CN109219019B - Train communication network multi-hop scheduling method based on Ethernet - Google Patents

Train communication network multi-hop scheduling method based on Ethernet Download PDF

Info

Publication number
CN109219019B
CN109219019B CN201811189486.2A CN201811189486A CN109219019B CN 109219019 B CN109219019 B CN 109219019B CN 201811189486 A CN201811189486 A CN 201811189486A CN 109219019 B CN109219019 B CN 109219019B
Authority
CN
China
Prior art keywords
data
scheduling
time
node
master node
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
Application number
CN201811189486.2A
Other languages
Chinese (zh)
Other versions
CN109219019A (en
Inventor
白璐瑶
张慧勇
尹晓旭
巫伟男
魏星原
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Institute of Specialized Machinery
Original Assignee
Beijing Institute of Specialized Machinery
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Institute of Specialized Machinery filed Critical Beijing Institute of Specialized Machinery
Priority to CN201811189486.2A priority Critical patent/CN109219019B/en
Publication of CN109219019A publication Critical patent/CN109219019A/en
Application granted granted Critical
Publication of CN109219019B publication Critical patent/CN109219019B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H04W4/42—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for mass transport vehicles, e.g. buses, trains or aircraft
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69—Spread spectrum techniques
    • H04B1/707—Spread spectrum techniques using direct sequence modulation
    • H04B1/7073—Synchronisation aspects
    • H04B1/7075—Synchronisation aspects with code phase acquisition
    • H04B1/70754—Setting of search window, i.e. range of code offsets to be searched
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • H04W40/04—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W56/00—Synchronisation arrangements
    • H04W56/001—Synchronization between nodes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/20—Control channels or signalling for resource management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/56—Allocation or scheduling criteria for wireless resources based on priority criteria
    • H04W72/566—Allocation or scheduling criteria for wireless resources based on priority criteria of the information or information source or recipient
    • H04W72/569—Allocation or scheduling criteria for wireless resources based on priority criteria of the information or information source or recipient of the traffic information

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Small-Scale Networks (AREA)

Abstract

The invention relates to a train communication network multi-hop scheduling method based on Ethernet, belonging to the technical field of train-mounted network system control. The invention introduces FTT-SE in the train communication network. FTT-SE adopts a time triggering mode to transmit periodic data and adopts an event triggering mode to transmit non-periodic data, so that the organic combination of time triggering and event triggering is realized; and a master-slave scheduling mode is adopted, so that the bandwidth efficiency is high, online scheduling and a multi-master system are supported, a standard industrial Ethernet switch is compatible, and the real-time performance of train Ethernet communication is improved.

Description

Train communication network multi-hop scheduling method based on Ethernet
Technical Field
The invention belongs to the technical field of train-mounted network system control, and particularly relates to a train communication network multi-hop scheduling method based on Ethernet.
Background
The real-time transmission of control commands and monitoring of system states is a basic task of a train control network, the data transmitted in the train communication network are various, a considerable part of the data is data related to control, detection, diagnosis and the like of train running safety, and the train control network must provide corresponding real-time communication to meet the time limit of information transmission. As train networks carry data continuously increasing, traditional train communication networks such as MVBs, WTBs, ARCNET and the like will have difficulty meeting the requirements. Ethernet has become one of the main research directions of train communication networks in the future due to its advantages of high transmission rate, good compatibility, low cost, etc. At present, the real-time problem of the Ethernet applied to the train communication network mainly exists in a data link layer, and the queuing mechanism of a switch brings uncertainty of data transmission. In order to meet the real-time requirements for train communication network data transmission in the IEC61375-3-4 standard, a specific real-time improvement mechanism is required.
Disclosure of Invention
Technical problem to be solved
The technical problem to be solved by the invention is as follows: how to improve the real-time performance of train ethernet communication.
(II) technical scheme
In order to solve the technical problem, the invention provides an Ethernet-based train communication network multi-hop scheduling method, which comprises the following steps:
firstly, reserving a management window at the beginning of each basic period for subsequent data scheduling; in a management window, each slave node informs a master node of data to be sent in a mode of 'multi-port to one-port', after the master node receives the data from the slave nodes, the master node broadcasts and sends a trigger message TM to each slave node in a mode of 'one-port to multi-port', wherein the trigger message TM comprises scheduling information and clock synchronization information of a current basic period, and the master node can calculate a non-cross transmission path of each message by providing the master node with information of each slave node related to data exchange, including an address, a port and a data type;
secondly, performing priority queuing on the global data and the local data, performing schedulability analysis on messages in a priority queue according to a periodic data scheduling algorithm, and determining all messages of the current EC which can be scheduled;
thirdly, constructing a train communication network topology;
and fourthly, scheduling the periodic information between the master node and the slave nodes based on the constructed train communication network topology according to the schedulability analysis result of the second step.
Preferably, the method further comprises a fifth step of scheduling non-periodic data between the master node and the slave node based on the constructed train communication network topology: the method comprises the steps that real-time non-periodic data and non-real-time data are transmitted in a non-periodic phase window, the non-periodic data are sporadic and do not need to be triggered by TM information, when the real-time non-periodic data are transmitted, and the time of the non-periodic phase window is larger than a preset value, the non-real-time data are transmitted in a first-in first-out mode, a polling method is adopted in the non-periodic data transmission mode, in the periodic data transmission process, a slave node reports non-periodic information sent in the current basic period to a master node through a reverse transmission link, the master node performs unified scheduling, and the non-periodic data are transmitted after the periodic phase is finished and the non-periodic phase arrives.
Preferably, in the second step, the principle of the schedulability analysis is: the method comprises the steps that messages which can be scheduled and transmitted by a current EC are coded into a TM, messages which cannot be scheduled are buffered into a next EC to be processed, in a management window at the beginning of each EC, a main node sends the TM in a subnet in a multicast mode, period information to be transmitted is activated, a message queue to be forwarded is updated, and after all messages which can be scheduled by the current EC are determined, the main node constructs the TM and broadcasts the TM to each slave node.
Preferably, the third step is specifically: in each section of vehicle of the train communication network, each electronic control unit device is connected to a networking switch CS according to a star topology structure, and the networking switch CS divides the train communication network into a plurality of independent conflict domains; wherein the vehicle control units VCUs of the head and the tail of the two vehicles are simultaneously connected to the switch ETBN of the CS and the train communication network to be used as central control units CCU1 and CCU2, the CS and the ETBN both adopt industrial Ethernet switches, and the equipment in the vehicles comprises: a vehicle control unit VCU and a brake control unit BCU, wherein the vehicle control units VCU in the head and tail two trains TC1 and TC2 simultaneously serve as a central control unit CCU of the whole train;
CS and terminal equipment directly connected with the CS in each carriage form a sub-network, ETBN1 and CCU1 form a sub-network, each switch is provided with a main node for scheduling, ETBN1 is defined as a root switch of the train communication network, CCU1 is a root main node, a brake control unit BCU in each carriage is selected as the main node for scheduling the sub-network, CCUs of the first and the last two trains are respectively used as the main nodes of ETBN1 and ETBN2, the train communication network is divided into different hierarchical structures, and ETBN1 and CCU1 form a first layer of a multi-hop network; let switch ETBN2 and switch CS5 and the sub-networks formed by them be the second layer of the multi-hop network, and at the same time ETBN2 and CS5 serve as the child nodes of ETBN1, and similarly, let CS6 serve as the child nodes of ETBN2, CS2 and CS1 serve as the child nodes of CS5, and CS3 and CS4 serve as the child nodes of CS 6.
Preferably, when scheduling is performed in the fourth step, clock synchronization is maintained among the sub-networks, clock synchronization among the sub-networks is achieved through clock synchronization between the master node of each sub-network and the root master node CCU1, in a management window of each basic period, the root master node CCU1 broadcasts global trigger information GTM, according to the hierarchical structure established in the third step, each master node is responsible for forwarding the GTM to the sub-master nodes, after receiving the GTM, each master node sends TM to initialize the basic period of each sub-network, after receiving the TM message, the slave nodes send scheduled data according to a specified time, the data encoded into the TM is triggered in the corresponding basic period, each slave node and the master node in the same sub-network maintain clock synchronization, and the clock synchronization is completed through the TM message.
Preferably, when scheduling is performed in the fourth step, a fixed deadline is set for each master node according to the relative positions of each master node BCU and ETBN2 and the root master node ETBN1, if the master node does not receive the GTM before the deadline arrives, a new GTM is generated by the master node and sent to the sub-master nodes, clock synchronization of the whole multi-hop network is continuously completed, because each networking switch divides each carriage into an independent conflict domain, the TM is only sent in the sub-network, and mutual interference between the sub-networks is avoided; unlike other master nodes, the root master node CCU1 sends the GTM directly at the very beginning of each fundamental cycle, does not determine whether the deadline has been reached, and does not need to receive the GTM.
Preferably, when scheduling is performed in the fourth step, all subnets that pass through during the transmission of the global data perform continuous scheduling on the subnets in a basic period.
Preferably, when the scheduling in the fourth step is performed, the method is implementedThe real-time period information transmitted in the train communication network is mathematically described, the number of the period information which needs to be transmitted in the network is assumed to be N, and the ith period information m isiThe transmission task is described as
γi＝{mi(Ci,Di,Ti,OiPKi,Si,DSi,Pi,Li,ni),i＝1…N} (1)
Wherein, CiRepresents period information miThe transmission time of (c); diAnd TiRespectively represents miOff-time and period of, OiIndicating the time at which the initial phase of the data is scheduled, PKiRepresents miThe size of the data packet; siAnd DSiRespectively represents miA source node and a destination node; piIndicating a priority; l isiRepresents miSet of links through, niRepresents miNumber of link segments passed, Li＝{lk|k＝1…n}。
Preferably, when scheduling is performed in the fourth step, real-time periodic data must be transmitted within a periodic phase window, and considering delay epsilon of the switch, the working time of all links cannot exceed a periodic phase LSW-epsilon, where LSW is the size of the periodic phase window, and for a sending link, it must be ensured that the sum of the transmission time of the data must not exceed the size LSW-epsilon of the periodic phase window; for the receiving link, it is ensured that the time f when the last data transmission is completed is smaller than the size LSW of the periodic phase window.
Preferably, when scheduling is performed in the fourth step, real-time cycle data m is setiThe time of arrival at the output port of the switch is AiIn queue ratio miData m with high priorityjThe reception completion time on the reception link RL is Fj(ii) a Wherein A isi＝Oi+Ci，OiIs the initial phase of the data, namely the transmission starting time of the data on the transmission link; to ensure reception of high priority data, miAt the beginning of reception on the reception link RL, max { F }j,AiH, so miUpon completion of the receive link transmissionEngraved max { Fj,Ai}+CiSo m isiCan be scheduled in the current basic cycle and should also satisfy
max{Fj,Ai}+Ci≤LSW (3)
When the formulas (2) and (3) cannot be satisfied simultaneously, miWill be postponed to the next basic cycle scheduling when the scheduled high priority data arrives later than miWhen m is allowediPrioritise these scheduled high priority data receptions on the reception link RL;
after receiving the TM message, each slave node decodes the TM message, prepares periodic data required to be transmitted in the current basic period according to the information contained in the TM message, initializes the data to be transmitted when the period comes, and simultaneously receives the periodic data transmitted by other nodes.
(III) advantageous effects
The invention introduces FTT-SE in the train communication network. FTT-SE adopts a time triggering mode to transmit periodic data and adopts an event triggering mode to transmit non-periodic data, so that the organic combination of time triggering and event triggering is realized; and a master-slave scheduling mode is adopted, so that the bandwidth efficiency is high, online scheduling and a multi-master system are supported, a standard industrial Ethernet switch is compatible, and the real-time performance of train Ethernet communication is improved.
Drawings
FIG. 1 is a schematic diagram of the FTT-SE communications model;
FIG. 2 is a schematic diagram of a train communication network architecture in an embodiment of the present invention;
FIG. 3 is a flow chart of master node scheduling of the present invention;
FIG. 4 is a slave node scheduling flow diagram of the present invention;
fig. 5 is a diagram of a multi-hop scheduling model of the present invention.
Detailed Description
In order to make the objects, contents, and advantages of the present invention clearer, the following detailed description of the embodiments of the present invention will be made in conjunction with the accompanying drawings and examples.
Data transmission of a train communication network exists in a vehicle and between vehicles, the data transmission between the vehicles usually needs to cross a plurality of switches, a transmission path is complex, uncertainty of data transmission is increased, and therefore an effective scheduling mechanism is needed. As the amount of communication increases, network architectures of "multi-subnet" and "multi-switch" are proposed successively, and the network requiring interconnection of a plurality of switches to realize communication is a "multi-hop network". The invention discloses an Ethernet-based train communication network, belongs to a multi-hop network architecture, and provides an FTT-SE-based vehicle-mounted Ethernet multi-hop scheduling method, aiming at improving the real-time performance of train Ethernet communication.
The FTT-SE mechanism adopts a master-slave scheduling manner, as shown in fig. 1, a network is composed of a master node and a plurality of other slave nodes, and the slave nodes are scheduled by one master node. The network is divided into a plurality of micro-network sections through a switch, and links in the network are divided into Sending Links (SL) and Receiving Links (RL) according to the transmission direction of data by adopting a full-duplex working mode; the switch ports connected with the two ends of each link are fixed. And the main node schedules the data to be sent according to the priority sequence. The transmission time of data is divided into fixed time slices called Elementary Cycles (EC), and each Elementary Cycle is divided into two windows, namely a periodic phase and an aperiodic phase, and is responsible for the transmission of periodic data and aperiodic data respectively.
The invention provides a train communication network multi-hop scheduling method based on the FTT-SE mechanism, aiming at the situation that data transmission in a vehicle-mounted Ethernet passes through a plurality of switches, which comprises the following steps:
the first step is to reserve a management window at the beginning of each basic period for subsequent data scheduling
In the management window, each slave node informs the master node of data to be sent in a 'multi-port-to-one-port' mode, and after receiving the data from the slave node, the master node broadcasts a sending Trigger Message (TM) to each slave node in a 'one-port-to-multi-port' mode, wherein the sending Trigger Message (TM) comprises scheduling information and clock synchronization information of the current basic cycle. By providing the information of each slave node related to data exchange, including the information of addresses, ports, data types and the like, to the master node, the master node can calculate the non-cross transmission path of each message (namely, the source node and the destination node are not overlapped with other messages), and construct a parallel scheduling table according to the non-cross transmission path of each message, thereby increasing the scheduling throughput.
The aim of the vehicle-mounted Ethernet multi-hop scheduling mechanism is to effectively schedule global data and local data in a specified time window and ensure the real-time performance of data transmission in and among vehicles. Under the condition that the topology of the train network is fixed, the source node and the destination node of data transmission can uniquely determine the transmission path of the data.
And secondly, performing priority queuing on the global data and the local data according to an RM algorithm or an EDF algorithm, performing schedulability analysis on messages in a priority queue according to a periodic data scheduling algorithm, coding messages which can be scheduled and transmitted by the current EC into a TM, and buffering messages which cannot be scheduled into the next EC for processing. And in the management window at the beginning of each EC, the main node transmits TM in the subnet in a multicast mode, activates the period information to be transmitted and updates the message queue to be forwarded. Finally, after determining all the messages that the current EC can be scheduled, the master builds the TM and broadcasts to the slave nodes.
Thirdly, constructing a train communication network topology
As shown in fig. 2, a subway control network is taken as a prototype, and in each vehicle, each important electronic control unit device is connected to a networking switch CS according to a star topology structure, and the networking switch CS divides a train communication network into several independent conflict domains; the VCUs of the vehicle control units of the head and the tail of the vehicle are simultaneously connected to the CS and the ETBN to serve as a central control unit CCU1 and a CCU2, and the CS and the ETBN both adopt industrial Ethernet switches. The main devices in the vehicle include: a Vehicle Control Unit (VCU) and a Brake Control Unit (BCU), etc. Wherein, the Vehicle Control Units (VCUs) in the head and tail two-train TC1 and TC2 are simultaneously used as the central control unit CCU of the whole train, and the integrated video monitoring system (PIS-CCTV) is also arranged in the head and tail two-train.
The train communication network shown in fig. 2 is a "multi-hop network" composed of a plurality of switches and terminal devices. If the CS in each compartment and the terminal equipment directly connected with the CS form a sub-network, the ETBN1 and the CCU1 form a sub-network, wherein each exchanger needs a special main node to be responsible for scheduling, the brake control unit BCU in each compartment is selected as the main node to be responsible for scheduling of the sub-network, and the CCUs of the first and the last vehicles are respectively used as the main nodes of the ETBN1 and the ETBN 2. For convenience of scheduling, the multi-hop network is divided into different hierarchical structures, the ETBN1 and the CCU1 form a first layer of the multi-hop network, the ETBN1 is defined as a root switch of the multi-hop network, and the CCU1 is defined as a root main node; ETBN2 and CS5 and the sub-networks formed by them are the second layer of the multi-hop network, and ETBN2 and CS5 are the child nodes of ETBN1, and similarly, CS6 is the child node of ETBN2, CS2 and CS1 are the child nodes of CS5, and CS3 and CS4 are the child nodes of CS 6.
Fourthly, scheduling the periodic information between the master node and the slave node based on the constructed train communication network topology according to the schedulability analysis result of the second step
Compared with data transmission in vehicles, data transmission paths between vehicles are complex, and besides clock synchronization of nodes inside the sub-networks, clock synchronization between the sub-networks is also required. Clock synchronization between subnets is achieved by clock synchronization of the master node of each subnet with the root master node CCU 1. In the management window of each basic period, the root master node CCU1 broadcasts Global Trigger information (GTM), according to the above hierarchical structure, each master node is responsible for forwarding the GTM to the child master nodes, and after receiving the GTM, each master node sends the TM to initialize the basic period of each subnet. The slave node transmits scheduled data at a prescribed time after receiving the TM message. The data encoded into the TM should be triggered in the corresponding fundamental cycle, which requires that the slave and master nodes within the same subnet maintain clock synchronization, which will be done by the TM message.
In order to prevent GTM loss, a fixed cutoff time is set for each main node according to the relative position of each main node (BCU and ETBN2) and the root main node (ETBN1), if the main node does not receive the GTM before the cutoff time (timeout) arrives, a new GTM is generated by the main node and is sent to the sub-main nodes, and the clock synchronization of the whole multi-hop network is continuously completed. Because each networking switch divides each compartment into an independent conflict domain, TM is only transmitted in the sub-networks, and the sub-networks cannot interfere with each other. Unlike other master nodes, the root master node CCU1 sends the GTM directly at the very beginning of each fundamental cycle, without having to determine whether the deadline has been reached, and without having to receive the GTM.
In the process of transmitting global data, all subnets that need to pass through continuously schedule the global data in a basic period, and when a certain subnet in a transmission path does not schedule the global data, the global data may miss the deadline. For example, when the VCU4 in M4 sends a message to the BCU5 in TC1, it needs to pass through switches CS4, CS6, ETBN2, ETBN1, and CS5, and at this time, master nodes BCU4, BCU6, CCU2, CCU1, and BCU5 need to perform continuous scheduling in the same basic cycle.
Performing mathematical description on real-time period information transmitted in a train communication network, assuming that the number of the period information to be transmitted in the network is N, and performing mathematical description on ith period information miThe transmission task can be described as
γi＝{mi(Ci,Di,Ti,OiPKi,Si,DSi,Pi,Li,ni),i＝1…N} (1)
Wherein, CiRepresents period information miTransmission time (data length/bandwidth); diAnd TiRespectively represents miOff-time and period of, OiIndicating the time at which the initial phase of the data is scheduled, PKiRepresents miThe size of the data packet; siAnd DSiRespectively represents miA source node and a destination node; piIndicating a priority; l isiRepresents miSet of links through, niRepresents miNumber of link segments passed, Li＝{lk|k＝1…n}。
The link working mode is full duplex, the transmitting link and the receiving link work simultaneously, and in order to fully utilize the bandwidth, each link considers the transmitting link SL and the receiving link RL at the same time. In the train communication network, each data transmission path is fixed, and links passing through the path are relatively fixed. Because the switch port connected with each link is fixed, the transmission speed of data on the link is the same, and in order to avoid data queuing in the switch, only one data is transmitted in each transmission direction in each link.
The master node generates a transmission queue of periodic data according to scheduling algorithms such as RM/EDL and the like, and the queue is encoded into a TM message and transmitted to each slave node by the master node. And each slave node transmits data to the switch through the transmission link of the slave node according to the scheduling table, and after the data arrives at the switch, the data is queued at the corresponding output port and is ready to be transmitted to the destination node to receive the link.
Being transmitted within the current EC requires that the transmission of data on both the transmit and receive links be completed within the synchronization window of the current EC. Since the switch adopts a full-duplex mode of operation, the SL and RL operate simultaneously. Several issues need to be noted when constructing a periodic schedule:
(1) the real-time periodic data must be transmitted within the periodic phase window, and the working time of all links cannot exceed the periodic phase LSW-epsilon, which is the size of the periodic phase window, considering the delay epsilon of the switch. Therefore, for the transmission link, it must be ensured that the sum of the transmission times of the data must not exceed the size LSW-epsilon of the periodic phase window; for the receiving link, it should be guaranteed that the time f at which the last data transmission is completed is smaller than the size LSW of the periodic phase window. In conclusion, the following results
Figure BDA0001827109100000111
(2) In order to calculate the time when the transmission of data on the receiving link is completed, the receiving completion time of the high priority data packet in the queue needs to be considered. Setting real-time period data miThe time of arrival at the output port of the switch is AiIn queue ratio miData m with high priorityjOn the receive chain RLThe receiving completion time is Fj. Wherein A isi＝Oi+Ci，OiIs the initial phase of the data, namely the transmission starting time of the data on the transmission link; to ensure reception of high priority data, miAt the beginning of reception on the reception link RL, max { F }j,AiH, so miAt the moment when the transmission of the receiving link is completed, max { F }j,Ai}+Ci. Therefore miCan be scheduled in the current basic cycle and should also satisfy
max{Fj,Ai}+Ci≤LSW (3)
When the formulas (2) and (3) cannot be satisfied simultaneously, miWill be postponed to the next basic cycle scheduling, in order to improve the utilization rate of the link bandwidth, on the basis of not influencing the real-time performance of the high priority data, when the arrival time of the scheduled high priority data is later than miWhen m is allowediThese data (scheduled high priority data) are received on the receive link RL in preference to.
The flow of the scheduling algorithm of the master node for real-time periodic data is shown in fig. 3 by integrating the transmission process of data in and between vehicles.
After receiving the TM message, each slave node decodes the TM message, prepares periodic data to be transmitted in the current basic period according to the information contained in the TM message, initializes the data to be transmitted when the period comes, and receives the periodic data transmitted from other nodes, where the working flow of the slave node is as shown in fig. 4.
Fifthly, scheduling non-periodic data between the master node and the slave node based on the constructed train communication network topology
The real-time non-periodic data and the non-real-time data are transmitted in the non-periodic phase window, the non-periodic data are sporadic and do not need to be triggered by TM messages, and when the transmission of the real-time non-periodic data is finished and the time of the non-periodic phase window is sufficient, the non-real-time data are transmitted in a first-in first-out mode. The transmission mode of the non-periodic data adopts a polling method. In the transmission process of the periodic data, the slave node reports the non-periodic information sent in the current basic period to the master node by using a reverse transmission link. And the master node performs unified scheduling, and transmits the non-periodic data after the periodic phase is finished and the non-periodic phase arrives. In this case, the master node may schedule the transmission timing sufficiently, but the synchronization delay increases the response time and is inefficient.
In summary, the ethernet-based train communication network multi-hop scheduling model is shown in fig. 5.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims (9)

1. A train communication network multi-hop scheduling method based on Ethernet is characterized by comprising the following steps:
firstly, reserving a management window at the beginning of each basic period for subsequent data scheduling; in a management window, each slave node informs a master node of data to be sent in a mode of 'multi-port to one-port', after the master node receives the data from the slave nodes, the master node broadcasts and sends a trigger message TM to each slave node in a mode of 'one-port to multi-port', wherein the trigger message TM comprises scheduling information and clock synchronization information of a current basic period, and the master node can calculate a non-cross transmission path of each message by providing the master node with information of each slave node related to data exchange, including an address, a port and a data type;
secondly, performing priority queuing on the global data and the local data, performing schedulability analysis on messages in a priority queue according to a periodic data scheduling algorithm, and determining all messages of the current EC which can be scheduled;
thirdly, constructing a train communication network topology;
fourthly, scheduling the periodic information between the master node and the slave nodes based on the constructed train communication network topology according to the schedulability analysis result of the second step;
in the second step, the principle of the schedulability analysis is as follows: the method comprises the steps that messages which can be scheduled and transmitted by a current EC are coded into a TM, messages which cannot be scheduled are buffered into a next EC to be processed, in a management window at the beginning of each EC, a main node sends the TM in a subnet in a multicast mode, period information to be transmitted is activated, a message queue to be forwarded is updated, and after all messages which can be scheduled by the current EC are determined, the main node constructs the TM and broadcasts the TM to each slave node.
2. The method according to claim 1, characterized in that the method further comprises a fifth step of scheduling aperiodic data between the master node and the slave node based on the constructed train communication network topology: the method comprises the steps that real-time non-periodic data and non-real-time data are transmitted in a non-periodic phase window, the non-periodic data are sporadic and do not need to be triggered by TM information, when the real-time non-periodic data are transmitted, and the time of the non-periodic phase window is larger than a preset value, the non-real-time data are transmitted in a first-in first-out mode, a polling method is adopted in the non-periodic data transmission mode, in the periodic data transmission process, a slave node reports non-periodic information sent in the current basic period to a master node through a reverse transmission link, the master node performs unified scheduling, and the non-periodic data are transmitted after the periodic phase is finished and the non-periodic phase arrives.
3. The method according to claim 1, characterized in that the third step is in particular: in each section of vehicle of the train communication network, each electronic control unit device is connected to a networking switch CS according to a star topology structure, and the networking switch CS divides the train communication network into a plurality of independent conflict domains; wherein the vehicle control units VCUs of the head and the tail of the two vehicles are simultaneously connected to the switch ETBN of the CS and the train communication network to be used as central control units CCU1 and CCU2, the CS and the ETBN both adopt industrial Ethernet switches, and the equipment in the vehicles comprises: a vehicle control unit VCU and a brake control unit BCU, wherein the vehicle control units VCU in the head and tail two trains TC1 and TC2 simultaneously serve as a central control unit CCU of the whole train;
CS and terminal equipment directly connected with the CS in each carriage form a sub-network, ETBN1 and CCU1 form a sub-network, each switch is provided with a main node for scheduling, ETBN1 is defined as a root switch of the train communication network, CCU1 is a root main node, a brake control unit BCU in each carriage is selected as the main node for scheduling the sub-network, CCUs of the first and the last two trains are respectively used as the main nodes of ETBN1 and ETBN2, the train communication network is divided into different hierarchical structures, and ETBN1 and CCU1 form a first layer of a multi-hop network; let switch ETBN2 and switch CS5 and the sub-networks formed by them be the second layer of the multi-hop network, and at the same time ETBN2 and CS5 serve as the child nodes of ETBN1, and similarly, let CS6 serve as the child nodes of ETBN2, CS2 and CS1 serve as the child nodes of CS5, and CS3 and CS4 serve as the child nodes of CS 6.
4. The method of claim 3, wherein in the fourth step, when performing scheduling, clock synchronization is maintained between the sub-networks, the clock synchronization between the sub-networks is achieved through clock synchronization between the master node of each sub-network and the root master node CCU1, during the management window of each basic period, the root master node CCU1 broadcasts global trigger information GTM, according to the hierarchical structure established in the third step, each master node is responsible for forwarding GTM to the sub-master nodes, each master node transmits TM after receiving GTM to initialize the basic period of each sub-network, the slave nodes transmit scheduled data according to a specified time after receiving TM message, the data encoded into TM is triggered in the corresponding basic period, and each slave node and master node in the same sub-network maintain clock synchronization, which is accomplished by TM message.
5. The method of claim 4, wherein during the scheduling in the fourth step, a fixed cutoff time is set for each master node according to the relative position of each master node BCU and ETBN2 and the root master node ETBN1, if the master node does not receive GTM before the cutoff time comes, a new GTM is generated by the master node and sent to the sub-master nodes, the clock synchronization of the whole multi-hop network is continuously completed, because each networking switch divides each car into an independent conflict domain, TM is only sent in the sub-network, and the sub-networks do not interfere with each other; unlike other master nodes, the root master node CCU1 sends the GTM directly at the very beginning of each fundamental cycle, does not determine whether the deadline has been reached, and does not need to receive the GTM.
6. The method of claim 1, wherein in the scheduling in the fourth step, all the subnets that pass through during the transmission of the global data are continuously scheduled in one basic period.
7. The method as claimed in claim 5, wherein in the scheduling in the fourth step, the real-time period information transmitted in the train communication network is mathematically described, assuming that the number of period information to be transmitted in the network is N, for the ith period information miThe transmission task is described as
γi＝{mi(Ci,Di,Ti,OiPKi,Si,DSi,Pi,Li,ni),i＝1…N} (1)
Wherein, CiRepresents period information miThe transmission time of (c); diAnd TiRespectively represents miOff-time and period of, OiIndicating the time at which the initial phase of the data is scheduled, PKiRepresents miThe size of the data packet; siAnd DSiRespectively represents miA source node and a destination node; piIndicating a priority; l isiRepresents miSet of links through, niRepresents miNumber of link segments passed, Li＝{lk|k＝1…n}。
8. The method of claim 5, wherein when scheduling in the fourth step, real-time periodic data must be transmitted within a periodic phase window, and considering the delay e of the switch, the working time of all links cannot exceed the periodic phase LSW-e, where LSW is the size of the periodic phase window, and for the sending link, it must be ensured that the sum of the transmission times of the data must not exceed the size LSW-e of the periodic phase window; for the receiving link, it is ensured that the time f when the last data transmission is completed is smaller than the size LSW of the periodic phase window.
9. The method of claim 5, wherein the scheduling in the fourth step sets real-time period data miThe time of arrival at the output port of the switch is AiIn queue ratio miData m with high priorityjThe reception completion time on the reception link RL is Fj(ii) a Wherein A isi＝Oi+Ci，OiIs the initial phase of the data, namely the transmission starting time of the data on the transmission link; to ensure reception of high priority data, miAt the beginning of reception on the reception link RL, max { F }j,AiH, so miAt the moment when the transmission of the receiving link is completed, max { F }j,Ai}+CiSo m isiCan be scheduled in the current basic cycle and should also satisfy
max{Fj,Ai}+Ci≤LSW (3)
When the formulas (2) and (3) cannot be satisfied simultaneously, miWill be postponed to the next basic cycle scheduling when the scheduled high priority data arrives later than miWhen m is allowediPrioritise these scheduled high priority data receptions on the reception link RL;
after receiving the TM message, each slave node decodes the TM message, prepares periodic data required to be transmitted in the current basic period according to the information contained in the TM message, initializes the data to be transmitted when the period comes, and simultaneously receives the periodic data transmitted by other nodes.
CN201811189486.2A 2018-10-12 2018-10-12 Train communication network multi-hop scheduling method based on Ethernet Active CN109219019B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811189486.2A CN109219019B (en) 2018-10-12 2018-10-12 Train communication network multi-hop scheduling method based on Ethernet

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811189486.2A CN109219019B (en) 2018-10-12 2018-10-12 Train communication network multi-hop scheduling method based on Ethernet

Publications (2)

Publication Number Publication Date
CN109219019A CN109219019A (en) 2019-01-15
CN109219019B true CN109219019B (en) 2021-02-09

Family

ID=64979570

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811189486.2A Active CN109219019B (en) 2018-10-12 2018-10-12 Train communication network multi-hop scheduling method based on Ethernet

Country Status (1)

Country Link
CN (1) CN109219019B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110271584A (en) * 2019-06-28 2019-09-24 四川迈铁龙科技有限公司 System is dispatched in the cooperative system of rail traffic scheduling and rail traffic
CN113541991B (en) * 2020-04-20 2022-09-27 中车唐山机车车辆有限公司 Train network control system and communication control method based on time sensitive network
CN111674428B (en) * 2020-06-05 2022-05-03 中车青岛四方车辆研究所有限公司 Train brake control unit Ethernet joint debugging system and joint debugging method
CN112492046B (en) * 2020-12-11 2024-07-02 交控科技股份有限公司 Train distributed TCMS (train control system) main-standby redundancy management method and system
CN113839988B (en) * 2021-08-24 2024-02-20 四川发展磁浮科技有限公司 Train multi-network integration network control system and control method
CN114363215B (en) * 2021-12-27 2024-05-28 北京特种机械研究所 Train communication network time delay analysis method based on supply and demand balance

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102371955A (en) * 2011-10-27 2012-03-14 苏州海格新能源汽车电控系统科技有限公司 Method for realizing time-sharing task of whole-vehicle controller
CN102970247A (en) * 2012-12-13 2013-03-13 中国航空无线电电子研究所 Effective communication time scheduling method of time-triggered network
WO2014009532A2 (en) * 2012-07-12 2014-01-16 Nec Europe Ltd. Method and system for supporting channel access of stations in a wireless communication network
CN106453140A (en) * 2016-09-29 2017-02-22 北京汽车股份有限公司 Message processing method and device based on electronic control units
CN107277855A (en) * 2017-04-20 2017-10-20 西安电子科技大学 A kind of car networking media access method for the class business that ensures safety
CN108173614A (en) * 2017-12-08 2018-06-15 同济大学 A kind of time synchronization and dispatching method of vehicle-mounted Ethernet
CN108631899A (en) * 2018-05-16 2018-10-09 常州海德克智能科技有限公司 It is forwarded and fault-tolerance approach and device based on the data packet in time triggered Ethernet

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070083491A1 (en) * 2004-05-27 2007-04-12 Silverbrook Research Pty Ltd Storage of key in non-volatile memory
US20160019501A1 (en) * 2014-07-17 2016-01-21 Dmitri Olechko Systems, methods and computer-program products for automation of dispatch of shipment delivery order

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102371955A (en) * 2011-10-27 2012-03-14 苏州海格新能源汽车电控系统科技有限公司 Method for realizing time-sharing task of whole-vehicle controller
WO2014009532A2 (en) * 2012-07-12 2014-01-16 Nec Europe Ltd. Method and system for supporting channel access of stations in a wireless communication network
CN102970247A (en) * 2012-12-13 2013-03-13 中国航空无线电电子研究所 Effective communication time scheduling method of time-triggered network
CN106453140A (en) * 2016-09-29 2017-02-22 北京汽车股份有限公司 Message processing method and device based on electronic control units
CN107277855A (en) * 2017-04-20 2017-10-20 西安电子科技大学 A kind of car networking media access method for the class business that ensures safety
CN108173614A (en) * 2017-12-08 2018-06-15 同济大学 A kind of time synchronization and dispatching method of vehicle-mounted Ethernet
CN108631899A (en) * 2018-05-16 2018-10-09 常州海德克智能科技有限公司 It is forwarded and fault-tolerance approach and device based on the data packet in time triggered Ethernet

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于交换式以太网的列车通信网络建模与性能分析;张玉琢;《通信学报》;20151201;全文 *

Also Published As

Publication number Publication date
CN109219019A (en) 2019-01-15

Similar Documents

Publication Publication Date Title
CN109219019B (en) Train communication network multi-hop scheduling method based on Ethernet
US7519076B2 (en) Method and node for using a communication network in parallel for real-time applications and non-real-time applications
US10530606B2 (en) Method for transmitting data via a serial communication bus, bus interface, and computer program
US11343117B2 (en) Method for transmitting data via a serial communication bus, correspondingly designed bus interface, and correspondingly designed computer program
CN114830611B (en) Centralized network configuration entity and time-sensitive network control system comprising same
CN110870285B (en) Method for high-performance data transmission in data networks with partial real-time requirements and device for carrying out said method
CN102801597B (en) Ethernet-based train communication network implementation method
CN110545241B (en) Message processing method and device
Lim et al. Performance comparison of ieee 802.1 q and ieee 802.1 avb in an ethernet-based in-vehicle network
CN109639560B (en) Method for improving availability of real-time computer network
Lee et al. Meeting the real-time constraints with standard Ethernet in an in-vehicle network
US11323380B2 (en) Method of synchronization of data packet transmission
CN103929377A (en) Wired network and wireless network combined dispatching method and system and related devices
US10389806B2 (en) Method for asynchronous data communication in a real-time capable ethernet data network
TW202046690A (en) Method implemented by computer means of a communicating entity in a packet-switched network, and computer program and computer-readable non-transient recording medium thereof, and communicating entity of a packet-switched network
JP6152425B2 (en) Real-time message transmission method and computer network for transmitting real-time messages
CN102164066A (en) Reliable data transmission method for CAN (control area network)-based measurement and control subsystem
CN115174370B (en) Distributed mixed data deterministic transmission device and method
Min et al. Performance Enhancement of In-Vehicle 10BASE-T1S Ethernet Using Node Prioritization and Packet Segmentation
Min et al. Interrupt-enabled Physical Layer Collision Avoidance for 10BASE-T1S Automotive Ethernet
Guo et al. A flow-grained end-to-end delay analysis for RC traffic in TTEthernet
KR101514398B1 (en) Gateway in-vehicle based on virtual bus concepet and message transmission method thereof
Shi et al. Research on bandwidth allocation strategy of train communication network based on time division multiplexing
Bayilmis et al. A CAN/IEEE 802.11 b wireless Lan local bridge design
JP2004048491A (en) Pon (passive optical network) system

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