CN110808920A

CN110808920A - Satellite communication flow control method for coexistence of CCSDS (consultative committee for space data system) frame and IP (Internet protocol) message

Info

Publication number: CN110808920A
Application number: CN201911035659.XA
Authority: CN
Inventors: 陈宇; 卞春江; 陈实; 周海; 李辉; 刘一腾
Original assignee: National Space Science Center of CAS
Current assignee: National Space Science Center of CAS
Priority date: 2019-10-29
Filing date: 2019-10-29
Publication date: 2020-02-18
Anticipated expiration: 2039-10-29
Also published as: CN110808920B

Abstract

The invention discloses a satellite communication flow control method for coexistence of a CCSDS frame and an IP message, which comprises the following steps: respectively identifying and marking the type of the received IP message and the CCSDS message; the messages of the same type are loaded into a CCSDS data frame; dynamically allocating bandwidth for various CCSDS data frames by utilizing token overall allocation; putting the CCSDS frame into a corresponding virtual channel; the virtual channels are scheduled based on dynamic scheduling of frame urgency. The method identifies and marks the types of the IP message and the CCSDS message according to the information of a plurality of specific domains of the message, performs overall bandwidth allocation on the CCSDS data frame based on the frame type, preferentially schedules a channel with high time delay requirement and high channel weight, and improves the service quality of satellite communication with the coexistence of the CCSDS protocol and the IP protocol; the method of the invention transmits according to the priority, namely transmits according to the requirement of QoS index, so that each type of service can meet the QoS index, and the service quality guarantee is provided.

Description

Satellite communication flow control method for coexistence of CCSDS (consultative committee for space data system) frame and IP (Internet protocol) message

Technical Field

The invention relates to the field of satellite communication, in particular to a satellite communication flow control method for coexistence of a CCSDS frame and an IP message.

Background

The existing satellite communication flow control method mainly comprises the following three methods:

a first-come first-serve flow control method: the satellite virtual channel is selected according to the time sequence of frame arrival, namely the virtual channel which arrives at the first frame at the current scheduling time is selected to occupy the physical channel, and the data frame is sent.

Secondly, a flow control method based on polling scheduling: the polling scheduling algorithm selects each virtual channel in turn in sequence to occupy the physical channel and send the data frame, that is, each virtual channel in the system is allocated with a fixed time slice, and the virtual channel occupies the physical channel and sends the data frame in the respective time slice.

Thirdly, the flow control method based on the static priority comprises the following steps: according to different requirements of each virtual channel on real-time performance and importance, each virtual channel is allocated with a static priority, and then the virtual channels are scheduled and selected according to the static priorities for transmission, namely, the virtual channel with the highest priority is scheduled and selected to occupy a physical channel and transmit a data frame.

In ground and space station communications, the link layer protocol uses the CCSDS data link layer protocol and the network layer protocol uses both the IP protocol and the CCSDS packetization protocol. Therefore, the information transmission scheme between the satellite and the ground must satisfy the service quality of the IP message and the CCSDS message at the same time. With respect to the communication tasks of the space station,

the existing satellite communication flow control method has two disadvantages.

1. The flow control cannot be performed on the IP packet and the CCSDS packet at the same time.

Specifically, for the spatial information transmitted by using the IP protocol format and the CCSDS protocol format, that is, the IP packet and the CCSDS packet, the prior art does not provide a specific transmission scheme that can satisfy the service quality of both the IP packet and the CCSDS packet.

2. Existing flow control methods do not provide quality of service (QoS).

The service quality refers to a series of service requirements that the network requires to meet when transmitting data streams, and may be specifically quantized to performance indexes such as bandwidth (bandwidth), delay (delay), delay jitter (delay jitter), packet loss rate (packet loss), throughput (throughput), and the like. The goal of QoS is to provide dedicated bandwidth for a specific service, reduce the transmission delay and delay jitter of a packet, reduce the packet loss rate, and provide a better and predictable network service. Services in a satellite communication scene and services of a ground network pay attention to the performance indexes, but the existing satellite communication flow control method cannot distinguish service types and cannot guarantee service quality according to specific service types.

Disclosure of Invention

The present invention aims to overcome the technical defects and provide a spatial information transmission traffic control method suitable for the coexistence of the CCSDS protocol and the IP protocol, so as to ensure the service quality of message transmission of the CCSDS protocol and the IP protocol.

In order to achieve the above object, the present invention provides a method for controlling satellite communication traffic flow in which a CCSDS frame and an IP packet coexist, the method comprising:

respectively identifying and marking the type of the received IP message and the CCSDS message;

the messages of the same type are loaded into a CCSDS data frame;

dynamically allocating bandwidth for various CCSDS data frames by utilizing token overall allocation;

putting the CCSDS frame into a corresponding virtual channel;

the virtual channels are scheduled based on dynamic scheduling of frame urgency.

As an improvement of the above method, the IP message is from a terrestrial lan or a spatial lan, and the CCSDS message is from a terrestrial lan or a spatial lan; the IP message carries data of a satellite service 7, a satellite service 6, a satellite service 5 and a satellite service 4; the CCSDS packet carries data of satellite service 3, satellite service 2, satellite service 1, and satellite service 0.

As an improvement of the above method, the performing type identification and marking on the received IP packet and CCSDS packet respectively specifically includes:

identifying the service type of data in the IP message; one or more of a source address, a destination address, a source port number, a destination port number, a protocol type and TOS information carried by the IP message;

identifying the service type of the CCSDS frame according to the frame type carried by the CCSDS message;

determining the priority for the IP message according to the service type of the data in the IP message, marking the IP message, and determining the priority for the CCSDS message according to the service type of the data in the CCSDS message;

the priority levels correspond to the service types one to one, each service type uniquely corresponds to one priority level, and each service type corresponds to a plurality of QoS description parameters including time delay, time delay jitter, packet loss rate and throughput.

As an improvement of the above method, the determining the priority is based on: according to the pre-stored corresponding relationship between the service type and the priority, the method specifically comprises the following steps:

identifying whether the message carries a priority;

if the message is carried, searching a corresponding relation between a prestored service type and the priority to find the priority corresponding to the service type, judging whether the carried priority is consistent with the found priority, and if not, modifying the priority of the message into the found priority;

if not, searching the corresponding relation between the pre-stored service type and the priority according to the identified service type to find the priority corresponding to the service type, and distributing the priority to the IP message.

As an improvement of the above method, the determining the priority is based on: according to the pre-stored corresponding relationship among the service types, the users and the priorities, the method specifically comprises the following steps:

identifying whether the message carries a priority;

if the IP message is carried, acquiring the information and the service type of the user from the message, searching the corresponding relation among the user, the service type and the priority stored in advance to find the priority corresponding to the user and the service type, judging whether the carried priority is consistent with the searched priority, and modifying the priority of the IP message into the searched priority if the carried priority is inconsistent;

if not, obtaining the user information and the service type from the message, searching the corresponding relation of the user, the service type and the priority stored in advance according to the user information and the identified service type, finding the priority corresponding to the user and the service type, and distributing the priority to the message.

As an improvement of the above method, after performing type identification and marking on the received IP packet and CCSDS packet, the method further includes: traffic shaping is used to limit the maximum traffic for each type of satellite traffic.

As an improvement of the above method, the loading the same type of messages into the CCSDS data frame specifically includes:

selecting IP messages of the same type and encapsulating the IP messages in a CCSDS frame; when the IP message is packaged, setting a virtual channel identification field in the AOS frame according to the priority of the IP message, thereby finishing the correspondence between the priority of the IP message and the virtual channel;

the CCSDS frames with the same priority are multiplexed and sectionally loaded into a packet area of a multiplexing protocol data unit with fixed length, and then an MPDU main head, a frame main head and a frame tail are respectively added, thereby generating a frame.

As an improvement of the above method, the overall allocation by using the token dynamically allocates bandwidth for each type of CCSDS data frame; the method specifically comprises the following steps:

configuring a token generator and a token bucket for each satellite service; setting an initial bandwidth for each service, and generating tokens at a fixed rate by each token producer according to the initial bandwidth; the satellite service 7 and the satellite service 6 are quota allocation, the satellite service 5, the satellite service 4 and the satellite service 3 can be lent for the quota allocation, and the satellite service 2, the satellite service 1 and the satellite service 0 can be expanded for balance allocation;

putting tokens generated by each token generator into a certain token bucket according to a token overall allocation algorithm;

when transmitting satellite service data, a token needs to be acquired from a token bucket of the satellite service to transmit a CCSDS message, and the number of tokens in the token bucket is reduced by 1; if the token bucket has no token, the CCSDS message cannot be transmitted.

As an improvement of the above method, the step of putting tokens generated by each token generator into a token bucket according to a token pool allocation algorithm specifically includes:

step S1), judging whether the satellite service token bucket is full, if so, entering step S2), otherwise, putting the token into the service token bucket, and ending;

step S2), judging whether the token is the one of quota distribution service, if yes, discarding the token, and ending; otherwise, go to step S3);

step S3), judging the weight of all balance expandable service token receivers to be 0, if so, resetting the weight of all balance expandable service token receivers; and point to the first balance to distribute extensible services; otherwise, go to step S4);

step S4), traversing all balance expandable service token buckets, finding out the token buckets with unsatisfied token buckets and non-zero weight, placing tokens and ending; if not, the token is discarded, and the process is ended.

As an improvement of the foregoing method, before placing the CCSDS message into the corresponding virtual channel, the method further includes: when bandwidth is allocated to data of which the service type is satellite service 3, satellite service 2, satellite service 1 or satellite service 0, scheduling the IP data packet specifically includes:

sending various CCSDS messages into corresponding IP queues; the IP queues adopt weighted polling queue scheduling, namely, the scheduling is carried out between each queue in turn, and the message in each queue can be ensured to be forwarded to a certain extent.

As an improvement of the above method, the CCSDS message is put into a corresponding virtual channel; the method specifically comprises the following steps:

acquiring the priority of the message from the CCSDS message, acquiring a corresponding virtual channel according to the corresponding relation between the pre-stored message priority and the virtual channel, and transmitting the acquired message through the virtual channel.

As an improvement of the foregoing method, the dynamic scheduling based on the frame urgency schedules a virtual channel, which specifically includes:

calculating the urgency F of the frame as:

f ═ Tmax-Tw × P where Tmax is the maximum scheduling delay allowed by the virtual channel, Tw is the time that a frame in the virtual channel has waited, the time difference between the current scheduling time k △ t of the virtual channel and the arrival time Ta of the current first frame in the virtual channel is defined as the time that a frame in the virtual channel has waited, that is:

tw-k Δ t-TaP is a static priority, the transmission real-time performance of each virtual channel before starting transmission is called as a static priority, a static priority with a value of 1 can be given to the virtual channel with high real-time performance requirement, otherwise, a static priority with a number of 10 is given; the maximum value of the static priority is 10, and the minimum value is 1;

firstly, scheduling a virtual channel with the minimum value F of the urgency degree of a transmission frame; if two or more virtual channels have the value F of the urgency degree of the minimum frame at the same time, the static priority of the virtual channels needs to be considered, the virtual channel with the minimum static priority value is scheduled to be transmitted preferentially, and if more than one channel with the minimum static priority occurs at the same time, the channel with the small sequence number is selected and served preferentially.

The invention has the advantages that:

1. according to the method, the types of the IP message and the CCSDS message are identified and marked according to the information of a plurality of specific domains of the message, the CCSDS data frame is subjected to overall bandwidth allocation based on the frame type, channels with high time delay requirements and high channel weights are scheduled preferentially, and the service quality of satellite communication with the coexistence of the CCSDS protocol and the IP protocol is improved;

2. the method of the invention transmits the spatial information transmitted by adopting the IP format and the CCSDS format according to the priority, namely according to the requirement of the QoS index, so that each type of service can meet the QoS index and the service quality guarantee is provided;

3. the method of the invention meets the QoS requirements of different services in satellite communication and also meets the requirement of dynamic bandwidth allocation of a satellite channel.

Drawings

FIG. 1 is a satellite communications protocol architecture;

FIG. 2 is a schematic diagram of a flow control method of the present invention;

FIG. 3 is a schematic diagram of the traffic shaping process of the present invention;

FIG. 4 is a schematic diagram of MPDU encapsulation for a fixed time slot-with padding packets;

FIG. 5 is a schematic view of MPDU encapsulation for a fixed time slot-just filled;

FIG. 6 is a schematic diagram of MPDU encapsulation for a fixed time slot-with a packet remaining;

FIG. 7 is a schematic diagram of IP Over CCSDS encapsulation;

FIG. 8 is a process for encapsulating CCSDS frames;

FIG. 9 is a flow diagram of dynamic bandwidth allocation based on token pooling allocation;

FIG. 10 is a flow chart of overall token distribution;

FIG. 11 is a flow chart of IP packet scheduling;

FIG. 12 is a flow chart of a virtual channel scheduling algorithm based on frame urgency.

Detailed Description

The technical solution of the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

Abbreviations and key term definitions are introduced first:

bandwidth: refers to the amount of data that the satellite link can transmit per unit of time.

Time delay: refers to the average elapsed time of network data from the entrance to the exit of the network. The factors that contribute to the delay include packet delay, queuing delay, switching delay, and propagation delay.

Time delay jitter: meaning that different packets in the same traffic stream exhibit different delays.

Packet loss rate: refers to the ratio of lost messages to the total number of messages sent.

Throughput: refers to the ability to efficiently transfer data between network users.

The invention provides a satellite communication flow control method with coexisting CCSDS frames and IP messages, which meets the requirements of different service QoS in satellite communication and the requirement of dynamic bandwidth allocation of satellite channels.

Compared with terrestrial communication, satellite communication transmission has the following three characteristics:

differentiation of real-time requirements of services, for example: the satellite remote control service is the most important life-saving data, and has the highest real-time requirement in relation to the normal flight and operation of the satellite; the telemetering service is satellite operation information transmitted from a satellite to a ground command center, and the requirement on real-time performance is high; voice and image services have general real-time requirements. The local area network service has no requirement on real-time performance.

The priority of the service is divided into a plurality of grades, the spatial information is strictly graded according to the urgency, and the high urgency information is guaranteed to be transmitted preferentially under the condition that the satellite communication resources are in shortage.

The spatial resources for traffic transmission are limited: satellite platforms have limited processing power and satellite communication links have limited bandwidth. Aiming at the three characteristics, the design target of the flow control method is as follows:

the real-time requirement of the time-sensitive service is met; the priority forwarding requirement of the high-priority service is met; the bandwidth utilization rate is improved, and the management and guarantee capability of the bandwidth is provided.

The satellite communication protocol architecture is shown in figure 1. The satellite communication protocol system adopts the CCSDS protocol and the IP protocol. In the protocol system, an application layer comprises the transmission of 8 types of satellite services, and the priority of the satellite services 0 to 7 is sequentially increased. At the network layer, the CCSDS space packet carries class 4 traffic, and the IP and CCSDS encapsulation packets carry class 4 traffic. In a data link layer, flow control is realized based on a CCSDS AOS protocol, and the method specifically comprises the following steps: frame encapsulation based on traffic priority, bandwidth allocation and virtual channel scheduling. According to the service application mode 4 in table 1, each service corresponds to one virtual channel, as shown in table 1.

Table 1: traffic types in satellite communications

First, data is classified into 8 types by classifying the data into satellite service types as shown in table 1. Distributing corresponding virtual channels for each type of satellite service; then, performing global overall dynamic bandwidth allocation on each virtual channel; and finally, scheduling the packaged message according to the priority of each service by adopting a virtual channel dynamic scheduling algorithm based on the urgency degree, and ensuring that important and low-delay data is sent preferentially. Therefore, the QoS requirements of different services in the air-space information transmission are met, and the dynamic bandwidth allocation requirement of the air-space information transmission is also met.

As shown in fig. 2, the method of the present invention comprises: traffic classification and marking, traffic shaping, packet channel multiplexing, bandwidth allocation, IP data packet scheduling, and virtual channel scheduling.

Step 1) service classification and marking:

and receiving the IP message from the ground local area network or the space local area network, and receiving the CCSDS message from the ground CCSDS equipment or the space CCSDS equipment.

The IP packet received by the flow control device carries data of the satellite service 7, the satellite service 6, the satellite service 5, and the satellite service 4. The received CCSDS message carries data of satellite service 3, satellite service 2, satellite service 1, and satellite service 0.

And identifying the service types of the data in the IP message and the CCSDS message.

And identifying the service type of the data in the IP message according to one or more of the information of the source address, the destination address, the source port number, the destination port number, the protocol type, the TOS and the like carried by the IP message.

And identifying the service type of the CCSDS frame according to the frame type carried by the CCSDS message.

And determining the priority for the IP message according to the service type of the data in the IP message, marking the IP message, and determining the priority for the CCSDS message according to the service type of the data in the CCSDS message.

The method for determining the priority of the message (IP message and CCSDS message) comprises two methods:

table 2: correspondence between service type and priority

The first method comprises the following steps: according to the pre-stored corresponding relationship between the service types and the priorities, as shown in Table 2

1. And identifying whether the message carries the priority, if so, searching the corresponding relation between the prestored service type and the priority to find the priority corresponding to the service type, judging whether the carried priority is consistent with the searched priority, and if not, modifying the priority of the message into the searched priority.

2. And identifying whether the message carries the priority, if not, searching a corresponding relation between the pre-stored service type and the priority according to the identified service type to find the priority corresponding to the service type, and distributing the priority to the IP message.

And the second method comprises the following steps: according to the pre-stored corresponding relation among the service types, the users and the priority

1. Identifying whether the message carries the priority or not, if so, acquiring the information and the service type of the user from the message, searching the corresponding relation among the prestored user, the service type and the priority to find the priority corresponding to the user and the service type, judging whether the carried priority is consistent with the found priority or not, and if not, modifying the priority of the IP message into the found priority.

2. Identifying whether the message carries the priority or not, if not, acquiring the information and the service type of the user from the message, searching the corresponding relation of the user, the service type and the priority which is stored in advance according to the information and the identified service type of the user, finding the priority corresponding to the user and the service type, and distributing the priority to the message.

Setting of priority: the priority levels correspond to the service types one by one, and each service type uniquely corresponds to one priority level. Each service type corresponds to a plurality of QoS description parameters, such as time delay, time delay jitter, packet loss rate and throughput.

Step 2) traffic shaping

Traffic shaping employs traffic shaping to limit the maximum traffic for each type of satellite traffic.

When the message arrives, the message is firstly classified according to a preset matching rule, and then the message enters a buffer area and a token bucket for processing. The token bucket places tokens into the bucket at a speed set by a user, if enough tokens in the token bucket can be used for sending messages, the messages are directly sent out continuously, and meanwhile, the quantity of the tokens in the token bucket is correspondingly reduced along with the sending of the messages. When the token in the token bucket is as few as the message can not be sent, the message is cached into the buffer (the subsequent message arriving at the buffer detects that the buffer has the message and directly enters the buffer queue, if the queue length reaches the upper limit, the message is directly discarded). When there is message in the buffer, the buffer takes out the message from the buffer queue according to a certain period to send, and the number of tokens in the token bucket is compared with the number of tokens in the token bucket for each sending until the number of tokens in the token bucket is reduced to the point that the message in the buffer queue can not be sent or the message in the buffer queue is completely sent. As shown in fig. 3.

Step 3) packet channel multiplexing

Frame generation refers to a mechanism for providing a shared virtual channel (frame) for various types of data packets generated by higher layer protocols.

An isochronous frame generation algorithm is employed. The isochronous frame generation algorithm is to encapsulate packets arriving at an upper layer into a frame at regular intervals and release the frame. During source packet encapsulation, three situations may occur:

the first method comprises the following steps: within the fixed framing waiting time Sw, if the number N of arriving packets is less than N, the remaining MPDU packet area needs to be filled with idle data packets, as shown in fig. 4;

and the second method comprises the following steps: within a fixed framing latency Sw, the number of arriving packets, N, is exactly equal to N, as shown in fig. 5;

and the third is that: within the fixed framing waiting time Sw, the number N of arriving packets is greater than N, at this time, only the first N CCSDS packets are encapsulated into one MPDU and transmitted, and the remaining packets are left in the next MPDU for transmission, as shown in fig. 6.

Packet channel multiplexing of IP messages

And selecting the IP messages of the same type and encapsulating the IP messages in a CCSDS frame. When the IP message is packaged, a VCID (virtual channel identification field) identification field in the AOS frame is set according to the priority of the IP message, so that the correspondence between the priority of the IP message and the virtual channel is completed. As shown in fig. 7.

Packet channel multiplexing of CCSDS frames: the CCSDS frames with the same priority are multiplexed and sectionally loaded into a packet area of a Multiplexing Protocol Data Unit (MPDU) with a fixed length, and then a MPDU main head, a frame main head and a frame tail are respectively added, thereby generating a frame. The data structure varies as shown in fig. 8, where the length of MPDUs is fixed within a specific task and fits exactly within the data area of a fixed-length Virtual Channel Data Unit (VCDU).

Step 4) Bandwidth Allocation

Since various types of satellite services in satellite communication have different priorities. The invention provides a dynamic bandwidth allocation method based on token overall allocation, and a schematic diagram is shown in fig. 9. The basic composition structure is as follows: the token generator: tokens are generated at a constant rate. The token overall distribution module: and carrying out overall distribution on the tokens in each token bucket. Token bucket: a container for tokens.

And configuring a token bucket shaper for each type of traffic flow to shape the traffic flow. The output flow of each service flow is controlled by a corresponding token bucket, and the rate and the bandwidth occupation ratio of the output flow are determined by the parameter setting of the token bucket. The token generator and the token bucket of the token bucket are divided, the token generator generates tokens at a fixed speed, the tokens are collected and are subjected to overall scheduling, and all generated tokens are put into the corresponding token bucket under the condition that the token bucket is not full, so that the emergency requirement of the data possibly generated is ensured. After the token bucket for the quota allocation service is full, the newly generated tokens are discarded. The full allocation may be used to transmit data instead of discarding newly generated tokens after the token bucket for loaning allocation traffic or balance allocation scalable traffic is full. When balance allocation can expand data to be sent in the service token bucket and cannot be sent due to insufficient tokens, newly generated redundant tokens can be called to the token bucket with insufficient tokens. Therefore, the link bandwidth sharing among multiple service types is realized, and the utilization rate of the network bandwidth is improved.

Firstly, bandwidth is distributed for each type of satellite service, and then bandwidth distribution is carried out on all services based on a token overall distribution mechanism.

Table 3: bandwidth allocation mode of satellite service

The bandwidth allocation mode is divided into 3 types: quota allocation, full allocation can be loaned, and balance allocation can be expanded.

Quota allocation: fixed bandwidth resources are allocated, the bandwidth resources of the service are not occupied by other services, and the bandwidth resources of other services are not used. For example, the satellite service 7 and the satellite service 6 are allocated in a quota manner.

The full allocation can be loaned: enough bandwidth is allocated initially, and when the traffic of the service is smaller than the initial bandwidth, the surplus bandwidth of the service can be lent to the balance allocation extensible service. When the service flow of the type increases, the loaned bandwidth must be returned to the service bandwidth of the type. The full allocation can lend the class service without borrowing other service bandwidth. For example, satellite services 5, 4, and 3 are loaned with a full allocation.

Balance allocation is extensible: the initially allocated bandwidth is the remaining bandwidth resource of the flow control device. When the service flow is smaller than the initial bandwidth, the surplus bandwidth can be lent to other balance distribution expandable services. When the service flow is increased, other balance distribution expandable services must return the initial bandwidth of the service, and can borrow the bandwidth from other services, and after the maximum bandwidth of each user is reached, other service bandwidths are not borrowed. For example, satellite service 2, satellite service 1, and satellite service 0 employ a balance allocation scalable approach.

The method specifically comprises the following steps:

step 4-1) configuration of token Generator and token bucket

Each service is provided with a token generator and a token bucket. The tokens generated by each token generator are put into a token bucket according to a coordinated allocation algorithm. When certain service data is transmitted, a token needs to be acquired from the service token bucket, a CCSDS message can be transmitted, and the number of tokens in the token bucket is reduced by 1. If the service token bucket has no token, the CCSDS message can not be transmitted.

Step 4-2) initial Bandwidth

An initial bandwidth is set for each service, and each token producer produces a token according to the initial bandwidth. The tokens of each service are preferably placed in the token bucket of the service. Thus, when there is a service flow, all services can obtain the initial bandwidth; when the token bucket is full, i.e. no traffic flows, surplus tokens are allocated to other token buckets, i.e. bandwidth is allocated to other traffic.

Step 4-3) token overall distribution

The overall distribution mode of the token is divided into 3 types:

1. fixed bandwidth

Tokens are only put into the corresponding token bucket, and if the token bucket is full, the tokens are discarded.

2. Bandwidth lending

The full allocation loanable service may loan the rich bandwidth to a balance allocation scalable service, and the balance allocation scalable service may loan the rich bandwidth to other balance allocation scalable services. Bandwidth lending is based on weighted priority rules. And presetting a weight for each service type, and scheduling surplus tokens according to the weight of each service when the tokens are distributed comprehensively. And ensuring that each service can obtain rich bandwidth allocation in proportion.

3. Bandwidth return

After the bandwidth is borrowed, when the traffic of the service is increased, the borrowed bandwidth returns to the service.

As shown in fig. 10, the token pool allocation algorithm is described as follows:

The overall token distribution strategy has the following characteristics:

1. the quota allocation mode ensures high-level service bandwidth: important services have high requirements on packet loss rate and the services occupy less bandwidth resources, so that a fixed bandwidth can be allocated to the important services in a bandwidth quota allocation mode, and the packet loss rate of the services is guaranteed to be the minimum value.

2. The bandwidth of the medium-grade service is ensured to be effectively utilized in a sufficient allocation loanable mode: and setting an initial bandwidth for each service type, wherein the initial bandwidth can ensure the normal operation of the corresponding service, and the initial bandwidth is configured according to the maximum value of the application scene. When the flow of a certain service is low, the bandwidth of the service can be lent to the balance extensible service. When the traffic flow increases to cause the remaining bandwidth of the traffic flow to be insufficient and the initial bandwidth of the traffic flow is occupied by other traffic, the initial bandwidth of the traffic flow must be returned. This mechanism ensures efficient use of the medium grade traffic bandwidth.

3. The balance allocation extensible mode ensures the minimum bandwidth of the low-level service: by the overall distribution strategy of the weighted priority token, surplus bandwidth is distributed to the service with insufficient bandwidth according to the weight proportion, so that the condition of starvation of the service does not exist, and the TCP global synchronization phenomenon during service recovery is avoided.

From the three characteristics, the dynamic bandwidth allocation strategy based on the token overall allocation provided by the invention firstly meets the initial bandwidth of normal operation of all services, and has the dynamic allocation capability of dynamically occupying unused bandwidth among the services, thereby improving the utilization rate of satellite communication bandwidth. Meanwhile, the dynamic allocation mechanism is based on the allocation of a weighted priority rule, and the stable transition of network flow is ensured.

Step 5) IP data packet scheduling

When bandwidth is allocated to data of traffic types VC3, VC2, VC1, and VC0, since these traffic are multi-user IP traffic, IP packet scheduling is required before entering a virtual channel. The scheduling method is shown in the figure. As shown in fig. 11, the specific steps are as follows:

and step 5-1) classifying the messages according to the types of the CCSDS messages. The satellite services 3-0 are all IP services, one or more IP packets are encapsulated in the CCSDS frame, and the priorities of the IP packets are consistent.

And 5-2) sending various CCSDS messages into corresponding IP queues. The IP queues adopt weighted polling queue scheduling, namely, the scheduling is carried out between each queue in turn, and the messages in each queue can be ensured to be forwarded to a certain extent.

As shown in the figure, the IP message output port has N queues, the transmission bandwidth of the output port is M, the priority is increased from 1 to N in sequence,firstly, setting corresponding weight value for each queue, and sequentially setting weight value as W₀-W_NThe weight value represents the proportion of the obtained bandwidth, and the obtained bandwidth of the queue x is M × W_xThe sum of weighted values. For example, the transmission bandwidth of one port is 1000M, and the weighted values set for 5 queues at the egress port are 4, 3, 1, and 1, respectively, then the forwarding bandwidths obtained by the corresponding queues are 400M, 300M, 100M, and 100M, respectively.

The WRR queues have the advantage that although the scheduling of each queue is in turn, the time for allocating service is not fixed, and when there is no message in a queue of a certain priority, the message in the queue of the next priority is continuously scheduled, so that the bandwidth resource can be fully utilized. The method is suitable for intermittent transmission of IP services.

And 5-3) putting the CCSDS message into a corresponding virtual channel.

Step 6) virtual channel scheduling

Since the service types transmitted by different virtual channels are different, and the service types corresponding to the priorities correspond to the virtual channels one to one, the virtual channels themselves have different service system levels. Therefore, the scheduling of the virtual channel also becomes one of the measures for providing quality of service guarantee for different application data.

When virtual channel scheduling is carried out, the waiting time of frames in virtual channels and the maximum time delay allowed by the frames in each virtual channel are considered, so that the real transmission urgency of the frames can be fully reflected, and a virtual channel dynamic scheduling algorithm based on the urgency of CCSDS frames is provided.

The dynamic scheduling algorithm based on the urgency of the frame involves 3 factors: maximum scheduling delay allowed by the virtual channel, the time that a frame has been waiting in the virtual channel, and static priority.

Maximum scheduling delay allowed by the virtual channel Tmax: each virtual channel has an allowed maximum scheduling delay, and the requirements of different signal sources on the maximum delay are different. When the frames in the virtual channel buffer wait longer than the maximum scheduling delay allowed by the virtual channel buffer, the frames will be lost.

The time Tw that the frame in the virtual channel has waited, the time difference between the current scheduling time k △ t of the virtual channel and the arrival time Ta of the current first frame in the virtual channel, is defined as the time that the frame in the virtual channel has waited, namely:

Tw＝kΔt–Ta (1)

static priority P: the real-time transmission of each virtual channel before transmission is called as static priority, the virtual channel with high real-time requirement can be endowed with a static priority with a smaller value, and conversely, the virtual channel with a larger value can be endowed with a static priority, so that different requirements of different virtual channels on transmission delay are reflected.

The urgency F of a frame may be defined as follows:

F＝(Tmax-Tw)×P (2)

substituting formula (1) into formula (2) to obtain:

F＝(Tmax-Tw)×P

＝(Tmax-kΔt+Ta)×P (3)

in the scheduling process, at any asynchronous scheduling moment, a virtual channel with the minimum value F of the urgency of a transmission frame is scheduled firstly. This is because the smaller the value F of the urgency of a frame is, the higher the real-time requirement of its virtual channel is. If two or more virtual channels have the value F of the urgency degree of the minimum frame at the same time, the static priority of the virtual channels needs to be considered, the virtual channel with the minimum static priority value is scheduled to be transmitted preferentially, and if more than one channel with the minimum static priority occurs at the same time, the channel with the small sequence number is selected and served preferentially. As shown in fig. 12.

Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and are not limited. Although the present invention has been described in detail with reference to the embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A satellite communication flow control method for coexistence of CCSDS frames and IP messages comprises the following steps:

the messages of the same type are loaded into a CCSDS data frame;

putting the CCSDS frame into a corresponding virtual channel;

2. The method according to claim 1, wherein the IP packet originates from a terrestrial local area network or a spatial local area network, and the CCSDS packet originates from a terrestrial local area network or a spatial local area network; the IP message carries data of a satellite service 7, a satellite service 6, a satellite service 5 and a satellite service 4; the CCSDS packet carries data of satellite service 3, satellite service 2, satellite service 1, and satellite service 0.

3. The method according to claim 2, wherein the performing type identification and marking on the received IP packet and CCSDS packet respectively comprises:

4. The method of claim 3, wherein the determining the priority is based on: according to the pre-stored corresponding relationship between the service type and the priority, the method specifically comprises the following steps:

identifying whether the message carries a priority;

5. The method of claim 3, wherein the determining the priority is based on: according to the pre-stored corresponding relationship among the service types, the users and the priorities, the method specifically comprises the following steps:

identifying whether the message carries a priority;

6. The method according to claim 1, wherein the identifying and marking the type of the received IP packet and the CCSDS packet further comprises: traffic shaping is used to limit the maximum traffic for each type of satellite traffic.

7. The method of claim 5, wherein the loading the messages of the same type into the CCSDS data frame specifically comprises:

8. The method of claim 7, wherein the pooled allocation of tokens dynamically allocates bandwidth for classes of CCSDS data frames; the method specifically comprises the following steps:

9. The method of claim 8, wherein the step of placing the tokens generated by each token generator into a token bucket according to a token pool allocation algorithm comprises:

10. The method of claim 9, wherein before placing the CCSDS packet into the corresponding virtual channel, further comprising: when bandwidth is allocated to data of which the service type is satellite service 3, satellite service 2, satellite service 1 or satellite service 0, scheduling the IP data packet specifically includes:

11. The method of claim 10, wherein the placing of the CCSDS packet into the corresponding virtual channel; the method specifically comprises the following steps:

12. The method of claim 11, wherein the dynamic frame-based urgency scheduling schedules virtual channels, and specifically comprises:

calculating the urgency F of the frame as:

F＝(Tmax-Tw)×P

tmax is the maximum scheduling delay allowed by the virtual channel, Tw is the waiting time of the frame in the virtual channel, and the time difference between the current scheduling time k △ t of the virtual channel and the arrival time Ta of the current first frame in the virtual channel is defined as the waiting time of the frame in the virtual channel, namely:

Tw＝kΔt–Ta

p is a static priority, the transmission real-time performance of each virtual channel before transmission is called as the static priority, the virtual channel with high real-time performance requirement can be endowed with the static priority with the numerical value of 1, and otherwise, the virtual channel with high real-time performance requirement is endowed with the static priority with the numerical value of 10; the maximum value of the static priority is 10, and the minimum value is 1;