0053]
In the third embodiment, in the second embodiment, the measurement of the frame length passing through the own station is not performed only at the time of starting the system as in the second embodiment, but at regular intervals, The buffer stage number k is corrected according to the measurement result.
[0054]
By doing this, the optimum number of stages of frequency deviation compensation buffers is always set for the frame length used by the system, and there is no more delay in the station than necessary, thereby preventing a decrease in transmission efficiency. .
[0055]
In the first embodiment as well, information for setting the stage value of the frequency compensation buffer by the master station a is set as a frame 11 at regular intervals or whenever the maximum length of a frame flowing in the network changes. The same effect can be obtained by issuing and notifying the slave stations b to m.
[0056]
【The invention's effect】
As described above, according to the present invention, the stage value of the buffer for frequency deviation compensation can be varied depending on the performance and state of the system in which the network is used. Efficiency can be improved.
[0057]
In addition, even when the maximum length of the frame used in the process of the network system is variable, the optimum number of buffer stages for frequency deviation compensation is always set for the frame length used by the system. As described above, there is no delay in the station, and a reduction in transmission efficiency can be prevented.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a schematic diagram of a ring network system according to a first embodiment.
FIG. 2 is a block configuration diagram centering on a deviation compensation buffer of each station in the first embodiment;
FIG. 3 is a diagram for explaining a frequency deviation compensation method using a deviation compensation buffer;
FIG. 4 is a block diagram of a configuration for realizing a second embodiment.
FIG. 5 is a block diagram showing a data transfer process in each station using a buffer for deviation compensation.
FIG. 6 is a diagram illustrating a frequency deviation compensation method by the deviation compensation buffer 102;
[Explanation of symbols]
1, 21 Data reception unit 2, 22 Frequency deviation compensation buffer 3, 23 Data transmission unit 4 Buffer stage number setting unit 10, 24 Ring network 11 Information for setting stage value frame 25 Frame data length counter circuit 26 Frame Maximum data length storage register 27 Frame length comparison circuit 28 Frame length buffer stage conversion circuit

Claims (8)

In a ring network system having one master station and one or more slave stations,
The master station
A maximum length notification means for generating and transmitting a frame including information based on a maximum length of a frame used by a system using the ring network;
The slave station
Buffer means for buffering received data from the network;
A stage number setting means for setting the number of stages of the buffer means based on information based on the maximum length of the frame included in the frame ;
A ring network system comprising:   2. The ring network system according to claim 1, wherein the stage number setting means sets the stage number of the buffer means by determining a timing for extracting stored data from the buffer means for transmission frame generation. A ring-type network station device, a frame length measuring means for measuring the length of a frame passing through the own station;
Buffer means for buffering received data from the network;
A stage number setting means for setting the number of stages of the buffer means based on the longest frame length measured by the frame length measurement means within a fixed period;
A ring network station apparatus comprising:   4. The stage number setting unit periodically sets the stage number of the buffer unit based on the longest frame length measured by the frame length measurement unit within a predetermined period. A ring network station. A frequency deviation compensation method in a ring network composed of a plurality of stations, wherein a first station transmits a frame for notifying information based on a maximum length of a frame used by a system using the ring network to the network,
Each station that has received the frame sets the number of stages of its own frequency deviation compensation buffer based on information based on the maximum length of the frame notified by the frame .
A frequency deviation compensation method in a ring network characterized by the above. A frequency deviation compensation method in a ring network composed of a plurality of stations, the station measuring a length of a transmission frame passing through the station for a certain period,
A frequency deviation compensation method in a ring network, characterized in that the number of stages of its own frequency deviation compensation buffer is set based on the length of the maximum transmission frame as a result of the measurement.   7. The ring network according to claim 5 or 6, wherein the setting of the number of stages of the frequency deviation compensation buffer is performed by determining a timing for extracting stored data from the frequency deviation compensation buffer for transmission frame generation. Frequency deviation compensation method in   8. The ring network according to claim 5, 6 or 7, wherein the station periodically performs the measurement, and sets the number of stages of its own frequency deviation compensation buffer based on the measurement result. Frequency deviation compensation method in

Images