This standard specifies a multilink procedure in which multiple parallel data links at the data link layer are used to provide data links of variable bandwidth between network layer entities. The multilink procedure (MLP) appears as a new higher sublayer in the data link layer, operating between the network layer and multiple single link protocol functions (SLPs) at the data link layer. This standard does not specify the means by which the SLPs indicate to the MLP that a multilink frame transmission has been successfully completed. The multilink procedure does not preclude the use of single link protocols with different delay characteristics and/or line rates before a multilink frame is formed. This may be achieved by agreeing in advance that all communications on the data link will use the protocol, by negotiating the use of the protocol by one of the SLPs, or by other means. The means by which both parties gain a common understanding as to whether or not to use the multilink procedure is to be used is not specified in this standard. GB/T 15124-1994 Multi-link specification for data communication in information processing systems GB/T15124-1994 Standard download decompression password: www.bzxz.net

National Standard of the People's Republic of China
Informatlon processing system
Data communication
Multilink procedures
GB/T15124—94
ISO74781987
This standard is equivalent to the international standard ISO7478-1987 "Multilink procedures for data communication in information processing systems". 0 Introduction
The multilink procedures involve the layers of the ISO Open Systems Interconnection (OSI) reference model, especially the physical layer, data link layer and network layer. The multilink procedures are located in the data link layer.
The multilink procedures provide the following functions: receiving data units from the network layer, scheduling the transmission and retransmission of data units on a set of parallel data links, and reordering the received data units and delivering them to the network layer. The multilink protocol has the following basic characteristics: it provides multiple connections between two data stations to make the service economical and reliable; a.
allows adding or reducing connections without interrupting the service provided by multiple connections; b.
optimizes the bandwidth utilization of a group of connections by balancing the load: when some connections fail, the service quality of the connection group decreases appropriately; d.
provides a single logical data link to the network layer for each multilink group; e.
f. if necessary, it can reorder the received data units before delivering them to the network layer. 1. Content and scope of application
This standard specifies the multilink protocol. In this protocol, multiple parallel data links at the data link layer are used to provide variable bandwidth data links between network layer entities. The multilink protocol (MLP) appears as a new higher sublayer in the data link layer. It operates between the network layer and multiple single data link protocol functions (SLP) of the data link layer (see Figure 1). This standard does not specify the method by which the SLP indicates to the MLP that the multilink frame transmission has been successfully completed. The multilink protocol does not preclude the use of single link protocols with different latency characteristics and/or line rates to form a multilink group. When the protocol specified in this standard is used on one or more parallel data links, both ends of the data link must understand that the protocol will be used before the first multilink frame is sent. This can be achieved by agreeing in advance that all communications on the data link will use the protocol, by negotiating the use of the protocol through one of the SLPs, or by other means. The method by which both parties obtain a common understanding of whether to use or not to use the multilink protocol: Not specified in this standard. Approved by the State Bureau of Technical Supervision on July 16, 1994 and implemented on March 1, 1995
2 Definitions and integers
Physical layer
GB/T1512494
Wuhan Railway Network
Multi-link connection
Multi-link single link operation
--Multi-link connection
Figure 1 Functional structure of multi-link protocol
SI.P ​​Single link protocol, MLF Multi-link protocol 2.1 MLP reset confirmation bit (C) MLP resct conlirmation bit (C) The network is
1 element
The MLP reset confirmation bit is used to respond to the setting of the \1\R bit to confirm that all state changes of the MLP have been reset. In normal communication, the C bit is set to "0", that is, no reset request has been initiated. The C bit is set to "1" to respond to the multilink signal with the R bit set to "1", indicating that the reset of the MLP state variable has been completed. When C is 1, the multilink signal has no data unit field. 2.2 Multilink procedures (MLP) multilink procedures (Mi.P) The protocol used to transmit data on one or more SLPs that constitute a group. 2.3 Multilink send sequence number [MN(S)] multilink xend sequence number [MN(S)] The value of the multilink sequence number assigned to the multilink frame. Note: The assignable number is in the range from 0 to 4095 (including 0 and 4095) and is used for reordering. At the multilink protocol (MLP) data receiving station, when reordering is required, the code can be used for reordering before delivering the data unit to the network layer, and it can be used to detect missing or duplicated multilink frames. 2.4 Lost multilink frame timer (MT1) lost. multilink frame timer (MT1) During periods of low traffic, the timing of this timer is used to determine that a multilink frame with MV(S) equal to MV(R) has been lost. 2.5 Grouphusy timer (MT2) Optional. On receive MLPs, the timing of this timer is used to determine if a blockage exists due to buffer exhaustion before reordering can be completed.
2.6 MLP reset confirmation timer (MT3) MIP rcset confirmalion limer (MF3) Timer, MT3 provides a means of identifying a multilink frame with the expected C bit set to "1" but not received after a multilink frame with the R bit set to "1" has been transmitted.
2.7 Multilink receive state variable [MV(R)] multilink Teceive state variable MV(R)] represents the MN(S) of the next in-sequence multilink frame to be transmitted to the network layer. Note: MV(R) is the lower edge of the receive window.
2.8 Multilink transmit state variable [MV(S)] mulilink send stale variable [MV(S)]CB/15124—94
represents the next MN(S) in sequence to be assigned to the multilink frame, and the multilink will resume sending on the multilink group. 2.9 Multilink frame acknowledgment state variable [MV(T)] represents the earliest multilink MN(S) waiting for an acknowledgment indication from the local SLP. Note: MV(T) is the lower limit of the send window
2.10 Multilink window size (MW) multilinkwindowsize (MW) For a given transmission direction, all sending MLPs should use the same MW. Note: The send window includes sequence numbers ranging from MV(T) to MV(T)+MW-1 (including both ends). The receive window includes sequence numbers ranging from MV(R) to MV(R)+MW-1 (including both ends). MW is a system parameter and should never exceed 4095MX: Factors affecting the value of parameter MW include, but are not limited to, link transmission and propagation delays, number of links, range of multilink recovery lengths, and number of single-link frames (number of retransmissions N, response time, and number of unresolved frames). 2.11 Receive MLP window guard zone (MX) receiveMI.Pwindowguard (MX) A range of sequence numbers of a certain size above the upper edge of the receive window. If a multilink frame is received with an MN(S) in this guard zone, then the unreceived frames L in the range from MV(R) to the received MN(S)-MW are lost: 2.12 Range of abnormal multilink frames (MZ) range of MN(S) sequence numbers that should not be received during normal operation. Note: Multilink frames received within the Mz range should be discarded. 2.13 Number of SLP retransmission attempts (N) Number of times a single link protocol (SLP) data station attempts to retransmit a multilink frame before notifying the MLI. Note: The subsequent action of the SLP depends on the design of the SLP. The subsequent action of the MLP should include allocating the multilink frame to the original SLP or one or more other SLP transmissions. 2.14 MLP reset request bit (R) MLP reset request bit (R) The MLP reset request bit is used to request the reset of the state variables of the MLP. In normal communication, the R bit should be set to "0", that is, there is no multilink reset request. The R bit is set to "1" to request the reset of the state variables of the receiving MLP. In the case of R-1, the multilink frame data unit segment does not contain the information of the underlying layer, but may contain an optional reason field that reflects the reset reason to indicate the reset reason. Note: In the DTE/DCE X.25 interface application, an 8-bit reason field is specified.The encoding of this 8-bit field is for further study. Refer to GB/T 14399
2.15 Sequence check option bit (S) The S bit is only meaningful when V=1 (indicating that the received multilink recovery does not need to be reordered). S=1 means that the MN (S) number has not been assigned, and S0 means that the MN (S) number has been assigned. Therefore, although reordering is not required: it is possible to check the duplication of multilink tops and identify the loss of multilinks.
2.16 Single link procedures (SI,P) single link procedures (SLP) Data link protocol used for the establishment, maintenance, transmission and termination of logical data links on a single data link. 2.17 Void sequencing bit (V) bit (V) The V bit indicates whether the received multilink frame is subject to sequencing constraints. V = 1 means that sequencing is not required. V0 means that sequencing is required. 3 Multilink frame format
In order to allow data units (e.g., packets) to be reordered, a multilink control (MLC) field is required. This field is transmitted as the first two octets of the information field of the SLP transmission unit. Figure 2 shows the (MLC) field and its relationship with the data unit and the S1.P header/part.
Note: If the SLP uses encoding-related procedures, some modifications to the format may be required. The two-octet MLC field should follow the SLP header and expand the data link layer encapsulation used to transmit data units between network layer entities. The single link and multilink control fields are only generated and used at the data link layer and cannot be forwarded to the network layer.
CB/T15124-94
The multilink sequence number MN(S) should consist of 12 bits, providing a modulo 4096 Count. MN(S) shall be divided into two fields, MNH(S) and MNL(S), as shown in Figure 2. Four control bits shall be provided in the M1.C field to provide additional multi-link control features. Handover SLF/Received from SLP The first bit of the SLP header with MN(S)
MNL(S>
The weight of the bit in MNrS1
East link digging Ning section
Old silicon width
Payed to SLP/Received from Si.P The received 1-bit
SLP header
Special starting value in MN(S)
Multilink rural market scanning
Multilink machinery
Figure 2 Multilink frame format
SLPE part
Data core
SI.P ​​part
MNH(S)-[2 bits Multilink send sequence number's 9th to 12th bits+MNL(S)-[2 bits Multilink send sequence number's 1st to 8th bits; V-no-sequencing bit; S-check sequence option bit; R-MI,P reset request bit; C..MLP reset confirmation bit ST,P-Single link protocol no-sequencing bit V is used to indicate that the received data units are not to be reordered into the order received from the network layer before being delivered to the network layer. When V=1, data units do not need to be reordered, and the receiving MLP should immediately deliver the data units to the network layer without considering its order in the receiving sequence. When V=0, the received data units should maintain the original order when delivered to the network layer. V=0 indicates that the MN(S) number exists. The check sequence option bit S is only meaningful when V=1 (indicating that the multilink frame does not need to be reordered). When V=0, the S bit is meaningless. When V=1 and S=1, the sending MLP should not assign the MN(S) value to the multilink frame. Therefore, the data units included in the multilink frame should be forwarded to the network layer without checking the sequence integrity or duplication. When V-1 and S-0, the sending MLP should assign the MN(S) value to the multilink frame to facilitate the detection of duplicate or lost multilink frames, but the sequence integrity does not need to be guaranteed. The data units of the duplicated multilink frames should not be delivered to the network layer. The MLP reset request bit R is used to initiate the multilink reset process. The MLP reset confirmation bit C is used to indicate the completion of the multilink reset process. During normal data exchange, the MLP reset request bit and the MLP reset confirmation bit should be set to "0". 4 Transmitter Operation
4.1 Overview
The transmitting MLP is responsible for controlling the flow of data units from higher layers (e.g., network layer) into the multilink frame and then sent to the SLP for transmission to the remote receiving MLP.
The functions of the transmitting MLP are:
Accepting data units from the network layer,
GB/T 1512494
Assign the multilink control field containing the appropriate sequence number MN(S) to the data unit: ensure that the assigned MN(S) does not exceed the MLP sending window, pass the resulting multilink frame to the SLP, transmit + accept the indication of successful transmission confirmation from the SLP. Monitor the transmission failure or difficulty that occurs in the SLP sublayer and recover from it; accept the stream control indication from the SLP. and take appropriate actions. 4.2 Transmitting Multilink Frames
When the sending MI.P receives a data unit from the network layer, it shall place the data unit in a multilink recovery frame and then send it to the SLP when necessary (V-(, or V-1 and S=D) set MN(S)=MV(S), and increase MV(S) by 1. The incremented sending and receiving state variables are in a continuous repetitive order: that is, for the modulo 4096 sequence, 4095 is 1 higher than 4094.0 is 1 higher than 4095.
If MN(S) is less than MV(T)+MW, and the remote data station has not indicated a busy state on all available data links, the sending MLP shall assign the multilink frame with the lowest MN(S) that has not been assigned to an available SLP. When the SI.P receives an acknowledgment from the remote that the multilink recovery transmission has been successfully completed, it shall indicate this to the sending MLP. Then send ML .P can delete confirmed multilink frames. When the sending MI.P receives an acknowledgment indication from the SLP., it should advance MV(T) to indicate the multilink frame with the lowest end number that has not yet been confirmed. The sending ML.P always allocates the multilink frame with the lowest MN(S) first. The sending MLP can also allocate a multilink frame to multiple SLPs.
If a multilink frame is sent on multiple data links (for example, to increase the probability of successful delivery), there is a possibility that one of these multilink frames (i.e., a duplicate multilink frame) may be delivered to the remote MLP after an earlier frame is confirmed. This earlier multilink frame may cause the receiving MLP to The sending MLP increments its MV(R). To ensure that old duplicate multilink frames are not mistaken by the receiving MLP as new multilink frames, the sending MLP M1.P must not send a new multilink frame with MN(S) equal to or greater than MN(S)-MW-MX before all SLPs have successfully transmitted the multilink frame or retransmitted the multilink frame up to the maximum number of times, where MN(S) is related to duplicate multilink frames transmitted on other SLPs. One way is to prohibit incrementing MV(T) before all SLPs have either successfully transmitted the multilink frame or retransmitted the frame up to the maximum number of times. Increment sequence number
Figure 3 Sender flow control
4.3 Sender flow control
GB/T 15124-94
Flow control is achieved using the window size parameter MW and by the remote SLP indicating the busy state. The sending MLP shall not assign a MN(S) greater than MV(T)+MW-1 to the multilink. When the next multilink frame to be allocated has MN(S) = MV(T) + MW, the sending MLP shall suspend that frame and the subsequent multilink frames until an acknowledgement indication to advance MV(T) is received.
In Figure 3, MV(S) has become equal to MV(T) + MW. At this point, the sending MLP shall not allocate any new multilink frames until MV(T) is advanced.
The receiving MLP performs flow control on the sending MLP by indicating a busy state on one or more SLPs. The number of SLPs that appear busy determines the degree of transmitter flow control that can be achieved. When the sending MLP receives a busy state indication from one or more of its SLPs, it shall reassign any unacknowledged multilink frames that were previously allocated to those SLPs. The sending MLP shall allocate the multilink frames containing the lowest MN(S) to the available SLPs in accordance with the above provisions. 4.4 Retransmission
When the sending SLP retransmits a single-link frame containing a multi-link frame N times and still does not receive confirmation of the single-link frame, the sending MLP will reallocate the multi-link frame to the original SLP or other SLPs.Transmit unless other SLPs have confirmed the restoration of the multilink.
4.5 Link Failure
If the sending SLP detects a data link failure, the SI.P shall report to the sending MIP that the data link cannot be used. All unconfirmed multilink frames shall be reallocated by the sending MIP. The sending MIP shall no longer allocate any multilink frames to the SLF until the data link is reestablished.
4.6 Transmitter Reassignment
The data link may not be able to provide service due to a disconnection of the physical layer or the data link layer. The sending SI.P shall indicate to the sending MLP that the data link cannot be operated. All unconfirmed multilinks allocated to the SLP shall be reallocated by the sending MLP. When the data link resumes service or provides service for the first time, the +SLP shall indicate to the sending MLP. The sending MLP can then allocate multilink frames to the SI.P.
5 Receiver Operation
5.1 Overview
When receiving a multilink frame from the receive SLP, the contents of the multilink control field shall be checked. When V=C, the multilink frames shall be reordered as described in 5.3 before the data unit is delivered to the network layer. In the event of a transmission error not detected by the receive SLP, or an internal error in the machine, multilink frames may be lost. The receive MLP shall detect such lost multilink frames as described in 5.4. It is envisaged that the higher layer recovery mechanisms are able to recover from this situation. If the receive MLP is about to run out of buffer resources, the flow control method described in 5.5 may be used. 5.2 Receive Multilink Frames
Multilink frames less than two octets in length shall be discarded by the receive MLP. If V=0, the multilink frames shall be reordered as described in 5.3. If V=1 and S=0, and MV(R) is less than or equal to MN(S)MN(S) is less than or equal to MV(R)+MW+MX-I, and MN(S) is not a duplicate MN(S). Then the data units contained in the multilink frame shall be delivered to the network layer immediately. If V=1 and S=1, the data units contained in the multilink frame shall be delivered to the network layer immediately. Compliance number
GB/T 15124—94
Exception area
Figure 4 Receive multilink frame
Receive window
The receive MLP sequence number used to control the receive multilink frame can be divided into three areas, as shown in Figure 4. MV(R) is the MN(S) of the next in-sequence multilink frame, the data units of which are to be delivered to the network layer. Multilink frames with higher MN(S) may have been received earlier and are suspended to wait for the expected multilink frame before being delivered to the network layer. The three sequence number zones associated with MV(R) are:
a. The receive window MW includes sequence numbers from MV(R) to MV(R)+MW-1: b. The protection zone MX includes sequence numbers from MV(R)+MW to MV(R)+MW+MX-1: c. The exception zone MZ contains sequence numbers from MV(R)+MW+MX to MV(R)-1. When a multilink frame with MN(S)=MV(R) is received, its data units shall be delivered to the higher layer (e.g., the network layer) regardless of whether it has been suspended in accordance with Section 5.3, and the receive MLP shall increment MV(R). 5.3 Reordering
The receive MLP reorders the multilink frames by the following method, that is, all multilink frames that need reordering and have MN(S) in the range of MV(R)+1 to MV(R)+MW-1 are suspended before their data units are delivered to the network layer. All multilink frames that need to be reordered and have sequence numbers less than MV(R) shall have been reordered and their data units delivered to the network layer. 5.4 Detection of lost multilink
Method:
MW is a fixed parameter known to both the sending and receiving MLPs, and the sending MLP window starts at the lowest multilink sequence number (modulo 4096) for which the SLP has not indicated an acknowledgment. Therefore, in addition to not receiving duplicate frames, the receiving MLP shall not receive multilink frames with MN(S) outside the receive window. When the receiving SLP accepts and acknowledges a transmission unit containing a multilink frame and the sending SLP indicates an acknowledgment to the sending MLP, the sending MLP may advance its window to allow the allocation of a multilink frame with a higher MN(S) to the SIP. If the received multilink frame is subsequently lost, the receiving MLP's receive window cannot be advanced. The receiving MLP shall wait for the lost multilink frame until: a. a multilink frame with a MN(S) in the protected area is received; or b. timer MT1 expires.
When a lost multilink frame is detected, the network layer can be notified. When a multilink frame located in the protected area is received by MN(S>), it can be considered valid. If a multilink frame from MV(R) to the received MN(S) minus MW has not been received, it should be considered lost; if it has been received, it should be passed to the network layer. The receiving window should be GB/T 15124-94
rotates so that MV(R) is equal to the received MN(S) minus MW+1. Then continue with normal processing of multilink frames as specified in Section 5.2 (see Figure 5). The
protection zone should be large enough for the receiving MLP to facilitate the receiving MLP to identify the highest MN(S) that can be legally received after the multilink frame loss occurs outside its receive window. MX should not be greater than MW, for example: a. In a system where the transmitting MLP sequentially assigns h sequentially consecutive multilink frames to the i-th SLP, MX should be greater than or equal to the sum of all h; +1-min, where h. is equal to the smallest h encountered. In the case where there are L SLPs in the multilink group, MX should be greater than or equal to
Zh, +1- hm: or
h, in a system where the sending MLP assigns h sequentially consecutive multilinks to each SLP in turn, the MX of the receiving MLP should be greater than or equal to h(L-1)+1, where L is the number of SLPs in the multilink group. MV(R) day Chen Lai
avoid adding pieces
receive N:
lost
delivered to the network layer
a new MV(R)
received MN(S)||tt ||Incremental sequence
Figure 5 Detection of lost multilink frames
If timer MT1 times out, a multilink frame with MN(S)=MV(R) may be declared lost. The receive window rotates as described above. Upon receiving a multilink frame with MN(S) greater than MV(R) and requiring reordering, timer MT1 shall be started. Upon receiving a multilink frame with MN(S)-MV(R) and upon receiving a multilink frame with MN(S) equal to the new MV(R) Timer MT1 should be restarted if there are other multilinks waiting for delivery before the receiving MLP indicates a busy state; otherwise, timer MT1 should be reset. Timer MT1 should be stopped when all SLPs indicate a busy state.
5.5 Receiver flow control
When flow control is required for the sending MLP, one or more receiving SLPs may indicate a busy state. If the receiving MLP may exhaust its receive buffer capacity before completing the reordering, timer MT2 may be installed. Timer MT2 should be started as long as the receiving MLP indicates a busy state on all SLPs and there are multilinks on the receiving MLP waiting for reordering.
Timer MT2 should be reset when the busy state is cleared on one or more SLPs. If timer MT2 times out, the multilink with MN(S) = MV(R) is blocked and considered lost. MV(R) should be incremented to the next sequence number not yet received, and the data units contained in the multilink with the same sequence number should be delivered to the network layer. If the busy state remains valid on all SLPs and multiple multilink sequences are waiting for reordering, timer MT2 should be restarted. The period of timer MT2 can be set independently by the implementer of each M1L.P. 6 Initialize MLF
Use the reset procedure described in Chapter 7 to initialize MV(S), MV(T) and MV(R) to 07 Reset MLP
The multilink sequence reset procedure provides a mechanism to synchronize the multilink sequence numbers of interconnected MLPs. After the multilink reset procedure is successful, the multilink sequence numbers in each direction should start from 0. The multilink sequence number R=1 is used to request a multilink reset. The multilink recovery response C=1 confirms that the multilink reset procedure has been completed. The MLP resets MV(S) and MV(T) to 0 based on the transmission of the multilink response R=1: resets MV(R) to 0 based on the receipt of the multilink recovery response R=1,When an MLP initiates the reset procedure, it shall remove all unacknowledged multilink frames pending in that MLP and in the associated SLPs and maintain control of these frames. Thereafter, the initiating MLP shall not transmit multilink frames with R=C= until the reset procedure is completed (one way to remove multilink frames in an SLP is to disconnect the data link of that SLP). The initiating MLP then resets its multilink transmit state variable MVS) and its transmit multilink frame acknowledgement state variable MV(T) to 0. The initiating MLP then transmits a multilink frame acknowledgement with R=1 on its SLP as a reset request and starts timer MT3. The value of the MN(S) field in R=1 may be arbitrary, since the receiving MLP ignores the MN(S) field when R=1. The initiating MLP shall continue to receive and process multilink frames from the remote MLP according to the procedure described in clause 5.2 until it receives a multilink frame with R=1 from the remote MLP. In normal communication, the MLP that receives the multilink side (reset request) with R=1 from the initiating MLP shall start the operation according to the above procedure. The MLP shall not receive the multilink frames with R=C=0 from the initiating MLP until the reset procedure is completed. Any such multilink frames received shall be discarded. When the MLP has initiated its own multilink reset procedure and transmits the multilink frame with R=1 to one of its SLPs for transmission, the MLP does not repeat the above operation when it receives the multilink frame with R=1 from another MLP. The receipt of the multilink frame (reset request) with R=1 causes the receiving MLP to deliver the data units of those multilinks that have been received to the higher level and identify those that have been allocated to the SLP but have not been confirmed. For any multilink frames whose number values ​​are in the interval from the original value of MV(R) to and including the maximum number value of the received multilink frames, if they are not received by the MLP, the MLP may report the loss of these frames to the higher layer. Then, the receiving MLP resets its multilink reception state variable MV(R) to 0. After the MLP assigns the R-1 multilink frame to one of its SLPs, it shall receive an indication of the success of the transmission from the SLP as one of the prerequisites for transmitting the C=1 multilink frame. When the initiating MLP subsequently receives the R=1 multilink and completes the above-mentioned multilink state variable reset operation, the initiating MLP shall transmit the C=1 multilink frame (reset confirmation) to the other MLP. When an MLP: a. receives the R=1 multilink,
b. transmits the R=1 multilink frame;
e. completes the above-mentioned multilink state variable reset operation, assuming that the indication of the success of the transmission of the R=1 multilink has been received from the SLP of the MLP, then the MLP shall transmit the C=1 multilink frame (reset confirmation) to the other MLP as soon as possible. The C=1 multilink frame is a response to the R=1 multilink frame. The value of the MN(S) field in the C.1 frame can be any value, because when C=1, the receiving MLP ignores the MN(S) field. After the multilink is reset, the received multilink sequence number MN(S) in each direction shall start from 0. When the MLP uses only one SLP to send multilink frames with R=1 and receive multilink frames with C=1, it may transmit multilink frames with C=1 immediately after transmitting multilink frames with R=1 without waiting for the SLP transmission completion indication. The MLP shall not retransmit multilink frames with R=1 or C=1 unless timer MT3 times out. The MLP may use two different SLPs, one for sending multilink frames with R=1 and the other for sending multilink frames with C=1 after receiving a successful or unsuccessful transmission indication of multilink with R=1. Multilink frames with R=C=1 shall never be used.
When the MLP receives multilink frames with C=1, it shall stop its timer MT3. Transmitting multilink frames with C=1 to the remote SLP and receiving multilink frames with C=1 from the remote MLP completes the multilink reset procedure of one MLP. For the first R_C=0 multilink reply transmitted, the sequence number value shall be 0. After transmitting a C=1 multilink frame to an SLP, one or more R=C=0 multilink frames may be received. After receiving a C=1 multilink reply, the MLP may transmit one or more R=C=0 multilink frames. When the MLP additionally receives one or more R=1 multilink replies between receiving an R=1 multilink reply and transmitting a C1 multilink reply, the MLP shall discard the extra R=1 multilink replies. When the MLP receives a C=1 multilink frame that is not a response to an R=1 multilink reply, the MLP shall discard the C=1 multilink reply. After the MLP transmits a C=1 multilink frame to one of its SLPs, the MLP may receive an R=1 multilink reply from another MLP. This MLP shall treat this R=1 multilink reply as a new reset request and start the multilink reset procedure from the beginning. When the MLP does not receive a multilink reply with R=1 but transmits a multilink frame with R=1 and thus receives a multilink reply with C=1, the MLP shall restart the reset procedure from the beginning. When the timer MT3 times out, the multilink reset procedure shall be restarted from the beginning. The value of the timer MT3 shall be large enough to include the transmission, retransmission, propagation delay of the SLP and the operation time of the MLP receiving the multilink reply with R=1 and responding with a multilink frame with C=1.
A1 Introduction
GB/T15124—94
Appendix A
Example of Multilink Reset Procedure
(Reference)
The following example introduces the use of the multilink reset procedure in the following situation. A single station initiates MLP reset, and
b. MLP reset is initiated by two stations at the same time
2 MLP reset initiated by a single station
No multi-link frame
MV(S)-I
MVTI-0
MV(R)-U
Note: 1) Confirm the delivery of SLP frame with R=1 multi-link town; 2) Confirm the delivery of SLP frame with C=1 multi-link frequency. 3. MLP reset initiated by two stations at the same time
especially multilink
NV(S)-0
NV(T)=
especially multilink
MV(Si-
MV(TI
MV(R =I
GB/T15124—94
Ro,c=n
Note: 1) Confirm delivery of R=1 multilink SI,P I2) Confirm delivery of C=1 multilink SLP Lai. Additional notes:
This standard was proposed by the Ministry of Machinery and Electronics Industry of the People's Republic of China. This standard is under the jurisdiction of the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry. B station
North China Link Building
MY(s)-o
MVRi-心
This standard was drafted by the Department of Electronic Engineering of Shanghai Jiaotong University and the North China Institute of Design and Technology of the Ministry of Machinery and Electronics Industry. The main drafters of this standard are Zhu Hongwen, Zhang Baodong, Zhong Shengyang, and Huang Jiaying.After receiving the multilink with R=1, it shall receive an indication of whether the transmission is successful from the SLP. This is one of the prerequisites for transmitting a multilink frame with C=1. When the initiating MLP subsequently receives a multilink with R=1 and completes the above multilink state variable reset operation, the initiating MLP shall transmit a multilink frame with C=1 (reset confirmation) to the other MLP. When an MLP: a. receives a multilink with R=1,
b. transmits a multilink frame with R=1;
e. completes the above multilink state variable reset operation, assuming that an indication of whether the transmission of the multilink with R=1 is successful or not has been received from the SLP of the MLP, then the MLP shall transmit a multilink frame with C=1 (reset confirmation) to the other MLP as soon as possible. The multilink frame with C=1 is a response to the multilink frame with R=1. The value of the MN(S) field in the above C.1 can be any value, because when C=1, the receiving MLP ignores the MN(S) field. After the multilink reset, the received multilink sequence number MN(S) in each direction shall start from 0. When the MLP uses only one SLP to send multilink frames with R=1 and receive multilink frames with C=1, it can immediately send multilink frames with C=1 after sending multilink frames with R=1 without waiting for the SLP transmission completion indication. The MLP shall not resend multilink frames with R=1 or C=1 unless timer MT3 times out. The MLP may use two different SLPs, one for sending multilink frames with R=1 and the other for sending multilink frames with C=1 after receiving a successful or unsuccessful transmission indication of multilink frames with R=1. Multilink frames with R=C=1 shall never be used.
When the MLP receives a multilink frame with C=1, it shall stop its timer MT3. Transmitting a multilink frame with C=1 to the remote SLP and receiving a multilink frame with C=1 from the remote MLP completes the multilink reset procedure of an MLP. For the first multilink frame with R=C=0 transmitted, its sequence number value shall be 0. After transmitting a C=1 multilink frame to an SLP, one or more R=C=0 multilink frames may be received. After receiving a C=1 multilink frame, the MLP may transmit one or more R=C=0 multilink frames. When the MLP additionally receives one or more R=1 multilink frames between receiving an R=1 multilink frame and transmitting a C1 multilink frame, the MLP shall discard the extra R=1 multilink frames. When the MLP receives a C=1 multilink frame that is not a response to an R=1 multilink frame, the MLP shall discard the C=1 multilink frame. After the MLP transmits a C=1 multilink frame to one of its SLPs, the MLP may receive an R=1 multilink frame from another MLP. This MLP shall treat this R=1 multilink frame as a new reset request and start the multilink reset procedure from the beginning. When the MLP does not receive a multilink reply with R=1 but transmits a multilink frame with R=1 and thus receives a multilink reply with C=1, the MLP shall restart the reset procedure from the beginning. When the timer MT3 times out, the multilink reset procedure shall be restarted from the beginning. The value of the timer MT3 shall be large enough to include the transmission, retransmission, propagation delay of the SLP and the operation time of the MLP receiving the multilink reply with R=1 and responding with a multilink frame with C=1.
A1 Introduction
GB/T15124—94
Appendix A
Example of Multilink Reset Procedure
(Reference)
The following example introduces the use of the multilink reset procedure in the following situation. A single station initiates MLP reset, and
b. MLP reset is initiated by two stations at the same time
2 MLP reset initiated by a single station
No multi-link frame
MV(S)-I
MVTI-0
MV(R)-U
Note: 1) Confirm the delivery of SLP frame with R=1 multi-link town; 2) Confirm the delivery of SLP frame with C=1 multi-link frequency. 3. MLP reset initiated by two stations at the same time
especially multilink
NV(S)-0
NV(T)=
especially multilink
MV(Si-
MV(TI
MV(R =I
GB/T15124—94
Ro,c=n
Note: 1) Confirm delivery of R=1 multilink SI,P I2) Confirm delivery of C=1 multilink SLP Lai. Additional notes:
This standard was proposed by the Ministry of Machinery and Electronics Industry of the People's Republic of China. This standard is under the jurisdiction of the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry. B station
North China Link Building
MY(s)-o
MVRi-心
This standard was drafted by the Department of Electronic Engineering of Shanghai Jiaotong University and the North China Institute of Design and Technology of the Ministry of Machinery and Electronics Industry. The main drafters of this standard are Zhu Hongwen, Zhang Baodong, Zhong Shengyang, and Huang Jiaying.After receiving the multilink with R=1, it shall receive an indication of whether the transmission is successful from the SLP. This is one of the prerequisites for transmitting a multilink frame with C=1. When the initiating MLP subsequently receives a multilink with R=1 and completes the above multilink state variable reset operation, the initiating MLP shall transmit a multilink frame with C=1 (reset confirmation) to the other MLP. When an MLP: a. receives a multilink with R=1,
b. transmits a multilink frame with R=1; Www.bzxZ.net
e. completes the above multilink state variable reset operation, assuming that an indication of whether the transmission of the multilink with R=1 is successful or not has been received from the SLP of the MLP, then the MLP shall transmit a multilink frame with C=1 (reset confirmation) to the other MLP as soon as possible. The multilink frame with C=1 is a response to the multilink frame with R=1. The value of the MN(S) field in the above C.1 can be any value, because when C=1, the receiving MLP ignores the MN(S) field. After the multilink reset, the received multilink sequence number MN(S) in each direction shall start from 0. When the MLP uses only one SLP to send multilink frames with R=1 and receive multilink frames with C=1, it can immediately send multilink frames with C=1 after sending multilink frames with R=1 without waiting for the SLP transmission completion indication. The MLP shall not resend multilink frames with R=1 or C=1 unless timer MT3 times out. The MLP may use two different SLPs, one for sending multilink frames with R=1 and the other for sending multilink frames with C=1 after receiving a successful or unsuccessful transmission indication of multilink frames with R=1. Multilink frames with R=C=1 shall never be used.
When the MLP receives a multilink frame with C=1, it shall stop its timer MT3. Transmitting a multilink frame with C=1 to the remote SLP and receiving a multilink frame with C=1 from the remote MLP completes the multilink reset procedure of an MLP. For the first multilink frame with R=C=0 transmitted, its sequence number value shall be 0. After transmitting a C=1 multilink frame to an SLP, one or more R=C=0 multilink frames may be received. After receiving a C=1 multilink frame, the MLP may transmit one or more R=C=0 multilink frames. When the MLP additionally receives one or more R=1 multilink frames between receiving an R=1 multilink frame and transmitting a C1 multilink frame, the MLP shall discard the extra R=1 multilink frames. When the MLP receives a C=1 multilink frame that is not a response to an R=1 multilink frame, the MLP shall discard the C=1 multilink frame. After the MLP transmits a C=1 multilink frame to one of its SLPs, the MLP may receive an R=1 multilink frame from another MLP. This MLP shall treat this R=1 multilink frame as a new reset request and start the multilink reset procedure from the beginning. When the MLP does not receive a multilink reply with R=1 but transmits a multilink frame with R=1 and thus receives a multilink reply with C=1, the MLP shall restart the reset procedure from the beginning. When the timer MT3 times out, the multilink reset procedure shall be restarted from the beginning. The value of the timer MT3 shall be large enough to include the transmission, retransmission, propagation delay of the SLP and the operation time of the MLP receiving the multilink reply with R=1 and responding with a multilink frame with C=1.
A1 Introduction
GB/T15124—94
Appendix A
Example of Multilink Reset Procedure
(Reference)
The following example introduces the use of the multilink reset procedure in the following situation. A single station initiates MLP reset, and
b. MLP reset is initiated by two stations at the same time
2 MLP reset initiated by a single station
No multi-link frame
MV(S)-I
MVTI-0
MV(R)-U
Note: 1) Confirm the delivery of SLP frame with R=1 multi-link town; 2) Confirm the delivery of SLP frame with C=1 multi-link frequency. 3. MLP reset initiated by two stations at the same time
especially multilink
NV(S)-0
NV(T)=
especially multilink
MV(Si-
MV(TI
MV(R =I
GB/T15124—94
Ro,c=n
Note: 1) Confirm delivery of R=1 multilink SI,P I2) Confirm delivery of C=1 multilink SLP Lai. Additional notes:
This standard was proposed by the Ministry of Machinery and Electronics Industry of the People's Republic of China. This standard is under the jurisdiction of the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry. B station
North China Link Building
MY(s)-o
MVRi-心
This standard was drafted by the Department of Electronic Engineering of Shanghai Jiaotong University and the North China Institute of Design and Technology of the Ministry of Machinery and Electronics Industry. The main drafters of this standard are Zhu Hongwen, Zhang Baodong, Zhong Shengyang, and Huang Jiaying.MLP reset initiated by two stations simultaneously
2MLP reset initiated by a single station
No multi-link frame
MV(S)-I
MVTI-0
MV(R)-U
Note: 1) Confirm the delivery of SLP frame with R=1 multi-link town; 2) Confirm the delivery of SLP frame with C=1 multi-link frequency. 3. MLP reset initiated by two stations at the same time
especially multilink
NV(S)-0
NV(T)=
especially multilink
MV(Si-
MV(TI
MV(R =I
GB/T15124—94
Ro,c=n
Note: 1) Confirm delivery of R=1 multilink SI,P I2) Confirm delivery of C=1 multilink SLP Lai. Additional notes:
This standard was proposed by the Ministry of Machinery and Electronics Industry of the People's Republic of China. This standard is under the jurisdiction of the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry. B station
North China Link Building
MY(s)-o
MVRi-心
This standard was drafted by the Department of Electronic Engineering of Shanghai Jiaotong University and the North China Institute of Design and Technology of the Ministry of Machinery and Electronics Industry. The main drafters of this standard are Zhu Hongwen, Zhang Baodong, Zhong Shengyang, and Huang Jiaying.MLP reset initiated by two stations simultaneously
2MLP reset initiated by a single station
No multi-link frame
MV(S)-I
MVTI-0
MV(R)-U
Note: 1) Confirm the delivery of SLP frame with R=1 multi-link town; 2) Confirm the delivery of SLP frame with C=1 multi-link frequency. 3. MLP reset initiated by two stations at the same time
especially multilink
NV(S)-0
NV(T)=
especially multilink
MV(Si-
MV(TI
MV(R =I
GB/T15124—94
Ro,c=n
Note: 1) Confirm delivery of R=1 multilink SI,P I2) Confirm delivery of C=1 multilink SLP Lai. Additional notes:
This standard was proposed by the Ministry of Machinery and Electronics Industry of the People's Republic of China. This standard is under the jurisdiction of the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry. B station
North China Link Building
MY(s)-o
MVRi-心
This standard was drafted by the Department of Electronic Engineering of Shanghai Jiaotong University and the North China Institute of Design and Technology of the Ministry of Machinery and Electronics Industry. The main drafters of this standard are Zhu Hongwen, Zhang Baodong, Zhong Shengyang, and Huang Jiaying.
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.