CN115412513B - ARINC664 network receiving terminal time integrity checking method - Google Patents
ARINC664 network receiving terminal time integrity checking method Download PDFInfo
- Publication number
- CN115412513B CN115412513B CN202210839595.4A CN202210839595A CN115412513B CN 115412513 B CN115412513 B CN 115412513B CN 202210839595 A CN202210839595 A CN 202210839595A CN 115412513 B CN115412513 B CN 115412513B
- Authority
- CN
- China
- Prior art keywords
- time
- switch
- message
- node
- delay
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000004044 response Effects 0.000 claims description 20
- 230000001174 ascending effect Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000001934 delay Effects 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 4
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 108700041286 delta Proteins 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 abstract description 6
- 238000004364 calculation method Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 230000007423 decrease Effects 0.000 description 3
- 230000003139 buffering effect Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 241000364483 Lipeurus epsilon Species 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/55—Prevention, detection or correction of errors
- H04L49/552—Prevention, detection or correction of errors by ensuring the integrity of packets received through redundant connections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/70—Virtual switches
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention belongs to the technical field of ARINC664 network high integrity design, and discloses a time integrity checking method for an ARINC664 network receiving terminal, which comprises the following steps: s1, acquiring a time DTS when a first byte of a message leaves a sending terminal; s2, calculating time deviation TOS between the transmitting terminal and the receiving terminal; s3, obtaining a delay upper bound T_word of the message in the switch network; s4, if the clock of the receiving end when the first byte of the message reaches the receiving terminal is within the upper limit of DTS+TOS+T_word, the time integrity check of the message is considered to pass; when the receiving end clock exceeds the upper bound of DTS+TOS+T_word, the receiving end discards the message if the time integrity check of the message is considered to be failed.
Description
Technical Field
The invention belongs to the technical field of ARINC664 network high-integrity design, and particularly relates to a time integrity checking method for ARINC664 network receiving terminals.
Background
In the process of data communication based on a transport layer service mechanism in an aircraft data network, any data transmission error has serious consequences for an upper layer application, so a message receiving end cannot wait for a message indefinitely in the process of data transmission, and after waiting for a certain time, if the message is not received, the message should not be received any more, and the message is ensured to be received within an effective delay time range, and the process is called time integrity check. Because the terminals in the a664 network are asynchronous terminals and have respective clock sources, the clock bias between the transmitting terminal and the receiving terminal must be calculated, so that a clock bias calculation method between the source terminal and the destination terminal needs to be defined, and message time integrity verification is completed.
Disclosure of Invention
Aiming at the problems in the background technology, the technical scheme of the invention provides a method for checking the time integrity of the ARINC664 network receiving terminal, which ensures the time integrity of the message when the message is transmitted in the network.
In order to achieve the above purpose, the invention is realized by adopting the following technical scheme.
An ARINC664 network receiving terminal time integrity checking method, the method comprising:
s1, acquiring a time DTS when a first byte of a message leaves a sending terminal;
s2, calculating time deviation TOS between the transmitting terminal and the receiving terminal;
s3, obtaining a delay upper bound T_word of the message in the switch network;
s4, if the clock of the receiving end when the first byte of the message reaches the receiving terminal is within the upper limit of DTS+TOS+T_word, the time integrity check of the message is considered to pass; when the receiving end clock exceeds the upper bound of DTS+TOS+T_word, the receiving end discards the message if the time integrity check of the message is considered to be failed.
The technical scheme of the invention is characterized in that:
(1) S1 specifically comprises the following steps:
the data DTS is obtained by time stamping at the moment when the first byte of the message leaves the transmitting terminal.
(2) S2, when calculating time deviation TOS between a transmitting terminal and a receiving terminal:
s21, firstly, an application software is resident on a general processing module GPM in an ARINC664 network and used for constructing and sending a time request message, and the general processing module GPM is used as a time Server;
s22, other devices capable of carrying out network communication in the ARINC664 network are set as time Client sides, each time Client side is resident with an application software and is responsible for constructing a time response message and sending the time response message to a time Server;
s23, defining a time Server to timestamp the message when the first byte of the message leaves the sending terminal, and a time Client to timestamp the message when the first byte of the time request message is received;
s24, constructing a time response message by the Client of the time Client, recording the values of the two time stamps in the time response message, and sending the time response message to the Server of the time Server;
s25, after receiving the time response message, the time Server constructs a time stamp list message, puts the time stamp data collected from each time Client into the time stamp list message, then sends the time stamp list message to each time Client, and finally the time Client calculates the time deviation.
(3) In S2, a terminal connected with a time-recording server is used as a transmitting terminal, and a terminal connected with a time Client1 and a Client2 is used as a receiving terminal;
assume that the time deviation among local clocks of the time service Server, the first time Client1 and the second time Client2 satisfies the following relationship:
the time of the first time Client1 is delta 1 faster than the time of the time Server; the time of the second time Client2 is delta 2 faster than the time of the time Server; and the time of the first time Client1 is faster than the time of the second time Client2, the time deviation tos=Δ1- Δ2 between the first time Client1 and the second time Client2.
(4) The record D1 represents the switch network path delay from the time Server to the first time Client1, and the record D3 represents the switch network path delay from the time Server to the second time Client 2;
the time Server sends a time request message to the first time Client1, and marks the time stamp of the message when the first byte of the time request message leaves, and marks the time stamp as T1, and the first time Client1 marks the time stamp of the message when the first byte of the time request message is received, and marks the time stamp as T2;
the time Server sends a time request message to the second time Client2, and marks the time stamp of the message when the first byte of the time request message leaves, and marks the time stamp as T1', and the second time Client2 marks the time stamp of the message when the first byte of the time request message is received, and marks the time stamp as T2';
then:
the time offset between the first time Client1 and the second time Client2 is:
TOS＝Δ1-Δ2
＝T2-T1-D1-(T2'-T1'-D3)
＝T2-T1-(T2'-T1')-D1+D3
(5) The maximum value of TOS is:
TOS max ＝T2-T1-(T2'-T1')-SC1delay min +SC2delay max
wherein SC1delay min Representing the minimum delay of the switch network from the time Server to the first time Client1, SC2delay max Representing the maximum delay of the switch network from the time Server to the second time Client2.
(6) The ARINC664 switch network consists of a plurality of switches; s3, obtaining a delay upper bound T_word of the message in the switch network, wherein the delay upper bound T_word is specifically:
s31, determining the maximum time interval from the time when the first byte of the data frame reaches the input port of a certain switch to the time when the last byte of the data frame reaches the input port of the switch, and marking as L1;
s32, determining the time interval from the last byte of the data frame to the input port of the switch to the time interval from the first byte of the data frame to the public buffer of the switch, and marking as L2;
s33, determining the maximum delay of the data frame at the output port of the switch, and marking as L3;
s34, calculating the maximum delay of the data frame at the switch according to the sum of L1, L2 and L3;
s35, determining the delay upper bound of the data frames in the ARINC664 switch network according to the number of switches contained in the ARINC664 switch network and the maximum delay of the data frames in each switch.
(7) Virtual links define a logically unidirectional connection from a source terminal to one or more target terminals, how many target terminals there are VL paths for a virtual link, and from source terminal to a target terminal are considered a VL Path;
s31:
wherein VL (VL) i Representing the ith virtual link, node representing one of the output ports of the switch, called switch node, VL i The e-node represents the set of virtual links that pass from the switch node,is the maximum frame length allowed by the ith virtual link, C i Is the bandwidth of the ith virtual link at its input port;
s32: the value of L2 is the maximum jitter time of the switch configuration.
(8) In S33, the specific procedure for calculating L3 is as follows:
(a) Calculating an arrival curve for each VLPath, one hop representing the exit from one device to the next;
the single virtual link arrival curve is expressed as: alpha (t) =sigma+ρt
Where σ is the maximum flow that can be reached in the bursty flow and ρ is the upper slope limit of the flow increase;
(b) Calculating a service profile for each switch, the service profile comprising: service profile beta for high priority data frames h (t) and service profile beta for low priority data frames l (t)；
β h (t)＝R[t-T] + ,R＝C,
β l (t)＝R[t-T] + ,
Wherein [ T-T ]] + Represents [ T-T ] when T-T is greater than zero] + Equal to T-T, when T-T is less than or equal to zero, [ T-T ]] + Is equal to zero and is equal to zero,for the maximum frame length of all low priority virtual links output from the same node,for the sum of the bandwidths of all high priority virtual links through the node, +.>To pass throughThe sum of the burst data volumes of all high priority virtual links of the node, T tech Is the technical delay of the switch, C is the bandwidth of the output port of the switch;
(c) Determining aggregate arrival curve alpha for all virtual links within a single input port through the same output port SL (t)；
Wherein SL represents all virtual links through one switch input port of the same switch node, iE SL represents virtual link VL through that input port i ；
The inflection point e of the aggregation curve is expressed as:
wherein sigma i Is the ith virtual link VL i Burst traffic through the input port; ρ i Is the ith virtual link VL i Is a bandwidth of (a); c is the rate of the input port;
(d) Sorting inflection points of arrival curves of all N input ports passing through the same output port in ascending order by using a grouping technology, wherein the sorted inflection points are marked as E g (g=1, 2,., N), N is the number of switch ports; the input port aggregate data flow after the grouping ascending sequence is converged at the switch node to obtain an arrival curve at the switch node;
by calculating the maximum horizontal distance between the arrival curve and the service curve of the switch node, the maximum horizontal delay D caused by queuing competition of the data flow at the switch node can be obtained max Denoted as T 3 。
(9) (a) calculating an arrival curve for each VLPath, specifically:
(a1) First jump to reach curve for each VLPath:
wherein S is max Representing the maximum frame length allowed by the virtual link, BAG being the intrinsic parameter minimum frame interval of the virtual link, max_jitter representing the maximum Jitter time on the virtual link, max_jitter being Jitter at the first hop switch ES Representing the existing jitter of the virtual link in the source terminal;
starting from the second hop, using the formula obtained in the following steps as an arrival curve of the virtual link;
(a2) Calculating an arrival curve of the VLPath next hop:
an arrival curve representing traffic entering switch node n, < >>Is the arrival curve of the next node passing after the nth node; they satisfy the following relationship:
wherein,is the maximum waiting time of the aggregate arrival curve in the node,/-, for>
(10) Maximum horizontal delay D max The calculation process of (1) is as follows:
with inflection point E x At E x To the left of (a), the slope of the arrival curve of the switch is equal to or greater than the slope of the service curve, at E x The slope of the arrival curve of the exchange is smaller than the slope of the service curve;
when t=e x Obtaining the maximum horizontal distance D between the arrival curve and the service curve of the switch node max Maximum delay at the switch output port as a data frame:
wherein T and R are parameters of the service curve of the switch; the intermediate parameter y is calculated according to the following equation:
in the method, in the process of the invention,m 1 is the number of 10Mbps ports in the switch, m 2 Is the number of 100Mbps ports in the switch, m 1 +m 2 ＝N-s+1，m 2 Firstly, 1 is reduced along with the addition of 1 to s, when m 2 After decreasing to 0, m 1 Begin to decrease 1, C as s increases by 1 1 Is 10Mbps, C 2 Is 100Mbps, r 0 ＝E 0 ＝0。
(11) The maximum delay of VL Path in a switch is calculated first, and the maximum delay of the switch through which VLi passes should be calculated according to the following formula:wherein (1)>Indicating the maximum delay for VLi to pass through switch j.
(12) The upper end-to-end delay bound of a VL Path within a network of switches is equal to the sum of the maximum delays of all switches that the VL Path passes:
wherein SW is j ∈PATH i Representing the ith VLPATH i A set of switches traversed;representing the ith VLPATH i Maximum delay through the jth switch.
The invention defines the time management function, and designs the ARINC664 network receiving terminal time integrity checking method, so that the message is received within a reasonable delay range, certain data transmission errors are avoided, and the time integrity of the ARINC664 network terminal is ensured.
Drawings
FIG. 1 is a schematic diagram of a time management function according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an interaction process between a time server and a time client according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a time service and a time client according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an arrival curve of a virtual link at an output port of a terminal system according to an embodiment of the present invention;
FIG. 5 is a schematic representation of the arrival profile of a polymeric stream through the same inlet provided by an embodiment of the present invention;
FIG. 6 is a schematic diagram of a corner ordering of an aggregate flow according to an embodiment of the present invention;
fig. 7 is a schematic diagram of an arrival curve after using a grouping technique according to an embodiment of the present invention;
FIG. 8 shows E provided by an embodiment of the invention x A delay calculation schematic diagram when the delay calculation is not less than T;
FIG. 9 shows E provided by an embodiment of the invention x <Delay calculation at T is schematic.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The invention defines a rule for guaranteeing the time integrity of a receiving terminal in the end-to-end data transmission process of an A664 network: if the clock of the receiving end when the first byte of the message arrives at the receiving terminal is within the upper limit of DTS+TOS+T_word, the time integrity check of the message is considered to pass; when the receiving end clock exceeds the upper bound of DTS+TOS+T_word, the receiving end discards the message if the time integrity check of the message is considered to be failed. Wherein: the DTS indicates the time when the first byte of the message leaves the sending terminal, and the data can be obtained by performing time stamping at the time when the first byte of the message leaves the sending terminal, t_word indicates the delay upper bound of the message in the switch network, and when performing time integrity check, t_word indicates the maximum delay of the switch network that the non-time management message passes from the sending terminal to the receiving terminal, and the calculation method will be described below. The present invention refers to the time request message, the time response message and the time stamp list message which are collectively called as time management message, and the time request message and the time response message are single frame data.
In order to calculate clock skew TOS, the present invention firstly constructs a network time management function, as shown in fig. 1, first, an application software resides on a General Processing Module (GPM) in the a664 network to construct and send a time request message, at this time, the GPM is used as a time Server, other devices capable of performing network communication in the a664 network are all set as time clients (clients), each time Client resides in an application software, and is responsible for constructing a time response message and sending the time response message to the time Server. Defining the time service end to timestamp the message at the moment when the first byte of the message leaves the sending terminal, and marking as T 1 The time client stamps the first byte of the time request message at the time of receipt of the message, denoted T 2 Then time clientThe end constructs a time response message and stamps T 1 、T 2 The values of the time stamp list information are recorded in the time response information, the time response information is sent to the time server, and the time server receives the time response information and then constructs the time stamp list information to collect the time stamp T from each time client 1 、 T 2 The data is put into a time stamp list message (for example, client1T1, client1T2, client2T1, client2T2, client3T1, client3T2 and …), the time stamp list message is sent to each time Client, and finally the time Client executes a time deviation calculation function.
The following describes a specific calculation procedure of the clock skew in the present invention.
Assume that the time offset between the local clocks of Server, client, client2 satisfies the following relationship: the Client1 time is faster than the Server time, i.e., the timing value is greater than Δ1; the Client2 time is faster than the Server time, i.e., the timing value is greater than Δ2; and Client1 is faster than Client2. (otherwise derived similarly) then the time offset between Client1 and Client2 is tos=Δ1- Δ2. As shown in fig. 3, let Δ1 be the length from point E to point B, the time at point a is T1, the time at point C is T2, and the time at point D is the time corresponding to the client when the client is time-stamped with a time stamp T2, that is, the path delay is the length from point a to point D, and also the length from point B to point C, and as shown in fig. 2, D1 represents the path delay from the server to the client1, and D3 represents the path delay from the server to the client2, it is possible to obtain:
the time offset between Client1 and Client2 is therefore:
TOS＝Δ1-Δ2
＝T2-T1-D1-(T2'-T1'-D3)
＝T2-T1-(T2'-T1')-D1+D3 (2)
both D1 and D3 in the formula are switch network path delays, which cannot be measured accurately, and because the path delays generated by the frame propagation in the switch network after the network topology of a664 and the network route determination of the frame may be both the maximum delay (queued at the end of the queue) and the minimum delay (queued at the front of the queue), if the time integrity of the receiving terminal is to be checked more reasonably, i.e. when the receiving terminal is not waiting for a message any more reasonably, D3 in the time offset calculation formula is amplified to the maximum delay of the switch network passing from Server to Client2, and D1 is reduced to the minimum delay of the switch network passing from Server to Client 1.
At this time, if the clock of the receiving end when the first byte of the defined message arrives at the receiving terminal and is within the upper bound of DTS+TOS+T_word, the time integrity check of the message is considered to pass; when the receiving end clock exceeds the upper bound of DTS+TOS+T_word, the receiving end discards the message if the time integrity check of the message is considered to be failed. At this time consider the upper bound TOS of the calculated TOS max The calculation formula is as follows:
TOS max ＝T2-T1-(T2'-T1')-SC1delay min +SC2delay max (3)
wherein SC1delay min Representing the minimum delay of the switch network from the time Server to the first time Client1, SC2delay max Representing the maximum delay of the switch network from the time Server to the second time Client2.
The specific calculation process of the maximum delay and the minimum delay of the switch network is as follows:
calculating the upper bound of delay generated when the data frame is transmitted by the switch network, namely calculating the upper bound of end-to-end delay of the data frame on any one VL Path in the switch network, wherein the upper bound is defined as the time difference between the following two events:
(1) The first byte of the data frame enters the input port of the first switch of the VL Path;
(2) The first byte of the data frame exits from the output port of the last switch of the VL Path.
The term Node (Node) is referred to, and the Node is explained as follows:
starting node: a source terminal of VL; termination node: a destination terminal of the VL; switch node: VL passes through a certain output port of the switch, and each switch has N nodes.
In particular, the time delay of the switch system for transmitting data is mainly divided into three parts,
the first part is the delay of the switch data receiving module, denoted as L 1 The duration between the following two events is described:
(1) The first byte of the frame (including the first 20 bytes, the beginning of the frame delimiter, and the inter-frame gap) arrives at the input port of the switch;
(2) The last byte of the frame arrives at the input port of the switch (including the 20 bytes of the preamble, the start of the frame delimiter, and the interframe gap);
L 1 representing the maximum value of the duration.
Frames passing through one switch output port (switch node) may come from one or more switch input ports having different bandwidths. L can be calculated according to the following formula 1 ：
VL i E node represents VL i From the point of view of the switch,is VL (VL) i Maximum frame length of C i Is VL (VL) i Bandwidth at the switch input port.
The second part is that when all ports of the exchanger are input with data frames with the maximum frame length, the interior of the exchanger needs to sequentially perform error filtering and flow control, the data frames are stored in a public buffer area, delay is introduced when waiting for the processing process of the data frames of other ports, and the delay is marked as L 2 Described are the durations of the following two events:
(1) The last byte of the frame arrives at the input port of the switch;
(2) The first byte of the frame arrives in the common buffer;
L 2 indicating that within this durationThe maximum jitter due to contention between frames is a fixed value, which is relevant to the switch design.
The third part is the delay of the output port of the exchanger, which is marked as L 3 The output buffer collects all VLs output from the ports, each output port consisting of two queues, one for buffering data frame traffic for high priority VLs and one for buffering data frame traffic for low priority VLs. In this module, the frames all use a first-in first-out scheduling pattern. For any priority, a network algorithm can be adopted to respectively give service curves of high-priority and low-priority traffic. In the high-low priority queues, all high-low priority VLs are aggregated into one data stream each: that is, for each priority, the output queue is modeled as an aggregation server that provides an arrival curve to aggregate all VLs within that priority.
Step one: an arrival curve of the first hop of each VL Path is calculated, one hop representing the exit from one device (switch or end system) to the next device for each VL Path.
The single virtual link arrival curve may be represented in the form: alpha (t) =sigma+ρt (5)
Wherein: σ is the maximum flow that can be reached in the bursty flow and ρ is the upper slope limit of the flow increase.
Fig. 4 is an arrival curve of a virtual link at an output port of a source terminal, where the upward shift of the curve is a rational amplification of the arrival traffic, and the first arrival curve of VL Path may be obtained as follows:
wherein:S max is the intrinsic parameter frame interval representing the maximum frame length, the BAG being the virtual link, and max_jitter should be Jitter at the first hop switch ES Representing VL i Jitter already present in the source terminal. The formula from step four should be used as the arrival curve for VL starting from the second hop.
Step two: calculating a service curve of each switch node, and for an aggregate data flow of a high priority VL, waiting for a frame of a low priority being transmitted due to non-preemption at one switch node; for the aggregated data stream of the low priority VL, it is necessary to wait for all the aggregated data streams of the high priority VL to be transmitted, so that a service profile of the aggregated data stream of each priority VL is obtained as follows:
where L ε L represents a low priority VL through the same node i l.epsilon.H represents a high priority VL through the same node i The method comprises the steps of carrying out a first treatment on the surface of the When T is more than or equal to 0 and less than or equal to T tech In the time-course of which the first and second contact surfaces,when t>T tech When (I)>T tech The technical delay of the switch is a fixed value and is determined by the design of the switch.
The service profile for each priority can be written in the form of:
β h (t)＝R[t-T] + ,R＝C,
β l (t)＝R[t-T] + ,
wherein,maximum value of maximum frame length of all low-priority VLs output from the same node;all high priority VLs for the node passing i Is a sum of bandwidths (bytes/ms). T (T) tech Is the technical delay of the switch. C is the bandwidth of the node.
Step three: and calculating an aggregate arrival curve of each node. For the output ports of the switch, different virtual links cannot be transmitted simultaneously, so that the data transmission is not only constrained by the bandwidth of the link but also constrained by the aggregate arrival curve, and the data flow traversing on the same node can be used as an aggregate flow, as shown in fig. 5, which shows the arrival curve α of VL aggregate flows passing through the same switch input ports SL (t) expressed as:
wherein SL represents all virtual links through one switch input port of the same switch node, iE SL represents virtual link VL through that input port i 。
The inflection point e is expressed as:
wherein: sigma (sigma) i Is VL (VL) i Burst traffic through the input port; ρ i Is VL (VL) i Is a bandwidth of (a); c is the rate of the input port.
When calculating the aggregate arrival curve, the inflection points of the arrival curves of all N input ports passing through the same output port are sequenced in ascending order by using a grouping technology, and the sequenced inflection points are marked as E g (g=1, 2,., N) (N is a switchAs shown in fig. 6). The input port aggregate data flow after the packet ascending order is converged at the switch node, and the arrival curve at the switch node can be obtained, as shown in fig. 7, wherein
Thus, the arrival curve and the service curve of the switch node are obtained, and the maximum delay caused by queuing competition of the data flow at the switch node can be obtained by calculating the maximum horizontal distance between the arrival curve and the service curve, as shown in fig. 7 and 8, and the delay is marked as D max 。
In both fig. 8 and 9 there is an inflection point E x On the left side of Ex, the slope of the arrival curve is equal to or greater than the slope of the service curve, and on the right side of Ex, the slope of the arrival curve is less than the slope of the service curve.
When t=ex, the maximum horizontal distance D between the arrival curve and the service curve can be obtained max Namely L 3 。
T and R are service curves of the switch nodes (β (T) =r [ T-T ]] + ) Is a parameter of (a).
The intermediate parameter y is calculated according to the following equation:
wherein: m1 is the number of 10Mbps ports in the switch, m2 is the number of 100Mbps ports in the switch, m1+m2=n-s+1, m2 decreases by 1 with s first with 1, when m2 decreases to 0, m1 starts decreasing by 1 with s with 1, C1 is 10Mbps, C2 is 100Mbps, r0=e0=0.
Step four: and calculating an arrival curve of the next node. By usingVL representing ingress node n i Data flow toReach the curve with->Representing VL i The data flow leaves the arrival curve of node n, i.e. the arrival curve of the next node after passing through node n. They satisfy the following relationship:
wherein,d is node n max Then it is possible to obtain: />
Wherein,
step five: the maximum delay of VL Path in a switch is calculated first, and the maximum delay of the switch through which VLi passes should be calculated according to the following formula:
wherein,representing the maximum delay of VLi through switch j, L1 represents the upper limit of frame delay in the input module of the switch; l2 represents the upper limit of frame jitter in the address queue; l3 represents the maximum horizontal distance D between the switch node arrival curve and the service curve max 。
The upper end-to-end delay bound of a VL Path within a switch network should be equal to the sum of the maximum delays of all switches that the VL Path passes through, i.e.:
SW j ∈PATH i representing VLPATH i A set of switches traversed.
The minimum delay of the switch is a case where no virtual link contention is considered when the frame passes through the switch, at which time L 1 、L 2 The calculation method is the same as that above, L 3 Is 0.
The invention designs the ARINC664 network receiving terminal time integrity checking method by defining the time management function, so that the message is received within a reasonable delay range, certain data transmission errors are avoided, and the time integrity of the ARINC664 network terminal is ensured.
While the preferred embodiments of the present invention have been described in detail, the present invention is not limited to the above embodiments, and various changes may be made without departing from the spirit of the present patent within the knowledge of those skilled in the art.
Claims (5)
1. An ARINC664 network receiving terminal time integrity checking method, comprising:
s1, acquiring a time DTS when a first byte of a message leaves a sending terminal;
s2, calculating time deviation TOS between the transmitting terminal and the receiving terminal;
s21, firstly, an application software is resident on a general processing module GPM in an ARINC664 network and used for constructing and sending a time request message, and the general processing module GPM is used as a time Server;
s22, other devices capable of carrying out network communication in the ARINC664 network are set as time Client sides, each time Client side is resident with an application software and is responsible for constructing a time response message and sending the time response message to a time Server;
s23, defining a time Server to timestamp the message when the first byte of the message leaves the sending terminal, and a time Client to timestamp the message when the first byte of the time request message is received;
s24, constructing a time response message by the Client of the time Client, recording the values of the two time stamps in the time response message, and sending the time response message to the Server of the time Server;
s25, after receiving the time response message, the time Server constructs a time stamp list message, puts the time stamp data collected from each time Client into the time stamp list message, then sends the time stamp list message to each time Client, and finally the time Client calculates the time deviation;
the terminal connected with the time counting server is used as a transmitting terminal, and the terminal connected with the time Client1 and the Client2 is used as a receiving terminal;
assume that the time deviation among local clocks of the time service Server, the first time Client1 and the second time Client2 satisfies the following relationship:
the time of the first time Client1 is delta 1 faster than the time of the time Server; the time of the second time Client2 is delta 2 faster than the time of the time Server; and the time of the first time Client1 is faster than the time of the second time Client2, the time deviation tos=Δ1- Δ2 between the first time Client1 and the second time Client 2;
the record D1 represents the switch network path delay from the time Server to the first time Client1, and the record D3 represents the switch network path delay from the time Server to the second time Client 2;
the time Server sends a time request message to the first time Client1, and marks the time stamp of the message when the first byte of the time request message leaves, and marks the time stamp as T1, and the first time Client1 marks the time stamp of the message when the first byte of the time request message is received, and marks the time stamp as T2;
the time Server sends a time request message to the second time Client2, and marks the time stamp of the message when the first byte of the time request message leaves, and marks the time stamp as T1', and the second time Client2 marks the time stamp of the message when the first byte of the time request message is received, and marks the time stamp as T2';
then:
the time offset between the first time Client1 and the second time Client2 is:
TOS＝Δ1-Δ2
＝T2-T1-D1-(T2'-T1'-D3)
＝T2-T1-(T2'-T1')-D1+D3；
s3, obtaining a delay upper bound T_word of the message in the switch network;
the ARINC664 switch network consists of a plurality of switches; the method comprises the steps of obtaining a delay upper bound T_word of a message in a switch network, wherein the delay upper bound T_word is specifically as follows:
s31, determining the maximum time interval from the time when the first byte of the data frame reaches the input port of a certain switch to the time when the last byte of the data frame reaches the input port of the switch, and marking as L1;
s32, determining the time interval from the last byte of the data frame to the input port of the switch to the time interval from the first byte of the data frame to the public buffer of the switch, and marking as L2;
s33, determining the maximum delay of the data frame at the output port of the switch, and marking as L3;
the specific procedure for calculating L3 is as follows:
(a) Calculating an arrival curve for each VLPath, one hop representing the exit from one device to the next;
the single virtual link arrival curve is expressed as: alpha (t) =sigma+ρt
Where σ is the maximum flow that can be reached in the bursty flow and ρ is the upper slope limit of the flow increase;
(b) Calculating a service profile for each switch, the service profile comprising: service profile beta for high priority data frames h (t) and service profile beta for low priority data frames l (t)；
Wherein [ T-T ]] + Represents [ T-T ] when T-T is greater than zero] + Equal to T-T, when T-T is less than or equal to zero, [ T-T ]] + Is equal to zero and is equal to zero,maximum frame length of all low priority virtual links output from the same node, +.>For the sum of the bandwidths of all high priority virtual links through the node, +.>T is the sum of the burst data volumes of all high priority virtual links through the node tech Is the technical delay of the switch, C is the bandwidth of the output port of the switch;
(c) Determining aggregate arrival curve alpha for all virtual links within a single input port through the same output port SL (t)；
Wherein SL represents all virtual links through one switch input port of the same switch node, iE SL represents virtual link VL through that input port i ；
The inflection point e of the aggregation curve is expressed as:
wherein sigma i Is the ith virtual link VL i Burst traffic through the input port; ρ i Is the ith virtual link VL i Is a bandwidth of (a); c is the rate of the input port;
(d) Sorting inflection points of arrival curves of all N input ports passing through the same output port in ascending order by using a grouping technology, wherein the sorted inflection points are marked as E g (g=1, 2,., N), N is the number of switch ports; the input port aggregate data flow after the grouping ascending sequence is converged at the switch node to obtain an arrival curve at the switch node;
by calculating the maximum horizontal distance between the arrival curve and the service curve of the switch node, the maximum horizontal delay D caused by queuing competition of the data flow at the switch node can be obtained max Denoted as L3;
s34, calculating the maximum delay of the data frame at the switch according to the sum of L1, L2 and L3;
s35, determining the delay upper bound of the data frames in the ARINC664 switch network according to the number of switches contained in the ARINC664 switch network and the maximum delay of the data frames in each switch; the upper end-to-end delay bound of a virtual link within a network of switches is equal to the sum of the maximum delays of all switches through which the virtual link passes:
wherein SW is j ∈PATH i Representing the ith virtual link VLPATH i A set of switches traversed;representing the ith virtual link VLPATH i Maximum delay through the jth switch;
s4, if the clock of the receiving end when the first byte of the message reaches the receiving terminal is within the upper limit of DTS+TOS+T_word, the time integrity check of the message is considered to pass; when the receiving end clock exceeds the upper bound of DTS+TOS+T_word, the receiving end discards the message if the time integrity check of the message is considered to be failed.
2. The method for checking the time integrity of an ARINC664 network receiving terminal according to claim 1, wherein S1 is specifically:
the data DTS is obtained by time stamping at the moment when the first byte of the message leaves the transmitting terminal.
3. The method for checking the time integrity of an ARINC664 network receiving terminal as claimed in claim 1, wherein TOS has an upper bound of:
TOS max ＝T2-T1-(T2'-T1')-SC1delay min +SC2delay max
wherein SC1delay min Representing the minimum delay of the switch network from the time Server to the first time Client1, SC2delay max Representing the maximum delay of the switch network from the time Server to the second time Client2.
4. The ARINC664 network reception terminal time integrity verification method according to claim 1, wherein the virtual link defines a logically unidirectional connection from a source terminal to one or more target terminals, how many target terminals there are VL paths for a virtual link, and from source terminal to target terminal is considered a VL Path;
s31:
wherein VL (VL) i Representing the ith virtual link, node representing one of the output ports of the switch, called switch node, VL i The e-node represents the set of virtual links that pass from the switch node,is the maximum frame length allowed by the ith virtual link, C i Is the bandwidth of the ith virtual link at its input port;
s32: the value of L2 is the maximum jitter time of the switch configuration.
5. The method of time integrity checking of ARINC664 network receiving terminals as claimed in claim 1, wherein (a) calculating an arrival curve of each VLPath comprises:
(a1) First jump to reach curve for each VLPath:
wherein S is max Representing the maximum frame length allowed by the virtual link, BAG being the intrinsic parameter minimum frame interval of the virtual link, max_jitter representing the maximum Jitter time on the virtual link, max_jitter being Jitter at the first hop switch ES Representing the existing jitter of the virtual link in the source terminal;
starting from the second hop, using the formula obtained in the following steps as an arrival curve of the virtual link;
(a2) Calculating an arrival curve of the VLPath next hop:
an arrival curve representing traffic entering switch node n, < >>Is the arrival curve of the next node passing after the nth node; they satisfy the following relationship:
wherein,is the maximum waiting time of the aggregate arrival curve in the node,/-, for>
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210839595.4A CN115412513B (en) | 2022-07-15 | 2022-07-15 | ARINC664 network receiving terminal time integrity checking method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210839595.4A CN115412513B (en) | 2022-07-15 | 2022-07-15 | ARINC664 network receiving terminal time integrity checking method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115412513A CN115412513A (en) | 2022-11-29 |
| CN115412513B true CN115412513B (en) | 2023-12-12 |
Family
ID=84157178
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210839595.4A Active CN115412513B (en) | 2022-07-15 | 2022-07-15 | ARINC664 network receiving terminal time integrity checking method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115412513B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102780581A (en) * | 2012-07-20 | 2012-11-14 | 北京航空航天大学 | AFDX (Avionics Full Duplex Switched Ethernet) end-to-end delay bound claculation method based on random network calculus |
| CN105337885A (en) * | 2015-09-28 | 2016-02-17 | 北京信息科技大学 | Multistage grouping worst delay calculation method suitable for credited shaping network |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011159210A1 (en) * | 2010-06-17 | 2011-12-22 | Saab Ab | Ethernet for avionics |
| US9769082B2 (en) * | 2015-04-01 | 2017-09-19 | Honeywell International Inc. | System and method for network bandwidth, buffers and timing management using hybrid scheduling of traffic with different priorities and guarantees |
-
2022
- 2022-07-15 CN CN202210839595.4A patent/CN115412513B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102780581A (en) * | 2012-07-20 | 2012-11-14 | 北京航空航天大学 | AFDX (Avionics Full Duplex Switched Ethernet) end-to-end delay bound claculation method based on random network calculus |
| CN105337885A (en) * | 2015-09-28 | 2016-02-17 | 北京信息科技大学 | Multistage grouping worst delay calculation method suitable for credited shaping network |
Non-Patent Citations (5)
| Title |
|---|
| "Optimal scheduling policy for AFDX End-Systems with virtual links of identical bandwidth allocation gap size";R. Nishi, G. Zhu and Y. Savaria;《2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)》;全文 * |
| "Comparison of AFDX and audio video bridging forwarding methods using network calculus approach";Zhao L, He F, Lu J.;《2017 IEEE/AIAA 36th Digital Avionics Systems Conference (DASC)》;全文 * |
| "Worst-case delay bounds with fixed priorities using network calculus";Bouillard A, Junier A.;《Proceedings of the 5th International ICST Conference on Performance Evaluation Methodologies and Tools》;全文 * |
| "航空电子AFDX与AVB传输实时性抗干扰对比";赵琳, 何锋, 熊华钢;《北京航空航天大学学报》;全文 * |
| 伍志韬 ; 黄宁 ; 王学望 ; 胡宁 ; ."基于随机型网络演算的AFDX端端时延分析方法".《***工程与电子技术》.2013,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115412513A (en) | 2022-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4356261B2 (en) | Priority setting method and apparatus | |
| JP4841674B2 (en) | Method and apparatus for controlling latency variation in a packet forwarding network | |
| US9608927B2 (en) | Packet exchanging device, transmission apparatus, and packet scheduling method | |
| CN104429029A (en) | Delay-based traffic rate control in networks with central controllers | |
| CN112019392A (en) | Multi-feature stream access method for TSN (traffic service network) | |
| KR101990235B1 (en) | Method and system for traffic management in a network node in a packet switched network | |
| CN110086728B (en) | Method for sending message, first network equipment and computer readable storage medium | |
| US20050144309A1 (en) | Systems and methods for controlling congestion using a time-stamp | |
| CN114615211A (en) | Method for managing and scheduling time-sensitive network queue in spacecraft with wired and wireless integration | |
| US6647413B1 (en) | Method and apparatus for measuring performance in packet-switched networks | |
| CN107070815B (en) | Synchronous message queuing delay control method and device in SDN network | |
| Soni et al. | Optimizing network calculus for switched ethernet network with deficit round robin | |
| EP2690821A1 (en) | Method And Apparatus For Packet Buffering Measurement | |
| CN115412513B (en) | ARINC664 network receiving terminal time integrity checking method | |
| Patel | Performance analysis and modeling of congestion control algorithms based on active queue management | |
| CN115604193B (en) | Deterministic resource scheduling method and system in hot rolling control system | |
| Soni et al. | Integrating offset in worst case delay analysis of switched ethernet network with deficit round robbin | |
| CN115643220A (en) | Deterministic service transmission method and device based on jitter time delay | |
| CN115174445B (en) | Delay upper bound analysis method for data frames in ARINC664 switch network | |
| CN108966028B (en) | Anti-jitter method for dynamically adjusting play speed based on network condition | |
| CN114598668B (en) | System for time-triggered data exchange | |
| Ganjalizadeh et al. | Reliability enhancement for hard real-time communication in industrial packet-switched networks | |
| WO2024016175A1 (en) | Deterministic stream transmission method and system, and electronic device | |
| CN115297031A (en) | ARINC664 terminal message delay upper bound analysis method based on FIFO scheduling | |
| Kwak et al. | A new starting potential fair queuing algorithm with O (1) virtual time computation complexity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |