RU2484521C1 - Apparatus for detecting and eliminating faults when transmitting binary signals over two optical channel lines - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: apparatus has a first unit for relaying signals in a channel, having an optical signal coupler and a switch for selecting channel lines, which consists of four interconnected optical switches, and a second unit for detecting, allowing or prohibiting passage of external signals to the first unit, having a first unit for detecting presence of signals in two lines at the same time in the channel, presence of a signal in one line only, absence of signals in both lines during a given time interval, a second unit consisting of four delay elements for external signals, and a third unit consisting of four switches.

EFFECT: broader functional capabilities owing to faster detection, elimination and location of faults when transmitting optical signals.

7 dwg

Description

The invention relates to the field of computer engineering, in particular to messaging devices transmitted by binary signals through a two-line optical channel connecting a group of messaging devices in series, and can be used to quickly detect and eliminate failures in the transmission of messages.

Known devices for detecting and eliminating failures when transmitting messages with binary signals. In particular, such a device adopted as a prototype is a device from the patent of the Russian Federation 2177674, which is designed to detect failures in the lines connecting the stations of the messaging system and quickly select transmission paths of messages that bypass the failures. The performance of the device is achieved by the fact that only hardware is used to detect failed lines and select new message transmission paths.

The disadvantages of the prototype are as follows

- the device is intended only to deal with failures in message lines, failures in the equipment of stations leading to failures in the exchange of messages with serviceable messaging lines, the device does not provide;

- the device does not contain means for quickly determining the locations of failures.

The objective of the present invention is

creation of a device for the rapid detection, elimination and location of failures in messaging equipment, leading to failures in messaging, and messaging lines when transmitting signals through an optical channel, including a ring, consisting of two lines with an opposite direction of signal transmission. In order to speed up the determination of the state of technical means of transmitting messages, the device must perform fast distributed calculations on the contents of messages transmitted through the channel without delaying the message for the calculation. Since it is important for many messaging systems to ensure low energy consumption in messaging equipment, the proposed device to save energy should not create new signals in the channel lines due to the device’s energy, but only change the path of further movement of the signals coming from the channel along the channel lines.

The technical result of the invention is the acceleration of the detection, elimination and location of failures during the transmission of optical signals when operating in two mutually exclusive modes, dynamically switched depending on the type of signals coming from the channel to the device.

1. The device prohibits external signals (signals of an external device transmitting a message to the channel) from affecting the optical signals relayed by the device in the channel.

2. If there are special signals in the message coming from the channel, the device allows external signals to act on the optical signals relayed by the device in the channel, including performing calculations on the signals relayed by the device without delaying the signals for the calculation.

The effect of external signals on the message transmitted through the channel only leads to a change in the choice of the channel line along which the signal going from the device to the channel will be transmitted, and the device does not need to generate new signals instead of those received from the channel. This reduces the energy costs inside the device. The primary source of optical signals in the channel is an external source that sends all devices optical signals in two lines of the optical channel.

The technical result is achieved by the fact that

a device for detecting and eliminating failures in the exchange of binary signals along two lines of the optical channel, into the gap of which this device is located, containing the first block of relaying signals in the channel and the second block of detecting, allowing or prohibiting the passage of external signals to the first block, while the first block contains a coupler optical signals, two inputs of which are connected to the channel lines, and two outputs are connected to the first node of the second block, and the switch for selecting the channel lines for the passage of signals and execution is calculated above them without delay of signals, consisting of four interconnected optical switches, two optical inputs of which are connected to the outputs of the coupler, four electrical control inputs are connected to the outputs of the third node of the second unit, two optical outputs are connected to the channel lines leaving the switch, and the second block contains the first detection node in the channel of the simultaneous presence of signals in two lines, the presence of a signal in only one line, the absence of signals in both lines for a given time interval, connected by outputs to the control inputs of the keys of the third node of the second block and to an external device, contains a second node consisting of four delay elements of external signals, connected by inputs to an external device and outputs to the inputs of keys of the third node of the second block, contains a third node of four keys .

The technical nature and principle of operation of the proposed device is illustrated by drawings.

Figure 1 - Optical switch based on the Mach-Zander interferometer.

Figure 2 - Functional diagram of an optical switch.

Figure 3 - Switch.

Figure 4 - The first block of the device.

Figure 5 - The second unit of the device.

6 - Node analysis of signals in the optical lines of the channel.

Fig.7 - Ring optical channel.

The proposed device for detecting and eliminating failures when transmitting binary signals along two lines of an optical channel contains two blocks: the first block is a block for receiving optical signals coming from the channel, relaying them further to the channel, calculating the signals received from the channel, and the second block is a detection block enable or disable the signals of an external device (WU) that control the first unit.

The above structural elements of the proposed device are as follows.

Figure 1 shows the optical switch based on the Mach-Zander interferometer created by IBM, where 1, 2 are input lines, 9, 10 are output lines, 3, 8 are input and output 50% signal splitters, 4, 5 are shoulders interferometer, 6 - transmission line of control electrical signals, 7 - electrodes of the arm of the interferometer. The signal enters node 3 only from line 1 or only from line 2. Then it is divided equally and enters both arms of the interferometer. A voltage is applied to the electrodes of the first arm, which changes the state of the physical medium of the arm, which leads to a phase shift of the optical signals passing through it. These signals and signals from the second arm then go to the node 8, which, depending on the phase shift of the signals, sends them to channel 9 or 10.

Below will be used the functional description of the switch of figure 1, presented in figure 2. Here, the designations 1, 2, 6, 9, 10 of FIG. 1 correspond to the designations 11, 13, 15, 13, 14. Elements 3, 4, 5, 7, 8 of FIG. 1, showing the physics of the operation of the switch, are represented by the switch 16. Next the switch of FIG. 2 will be designated as Mi.

Figure 3 shows the first block of the proposed device, the center of which is a switch assembled from four switches (figures 1 and 2), designated as Mi (M1-M4). Here the notation is also accepted: a - the entrance to the switch of the first line of the optical channel, the same letter indicates the output of this line from the switch, b - correspond to the input and output of the second line of the channel, and 'and b' are the additional outputs of the switch.

The input lines a and b of the channel are supplied to passive optical couplers 17, which divert part of the signal energy in lines c and d. Mi receives external electrical signals that control the phase shift of the optical signal. When such a control signal with a value of 1 is supplied to the switch or if it is absent (a signal with a value of 0), the optical signal from any input of the switch will go to its outputs, in accordance with the table. The control signals supplied to Mi are indicated in the table in the same way as the switches to which the signal is applied.

When applying the sets of signals Mi specified in the first four rows of the table, the listed combinations of control signals allow you to perform all possible switching inputs / outputs a and b. The next four rows of the table provide the same actions for outputs a 'and b'. Thus, the design of FIG. 3 converts messages represented by optical signals by switching lines. Indeed, we will consider the presence of a signal on line a as a binary unit, and the presence of a signal on line a as binary zero. The presence of a signal simultaneously on two lines is denoted as signal C, the absence of signals simultaneously on two lines is denoted as signal Z. Let either 1 or 0 be input to inputs a and b. Then the set of control signals in the first row of the table does not change the signal value at the output of the switch . The second line replaces output 1 with 0.0 for 1 (invert). The third line turns 1 and 0 into 1. The fourth line turns 1 and 0 into 0. Performing these transformations, as will be shown in the section showing the use of the proposed device, allows us to perform the computational operations necessary for detecting failures without delaying signals for the operation.

Lines a 'and b' can be used to change the path of movement of signals in the system and to output signals from the system at any location.

It can be seen from the table that new signals are created only by moving signals coming from the channel from one channel line to the second and do not require energy from the control unit to create a new signal.

Table M1 M2 M3 M4 entrance Exit 0 0 0 0 a; b a; b one one 0 0 a; b b; a 0 one 0 1/0 a; b a; a one 0 1/0 0 a; b b; b 0 0 one one a; b but'; b ' one one one one a; b b ': a' 0 one one 1/0 a; b but'; but' one 0 1/0 one a; b b '; b ' 0 0 0 one a; b a; b ' 0 0 one 0 a; b but'; b

Important to ensure fault tolerance, the property of both the switch (figure 2) and the switch (figure 3), as can be seen from the table, is that in the absence of control signals on these devices, the inputs and outputs of the corresponding channel lines are automatically connected to each other.

The second block is shown in figure 4. It contains a node 18 for detecting at the input from the channel the simultaneous arrival of signals on two lines, the arrival of a signal from only one line, the absence of signals in both lines for a given time interval, it contains a node 19, consisting of four delay elements of electrical signals, contains a node 20 of four electronic keys, the state of which is controlled by the node 18, and the inputs of the delay elements of the node 19 receive electrical signals from an external source, the outputs of the delay elements are connected to the inputs of the electronic keys of the node and 20, and the outputs of these keys are connected to the control inputs of the switches of the first block (Fig. 4), the switch of which is shown here as 21, and the signal couplers are indicated as 22. The node 18 informs the WU about the presence of signals on the lines of the optical channel.

Nodes 19 and 20 do not require further detail. The node 18 is presented in figure 5.

Here, lines c and d are connected to photodetectors 23 and 24, which convert optical signals into electrical signals that are sent to the slave (lines 30 and 31) for information about the state of the channel, to element 25, a logical I and to timer 26. The timer also receives the signal from the output 25. The signals from the outputs (25) and (26) through the logical OR element (27) are received through line 28 to the keys 20 of Fig. 4, passing signals from the input of the keys to the outputs. The signal from output 25 informs WU about the presence of signal C in the channel (line 31).

An embodiment of the assembly 26 is shown in Fig.6. Node 26 consists of the following elements: 36 - logical OR-NOT, 37 - timer, consisting of a controlled pulse generator and a counter for the number of these pulses, 38 - trigger, 40 - logical OR. Only the presence of signal Z on lines a and b (simultaneous absence of signals at the outputs of nodes 23 and 24) through lines 34 and 35 acts on element 36, which will start the generator in timer 37. When a given number of pulses arrives at the timer counter, a signal appears on its output , which will turn on the trigger 38. The signal from the output 39 of the trigger, passing through the element 27 and line 28, will go to the keys 20 and allow the passage of the signals of the WU through the keys.

The elements of the node 37 and 38 go into the initial state in the following cases:

- when signal C appears on lines a and b, elements 37 and 38 go back to the initial state;

- when any signals other than Z appear on lines a and b, element 37 goes to the initial state.

Thus, the node 18 opens the keys 20 only when the signal C appears on the channel lines in any case and the signal Z appears within a given time interval.

In the latter case, the keys 20 remain open if there are signals 1, 0, Z in the channel. After the signal C appears in the channel, the node 18 ceases to affect the state of the keys 20.

The proposed device is used as follows.

First of all, consider the connection structure of devices involved in messaging. The structure focused on the use of the capabilities proposed in the application of the device shown in Fig.7. It uses two ring optical communication channels with the transmission of signals in opposite directions. Each channel, as indicated above, consists of two lines a and b. The leader control device (41) sends command messages initiating the operation of the group of slaves. 7, these are devices 42-46. The leader and the VU do not generate signals, they all receive two optical signals sent to lines a and b by an external source 47. The leader and the VU, by moving the signal from one channel line to the second, turn the source signals into message signals or transmit them further in the form of signal C .

The command is relayed by all WUs. The WU team and message return to the leader. The leader breaks the ring channel to prevent repeated signal circulation.

Each WU contains in each channel the two blocks discussed above, providing fault-tolerant signal transmission.

We show the implementation and use of the above two modes.

Consider the first mode: the device prevents the external source from distorting the signals relayed by the device in the channel.

In this mode, the WU source of the message transmitted to the channel at the command of the leader must act in accordance with the last two rows of the table: it must generate binary signals 1 and 0 from two signals simultaneously arriving along lines a and b (signal C). Not used further, one of the two signals C is output to the output a 'or b'.

The device for detecting and eliminating failures should prohibit the VU from transmitting a message to the channel if signals C do not arrive from the channel to the device. Only if there are signals C on lines a and b will a signal appear at the output of element 25, which, passing through element 27 to line 28, will allow the WU control signals to pass through the keys 20. In this case, the WU signals in accordance with the last two rows of the table will form binary message signals in the channel from signal C. In the absence of signal C, the keys 20 will be closed. As a result of these actions, the slave will transmit its message only if there are no binary message signals from another slave in the channel.

The delay elements 19 take into account the inertia of the node 18. The node 18 on the lines 29-31 informs WU about the state of lines a and b.

As a result, it is forbidden to transmit messages to the channel from the WU located after the WU, which is selected by the leader for signal transmission. WUs located closer to the leader than WU chosen by the leader can disrupt the transfer of the latter, but they cannot interfere with the transfer of the leader’s commands.

Given the presence of two annular channels, we obtain the following result. Any number of slaves trying to perform unauthorized signaling cannot distort the message required by the leader if the indicated slaves in each channel are between the source of the allowed message and the leader. Any sources of unauthorized signaling cannot distort the leader’s team.

Consider the second mode: the device allows an external source to change the signals relayed in the channel, in particular, to perform calculations on messages transmitted by relay signals.

To perform this mode, the node 18 requires a reaction that is different from its actions in the first mode. At the same time, in order to ensure fault tolerance, it is impossible to switch the state of the node 18 at the command of the WU device complex and therefore less reliable than the proposed device. It is required to form the leader’s team transmitted over the channel so that it acts only directly on node 18, switching it to actions in the first or second mode.

To do this, the leader team requiring distributed computing must have a field in which the result of the distributed calculation will be placed, and the signal Z should be placed at the beginning of this field. This signal will put block 18 into the mode allowing the control unit to control the switch 21 to convert binary signals from the purpose of the calculation. The specified field must end with a symbol C, which will return unit 18 to the blocking mode of unauthorized transmission of signals from the control unit. Thus, the pair of signals Z and C at the beginning and end of the field acts like opening and closing brackets, changing the operation mode of node 18 inside the field located between the “brackets”.

It is necessary to specify the duration of signal Z. A gap is allowed between signals 1 and 0 - the simultaneous absence of signals in lines a and b. This may be required, for example, to separate adjacent bits in an asynchronously transmitted message. Therefore, the duration of the signal Z in the role of brackets should exceed the duration of the specified gap between the message signals, and only to such a signal the node 18 should respond as a bracket. As shown above, this function is performed by the node 26.

The second mode allows without delay in the calculation to carry out operations that determine the state of the system, examples of which are given below.

1. Finding max. It is required to compare the number in the leader’s team field with the number stored in a particular WU and leave the largest of the numbers in the team field.

Numbers are processed bitwise, starting with the highest order. If the current bit of the number of slaves is 1, then 1 should be entered in the same bit of the number in the command field, regardless of the value of the bit coming from the channel (row 3 of the table is executed).

If at the same time the discharge received from the command field is equal to 0, then all other digits of the WU number should replace the digits of the number in the command field. This means that you must act in accordance with line 3 of the table for writing 1, and with line 4 for writing 0. Otherwise, the control unit proceeds to the same actions with the next discharge.

If the digit number in the WU is 0, then the digit value in the command field does not change. Moreover, if the discharge in the command field is 1, then this WU contains a number less than the number in the team field, and the WU ceases to participate in the operation. Otherwise, the WU in the described manner processes the next bit of the number in the command field. After a team traverses all the control centers, the maximum number of numbers stored in all control centers will be in its field. The operation was not required to delay the transmitted command. Similarly, the min operation is performed with the values of the numbers being checked changed from 1 to 0.

2. Operation Account. Each WU should add 1 to the number in the team field.

Numbers are processed starting from the lower order. An exclusive OR operation is performed on each bit: if the bit in the control unit is 0, then the bit in the command field is passed further into the channel without changes (row 1 of the table is executed). If the bit in the control unit is 1, then the discharge value in the command field is inverted (row 2 of the table), and the control unit for processing the next discharge in the command field should adjust its discharge taking into account the resulting transfer 1. The account is executed without message delay.

By switching in accordance with the table, you can perform all logical operations, arithmetic addition, subtraction, multiplication, determine max and min over the content of the message transmitted through the channel without additional delay for the calculation.

Distributed computing not only accelerates the failure detection of messaging tools, as will be shown below, but in some cases much faster than generally accepted solutions, the state of the entire distributed object serviced by a measurement or control system is determined. So, using the max / min operations, during the time the team moves along the ring channel, the leader will discover system objects for which the values of the measured parameters reach max / min or go beyond the specified limits. The Account operation allows you to determine the number of objects with the indicated deviations, find the object closest to the leader or the most distant from it. Within only one command containing several given fields, it is possible to perform a complex sequence of calculations without delaying the message.

3. Detection and localization of WUs that do not send signals to the channel.

Let it be required to detect WUs that do not send signals to the channel in response to the commands of the leader. To do this, apply the following algorithm.

The leader has a serial number of 0, the numbers of WU increase by 1 as you move away from the leader. Numbering is set in advance or is carried out according to the team leader, performing the Account.

The command sent by the leader to the channel has a sequence of fields abcd. In the acd fields, the leader enters the unit. In field b, the leader enters 0.

Each slave operates with abcd fields as follows. WU adds 1 to the number in the field a and the result is relayed.

If the field a received from the channel contains a number that matches the WU serial number, then the same number is entered in the cd fields, which placed the WU in the field a.

If there is no indicated coincidence of numbers (there are failures between the leader and this WU), then 1 is entered in field b and at the same time it is checked if this field did not contain 1 before recording. If 1 was, then the content of the fields cd does not change, otherwise the number from the field a is entered only in the field c.

When a team returns in a ring to the leader, he determines the number of working units by the number contained in field a. If there is 1 in the field b, the field d contains the number of the inoperative unit which is closest to the leader, and the field c contains the number of the first operable unit which is located behind the inoperative unit or a continuous sequence of inoperative units.

If the number received by the leader in the field a does not coincide with the number of slaves connected to the channels, and the field b contains 0, then the slave last in order is faulty.

The stated solution does not allow to determine the location of multiple failures that occurred in various places of the channel. For such a definition, the leader sends a broadcast command check WU, which houses a sequence of bits 0 with the number of bits in it, corresponding to the number of WU. Each operational WU in the corresponding category without delaying the message replaces 0 by 1, which allows sending a single command to determine all the faulty WUs.

4. Detection and finding places of rupture of the lines of the annular channel.

The leader and each WU detects a break in the channel lines by the absence of signals in the channel lines during a time interval exceeding the permissible one. The leader, using a working ring channel, sends a command with a request to indicate the beginning of the gap. The command contains a field containing zero bits, and each control unit that detects the absence of signals in the lines enters 1 in the discharge assigned to it in the field, which allows the leader to determine the location of damage to the lines.

The device has the following advantages.

1. The device detects a prohibited attempt to change the signal relayed in the channel and prohibits it.

2. When using two ring channels with the opposite direction of signal transmission to any number of WUs trying to perform forbidden signal transmission, the device does not allow to distort the message requested by the leader if the specified WUs in each channel are between the source of the authorized message and the leader. Any sources of prohibited signaling cannot distort the leader’s team.

3. The device allows you to convert the signals transmitted through the channel, including performing calculations on them without delaying the signals for the conversion, which ensures a high speed of obtaining information about the state of the message transfer means and the state of other system tools and external objects served by the system.

4. The signals transmitted over the channel are created only by an external source of optical signals. The proposed device performs all signal conversion according to claim 3 by switching sections of the channel lines along which the signal moves. This reduces the power consumption of the WU and the dissipation of energy in them.

Claims (1)

A device for detecting and eliminating failures in the exchange of binary signals along two lines of the optical channel, into the gap of which this device is located, containing the first block of relaying signals in the channel and the second block of detecting, allowing or prohibiting the passage of external signals to the first block, characterized in that the first block contains an optical signal coupler, two inputs of which are connected to the channel lines, and two outputs are connected to the first node of the second block, and a switch for selecting channel lines for the passage of signals and computation above them without signal delay, consisting of four interconnected optical switches, two optical inputs of which are connected to the coupler outputs, four electrical control inputs are connected to the outputs of the third node of the second unit, two optical outputs are connected to the channel lines leaving the switch, and the second block contains the first detection node in the channel of the simultaneous presence of signals in two lines, the presence of a signal in only one line, the absence of signals in both lines for a given the time shaft, connected by the outputs to the control inputs of the keys of the third node of the second block and to the external device, contains a second node consisting of four delay elements of external signals connected by inputs to the external device and the outputs to the key inputs of the third node of the second block, contains the third node, consisting of four keys.

Images