CN116232427A

CN116232427A - Forward bandwidth allocation method for satellite network

Info

Publication number: CN116232427A
Application number: CN202310043466.9A
Authority: CN
Inventors: 徐钧; 刘浩
Original assignee: Space Engineering Network Technology Development Hangzhou Co ltd
Current assignee: Space Engineering Network Technology Development Hangzhou Co ltd
Priority date: 2023-01-29
Filing date: 2023-01-29
Publication date: 2023-06-06
Anticipated expiration: 2043-01-29
Also published as: CN116232427B

Abstract

The invention relates to the technical field of satellite networks, in particular to a forward bandwidth allocation method for a satellite network, which aims to solve the problem that dynamic resource allocation of a current forward link gateway and a three-layer gateway serving the current forward link gateway is difficult to balance. To this end, the method provided by the invention comprises the following steps: when a single forward link gateway allocates bandwidth for the three-layer gateway, the forward link gateway allocates bandwidth for the three-layer gateway according to the bandwidth requirements and the priority ordering of the three-layer gateway and the bandwidth capacity of the forward link gateway; when the multi-forward link gateway allocates bandwidth for the three-layer gateway, the bandwidth manager dynamically adjusts the bandwidth allocation values of the plurality of forward link gateways by using the bandwidth manager according to the bandwidth requirements and the priority ordering of the three-layer gateway, and sends the bandwidth allocation values to the plurality of three-layer gateways. The bandwidth manager dynamically calculates the connection topology of the three-layer gateway and the two-layer gateway by collecting the data backlog value of the three-layer gateway reported by the forward link gateway, and the service gateway can be dynamically increased or decreased.

Description

Forward bandwidth allocation method for satellite network

Technical Field

The invention relates to the technical field of satellite networks, and particularly provides a forward bandwidth allocation method for a satellite network.

Background

Satellite communication is one of important applications of satellite networks, and has the advantages of wide coverage, flexible bandwidth allocation, high resource utilization rate, convenient user access, no limitation of various regional conditions and the like, and has wide application range, and can meet the requirement of mass data communication. Thus, a sanitary communication system will play an important role in network applications as well.

Currently, researches on a satellite network forward bandwidth allocation method mainly comprise: the research is directed to the situation that one forward link gateway (MCS) serves a plurality of three-layer gateways (IPGWs), and if the allocatable bandwidth of one forward link gateway is less than the total bandwidth applied by the three-layer gateways, the forward bandwidth is firstly allocated to the three-layer gateway with the front priority based on the service priority. However, in a satellite network system, there are typically multiple three-layer gateways and multiple forward link gateways. One forward link gateway may serve multiple tri-layer gateways, and one tri-layer gateway may also have multiple forward link gateways to serve it, so the forward bandwidth needs to be dynamically adjusted between the multiple forward link gateways and the multiple tri-layer gateways to achieve load balancing.

Accordingly, there is a need in the art for a new forward bandwidth allocation scheme for satellite networks to address the above-described problems.

Disclosure of Invention

The invention aims to solve the technical problems, namely the problem that dynamic resource allocation of a current forward link gateway and a three-layer gateway served by the current forward link gateway is difficult to balance, and provides a forward bandwidth allocation method for a satellite network.

The invention provides a forward bandwidth allocation method for a satellite network, which comprises the following steps:

when one forward link gateway distributes bandwidth for a plurality of three-layer gateways, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth distribution value according to the bandwidth requirement and the priority order of each three-layer gateway, and distributes the bandwidth distribution value to the plurality of three-layer gateways;

when the forward link gateways allocate bandwidths for the three-layer gateways, bandwidth allocation values of the three-layer gateways connected with each forward link gateway are calculated according to the bandwidth requirement and the priority order of each three-layer gateway, and the bandwidth manager dynamically adjusts the bandwidth allocation values of the forward link gateways according to the bandwidth capacity of each forward link gateway to obtain adjusted bandwidth allocation values and returns the adjusted bandwidth allocation values to each forward link gateway, and each forward link gateway sends the adjusted bandwidth allocation values to the three-layer gateways.

In one technical scheme of the foregoing forward bandwidth allocation method for a satellite network, when one forward link gateway allocates bandwidth to a plurality of three-layer gateways, the forward link gateway calculates a bandwidth allocation value according to a certain ratio of bandwidth capacity according to bandwidth requirements and priority ranks of each three-layer gateway, and allocates the bandwidth allocation value to the plurality of three-layer gateways, including:

when the system is just started or a new three-layer gateway is added, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth allocation value, wherein the bandwidth allocation value is as follows:

/>

wherein ,λ_total Total service bandwidth, w, provided for all priorities of forward link gateway _i For the weight of priority i in the forward link gateway, J is the bandwidth application of J three-layer gateways, G _ij The bandwidth allocated to the three-layer gateway j priority i;

when the total bandwidth applied by the three-layer gateways is smaller than or equal to the total bandwidth provided by one forward link gateway, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth allocation value, which comprises the following steps:

calculating the ratio of the forward link gateway provided bandwidth to the bandwidth requirements of the plurality of three-layer gateways, wherein the ratio is as follows:

wherein M is M priorities, R _ij The bandwidth of the priority i applied for the three-layer gateway j;

computing forward link gateway to allocate bandwidth to multiple tri-layer gatewaysThe method comprises the following steps: g _ij ＝η·R _ij ；

When the total bandwidth of the plurality of three-layer gateway applications is greater than the total bandwidth provided by the forward link gateway and less than or equal to the expansion ratio value of the total bandwidth provided by the forward link gateway

And when the Θ is a switching proportion value of a bandwidth allocation algorithm, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain the bandwidth allocation value, which comprises the following steps:

calculating the sum of the application bandwidths of each priority i, wherein the sum is as follows:

the unused bandwidth of each priority i is calculated as: z is Z _i ＝λ _i -R _i, wherein ,λ_i Service bandwidth provided for priority i on behalf of the forward link gateway;

wherein if Z _i Negative, then indicates insufficient allocation; if Z _i Positive, then indicates adequate allocation;

will Z _i <I of 0 joins set I ^- Will Z _i >I of 0 joins set I ⁺ ；

For set I ⁺ Is: g _i ＝R _i ；

And calculates the sum of all the residual bandwidths as:

wherein ,N₊ For set I ⁺ The number of elements;

set I according to the forward link gateway capability ^- Initial bandwidth is allocated for priority of (a): g _i ＝λ _i ；

For set I ^- Is assigned a weight by priority calculation of (1) set ^- The number of elements of (2) is N _- The formula for assigning weights is:

or->

wherein ,

for the ith priority in set I ^- The assigned weights of (a);

for set I ^- Other priority residual bandwidths are allocated to the priorities of the (a) and the (b) is calculated as follows:

if Z _i Not less than 0, remove I ^- ；

If Z ₊ >0, indicating that the last round of adjustment does not completely allocate the residual bandwidth, and continuing to repeat the above steps until Z ₊＝0 or I^- Is an empty set;

the bandwidth is allocated to each priority of each three-layer gateway, and the calculation formula is as follows:

when the total bandwidth of the plurality of tri-layer gateway applications is greater than the total bandwidth provided by the forward link gateway and greater than the expansion of the total bandwidth provided by the forward link gatewayLarge scale value

When the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth allocation value, the method comprises the following steps:

for the priorities 1 to M-1, a calculation step is used, wherein when the total bandwidth applied by the three-layer gateways is larger than the total bandwidth provided by the forward link gateway and smaller than or equal to the expansion proportion value of the total bandwidth provided by the forward link gateway, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth allocation value;

for priority M, the calculation method is:

calculating the bandwidth value that priority M can allocate:

According to the bandwidth requirement of the three-layer gateway and the service plan of the user, calculating the service weight of each IP gateway as follows:

wherein ,u_hj The number of service users planned in the h service for the j-th IP gateway;

maximum transmission bandwidth for the service plan; g _h Satisfaction weight for the service plan;

calculating the service weight sum of a plurality of three-layer gateways served by one forward link gateway, wherein the service weight sum is as follows:

wherein ,ur_j The service weight of the jth three-layer gateway serving the forward link gateway;

allocating bandwidth for the priority M of each three-layer gateway, wherein the bandwidth is as follows:

in one technical solution of the foregoing forward bandwidth allocation method for a satellite network, when a plurality of forward link gateways allocate bandwidths for a plurality of tri-layer gateways, according to bandwidth requirements and priority ranks of each tri-layer gateway, calculating a bandwidth allocation value of each forward link gateway to the tri-layer gateway connected thereto, and dynamically adjusting, by a bandwidth manager, the bandwidth allocation values of the plurality of forward link gateways according to bandwidth capacities of each forward link gateway, to obtain adjusted bandwidth allocation values and return the adjusted bandwidth allocation values to each forward link gateway, where each forward link gateway sends the adjusted bandwidth allocation values to the plurality of tri-layer gateways, including:

The forward link gateway sends a heartbeat message to the three-layer gateway in a first period;

the three-layer gateway collects bandwidth requirement values of the three-layer gateway in a second period, encapsulates the collected forward bandwidth requirement values of the three-layer gateway into BW solicitation messages to be sent to the forward link gateway after receiving the heartbeat messages sent by the forward link gateway, and applies forward bandwidth to the forward link gateway;

after receiving the BW Solict message, the forward link gateway calculates a forward bandwidth allocation plan;

if the bandwidth manager is not on-line, the forward link gateway directly transmits the calculated forward bandwidth allocation plan to the three-layer gateway;

if the bandwidth manager is online, the bandwidth manager acquires connection topology of the forward link gateway and the three-layer gateway in a third period, the forward link gateway sends the calculated forward bandwidth allocation plan to the bandwidth manager according to the connection topology of the forward link gateway and the three-layer gateway, the bandwidth manager dynamically adjusts among the forward link gateways by checking the promised service upper limit of the three-layer gateway to obtain an adjusted forward bandwidth allocation plan, the bandwidth manager returns the adjusted forward bandwidth allocation plan to the forward link gateway, and the forward link gateway sends the adjusted forward bandwidth allocation plan to the three-layer gateway.

In one aspect of the foregoing method for allocating forward bandwidth for a satellite network, the bandwidth manager obtains connection topologies of a forward link gateway and a tri-layer gateway in a third period, including:

and the bandwidth manager samples the connection topology of the forward link gateway and the three-layer gateway in a third period, calculates a topology configuration value according to the sampled data, and updates the connection topology of the forward link gateway and the three-layer gateway if the topology configuration value is updated.

In one technical solution of the foregoing forward bandwidth allocation method for a satellite network, the bandwidth manager samples connection topologies of a forward link gateway and a tri-layer gateway in a third period, calculates a topology configuration value according to data obtained after sampling, and updates connection topologies of the forward link gateway and the tri-layer gateway if the topology configuration value is updated, including:

starting sampling, and resetting the sampling times of the last third period;

the bandwidth manager sends a heartbeat message to the forward link gateway, and calculates a connection matrix A (i) according to the data backlog of the three layers of gateways collected by the forward link gateway;

adding the sampled A (i) to obtain a connection matrix A, and comparing the connection matrix A with a threshold to obtain a connection topology ANorm;

When bandwidth allocation is adjusted, the bandwidth manager generates a connection matrix Etopo according to data backlog reported by the forward link gateway;

judging whether the topology is changed according to the connection matrix Etopo and the connection topology ANorm, and if so, updating the connection topology of the forward link gateway and the three-layer gateway in time.

In one aspect of the foregoing method for allocating forward bandwidth in a satellite network, according to a connection topology of a forward link gateway and a tri-layer gateway, the forward link gateway sends a calculated forward bandwidth allocation plan to a bandwidth manager, and the bandwidth manager dynamically adjusts between a plurality of forward link gateways by checking a guaranteed service upper limit of the tri-layer gateway, to obtain an adjusted forward bandwidth allocation plan, including:

summarizing the data backlog and bandwidth allocation of each forward link gateway;

determining three layers of gateways to be adjusted;

reducing the allocation value of the three-layer gateway with the allocation bandwidth larger than the upper limit of the promised bandwidth;

the allocation value of the three-layer gateway with the allocation bandwidth smaller than the upper limit of the promised bandwidth is increased.

In one aspect of the foregoing method for allocating forward bandwidth for a satellite network, the summarizing the data backlog and the bandwidth allocation of each forward link gateway includes:

Calculating total data backlog reported to the ith forward link gateway by the jth three-layer gateway, and summarizing the corresponding priority to obtain:

wherein M represents priority, and the value range is more than or equal to 1 and less than or equal to M; j represents a three-layer gateway, and the value range is more than or equal to 1 and less than or equal to J; i represents a forward link gateway, and the value range is 1-N; q (Q) _ij (m) represents the total backlog of the mth priority queue reported by the jth tier gateway to the ith forward link gateway;

calculating the total bandwidth allocated to the jth three-layer gateway by the ith forward link gateway, and summarizing the corresponding priority to obtain

wherein ,C_ij (m) bandwidth allocated to the mth priority queue for the ith forward link gateway to the jth tri-layer gateway;

the total bandwidth of the ith forward link gateway is calculated, and the three-layer gateways connected to each forward link gateway are summarized,

the total bandwidth allocated by the j-th three-layer gateway is calculated, the forward link gateway connected with each three-layer gateway is summarized,

in one technical solution of the foregoing forward bandwidth allocation method for a satellite network, the determining a three-layer gateway to be adjusted includes:

allocating bandwidth allocation value as initial value of all bandwidths of the ith MCS allocated to the jth IP gateway after adjustment

The method comprises the following steps: />

And the three-layer gateway set with the allocation bandwidth larger than the upper limit of the promised bandwidth is as follows:

the three-layer gateway set with the allocated bandwidth smaller than the upper limit of the promised bandwidth is as follows:

wherein ,/>

The upper limit of promised service for the jth three-layer gateway;

if Z ⁺ If the set is an empty set, the output parameters are returned without adjustment:

if Z ⁺ And not the empty set, then the adjustment is made.

In one aspect of the foregoing method for allocating forward bandwidth for a satellite network, the reducing allocation values of the three-layer gateway with allocation bandwidth greater than the upper limit of the guaranteed bandwidth includes:

initializing per forward link gateway bandwidth allocation reduction value D _ij ＝0，j∈Z ⁺ ；

The upper limit of the promised bandwidth of the three-layer gateway is fairly distributed to the forward link gateway connected with the three-layer gateway according to the lengths of the data backlog queues of the different forward link gateways,

when meeting H _kj =1, then it indicates that the forward link gateway is connected to the tri-layer gateway;

the current bandwidth update value for the three-tier gateway j is initialized,

obtaining the reduced value of all forward link gateway bandwidth allocation connected with the three-layer gateway j when the condition is that

Satisfying, the following loop steps are circularly executed;

for meeting H _kj K=1, the following loop steps are performed in a loop;

if it is

The bandwidth allocated by the forward link gateway may be reduced to：

If all k have been satisfied

Or->

Meeting, jumping out of the circulation step;

recording the distribution reduction value of all forward link gateway bandwidths connected by the three layers of gateways j as follows:

if Z ^- If the bandwidth is empty, the adjustment is finished, new adjusted bandwidth is redistributed according to the original priority distribution proportion, other parameters are unchanged, and the output parameters are output

If Z ^- Instead of an empty set, a summary of each forward link gateway bandwidth allocation reduction value is calculated, which may be adjusted to set Z ^- The total value of the three-layer gateway is as follows:

in one aspect of the foregoing method for allocating forward bandwidth for a satellite network, the allocating values of the three-layer gateway that increase the allocated bandwidth to be less than the upper limit of the guaranteed bandwidth include:

for meeting in MCS

The index numbers of (2) are ordered to obtain an ordered list set N _sort Wherein three layers of gateways connected with a forward link gateway are arranged in front in a small number;

for i epsilon N _sort The following first loop steps are performed in a traversal manner:

when (when)

The following second loop step is performed under the condition:

handle

Adjusting unallocated bandwidth to be preferentially allocated to the three-layer gateway which does not meet the bandwidth application;

For all j E Z ^- The third loop step is circularly executed:

if H _ij＝1 and

if j satisfies

Then from set Z ^- Removing j;

if all j E Z ^- All have satisfied

The third cycle step is skipped;

if setZ is in the same position ^- If the first circulation step is empty, the second circulation step is jumped out;

if it is

Jumping out of the second circulation step;

if it is

The rest is that the unallocated bandwidth is regulated and then allocated to the IP gateway which does not reach the upper limit of the promised bandwidth of the three-layer gateway;

for all j E Z ^- Performing a fourth loop step:

if H _ij ＝1

If j satisfies

Then from set Z ^- Removing j;

if set Z ^- If the first circulation step is empty, the second circulation step is jumped out;

if it is

Jumping out of the second circulation step; />

And reallocating the new adjusted bandwidth according to the original priority allocation proportion, and returning to output parameters:

one or more of the above technical solutions of the present invention at least has one or more of the following

The beneficial effects are that:

in the technical scheme of implementing the invention, a forward bandwidth allocation method for a satellite network is provided, the method selectively uses different bandwidth allocation methods to obtain the bandwidth allocation value of the forward link gateway to the three-layer gateway according to the actual situation that one forward link gateway allocates bandwidth to a plurality of three-layer gateways or a plurality of forward link gateways allocate bandwidth to a plurality of three-layer gateways, the method solves the problem that the dynamic resource allocation of the current forward link gateway and the three-layer gateway served by the current forward link gateway is difficult to balance, provides service quality guarantee for user terminals in the network under the condition of limited bandwidth, and basically starts to flexibly allocate and dynamically adjust the available resources of wireless channels and the network under the conditions of uneven network data volume distribution, fluctuation change of channel characteristics due to channel weakness and interference and the like, thereby maximally improving the utilization rate of inter-satellite links and inter-satellite links, preventing network congestion and keeping the signaling load as small as possible. A bandwidth manager is necessary on the forward link to perform overall planning and dynamic adjustment for bandwidth application and allocation of multiple two-layer gateways and multiple three-layer gateways. The load balancing of the link bandwidth in the three-layer gateway and the two-layer gateway can be realized through the dynamic adjustment of the bandwidth manager. And under the condition of the same physical link resources, the plurality of groups of users are maximally served.

Drawings

The present disclosure will become more readily understood with reference to the accompanying drawings. As will be readily appreciated by those skilled in the art: the drawings are for illustrative purposes only and are not intended to limit the scope of the present invention. Moreover, like numerals in the figures are used to designate like parts, wherein:

fig. 1 is a flow chart illustrating the main steps of a forward bandwidth allocation method for a satellite network according to one embodiment of the present invention;

FIG. 2 is a flow diagram of interoperation data and control of a plurality of tri-layer gateways and a forward link gateway according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a framework of multiple tri-layer gateways and multiple forward link gateway bandwidth requests and allocations in accordance with one embodiment of the present invention;

FIG. 4 is a diagram of bandwidth manager BM topology sampling update according to one embodiment of the present invention;

fig. 5 is a flow chart of a bandwidth allocation message synchronization mechanism in a forward bandwidth allocation method for a satellite network according to one embodiment of the present invention.

Detailed Description

Some embodiments of the invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.

Referring to fig. 1, fig. 1 is a flow chart illustrating main steps of a forward bandwidth allocation method for a satellite network according to an embodiment of the present invention. As shown in fig. 1, the forward bandwidth allocation method for a satellite network according to the embodiment of the present invention mainly includes the following steps S11 to S12.

Step S11: when one forward link gateway distributes bandwidth for a plurality of three-layer gateways, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth distribution value according to the bandwidth requirement and the priority order of each three-layer gateway, and distributes the bandwidth distribution value to the plurality of three-layer gateways;

step S12: when the forward link gateways allocate bandwidths for the three-layer gateways, bandwidth allocation values of the three-layer gateways connected with each forward link gateway are calculated according to the bandwidth requirement and the priority order of each three-layer gateway, and the bandwidth manager dynamically adjusts the bandwidth allocation values of the forward link gateways according to the bandwidth capacity of each forward link gateway to obtain adjusted bandwidth allocation values and returns the adjusted bandwidth allocation values to each forward link gateway, and each forward link gateway sends the adjusted bandwidth allocation values to the three-layer gateways.

In this embodiment, in the satellite network system, the Gateway GW (Gateway) system of the terrestrial segment includes a three-layer Gateway IPGW (IP Gateway), and a forward link Gateway MCS (MODCOD Servicing System).

The three-layer gateway IPGW applies forward bandwidth according to the service plan and the data backlog quantity of the user, and the forward link gateway distributes the forward bandwidth. One forward link gateway may allocate forward bandwidth to multiple tri-layer gateways, and one tri-layer gateway may also have multiple forward link gateways allocated forward bandwidth.

When one forward link gateway distributes bandwidth for a plurality of three-layer gateways, and each three-layer gateway only has one forward link for distributing bandwidth, the forward link gateway calculates bandwidth distribution values according to a certain proportion according to the bandwidth capacity of the forward link gateway and the bandwidth requirement and priority order of each three-layer gateway, and distributes the bandwidth distribution values to the plurality of three-layer gateways connected with the forward link gateway.

Specifically, the data source of the forward link gateway MCS is an IP gateway and the destination is a physical layer modulation board card. The MCS needs to perform data packet and forwarding according to the transmission capability of the physical layer modulation card. The MCS needs to collect the sending rate while forwarding the data, estimate the physical channel bandwidth and report to its own bandwidth allocation manager. All IPGWs of the MCS service send bandwidth requirements to the bandwidth allocation manager of the MCS, which is responsible for allocating the bandwidth provided by the MCS to different IPGWs according to a certain ratio. And the IP gateway performs data scheduling according to the output of the bandwidth allocation manager to realize flow control. The IP gateway and the MCS cooperate with each other to perform data transmission with maximum efficiency. IP gateway and MCS interoperation data and control flows are shown in fig. 2.

In one implementation of the embodiment of the present invention, the step of allocating bandwidth by a single forward link gateway to a plurality of tri-layer gateways includes steps S111-S114.

First, a definition of the letter representatives used in the steps is given.

M: the default (highest value of priority in this embodiment) is 4, and the system does not change dynamically during operation (the lower corner i indicates priority);

j: j IPGWs are indicated to participate in bandwidth application, and dynamic increase and decrease can occur in system operation (the lower corner mark J indicates an IP gateway);

n: indicating that there are N forward link gateway MCS services, dynamic increase or decrease occurs in the system operation (the forward link gateway allocation algorithm considers only one MCS, i.e., n=1);

h: representing that there are H Service plans, each Service plan having a maximum Service rate, all users having their own Service plan, service plan (the Service plan is represented by the subscript H);

maximum transmission bandwidth of service plan, unit kbps, satisfies +.>

A network management configuration value;

g _h : the satisfaction degree weight of the service plan, the network management configuration value, default to 1;

u _hj : the j-th IP gateway reports the value of the number of service users in the h-th service plan;

w _i : priority i weight in MCS;

λ _i : MCS may be the service bandwidth provided by priority i, in kbps (algorithm input value, MCS provided);

R _ij : the bandwidth of priority i applied by IP gateway j, unit kbps (algorithm input value, IPGW provides);

Θ: the bandwidth allocation algorithm switches the proportional value, the network management configuration parameter is 2.0 by default;

λ _total : the total service bandwidth provided by all the priorities of the MCS is calculated as the following formula in kbps:

G _ij : the IP gateway j priority i is allocated to the bandwidth, the unit kbps (algorithm output value)

Step S111: when the system is just started or a new three-layer gateway is added, the forward link gateway calculates the bandwidth capacity according to a certain proportion to obtain a bandwidth allocation value.

Specifically, the execution condition may be satisfied by either one of the following:

at system start-up, R is present for all i and all j _ij ＝0；

There is a new IPGW addition, i.e., J is increased by 1;

the average calculation allocates bandwidth as:

step S112: when the total bandwidth of the plurality of three-layer gateway applications is less than or equal to the total bandwidth provided by one forward link gateway, i.e., the execution condition is that the total bandwidth of all IPGW applications is less than or equal to the total bandwidth provided by the MCS,

Calculating the ratio of the MCS provided bandwidth to the IPGW applied bandwidth as follows:

calculating an allocated bandwidth as: g _ij ＝η·R _ij ；

Step S113: when the total bandwidth applied by the three-layer gateways is greater than the total bandwidth provided by the forward link gateway and is less than or equal to the expansion ratio value of the total bandwidth provided by the forward link gateway, the execution condition is that: the total bandwidth of all IPGW applications is greater than the total bandwidth provided by the MCS, but less than or equal to the expansion ratio value of the provided bandwidth,

when the forward chainThe method for calculating the bandwidth allocation value by the gateway according to a certain proportion comprises the following steps:

1. calculating the sum of the application bandwidths of each priority i, wherein the sum is as follows:

2. calculating the unused bandwidth of each priority i, if the number of the unused bandwidths is not allocated enough, the unused bandwidth is expressed as a negative number, the unused bandwidth is expressed as a positive number, the unused bandwidth is remained, and i meets Z _i >0 then add set I ⁺ I satisfies Z _i <0 then add set I ^- The method comprises the following steps: z is Z _i ＝λ _i -R _i ；

3. For set I ⁺ And calculates the sum of all the remaining bandwidths, N ₊ For set I ⁺ The number of elements is as follows: g _i ＝R _i ；

4. Capability per MCS is set I ^- Initial bandwidth is allocated for priority of (a): g _i ＝λ _i ；

5. For set I ^- Is assigned a weight by priority calculation of (1) set ^- The number of elements of (2) is N _- One of the allocation formulas is selected, and the first formula is allocated according to the weight, and is as follows:

The second formula is distributed according to the number average, and is: />

wherein ,/>

For the ith priority in set I ^- Is assigned a weight. Specifically selecting which algorithm is determined according to the network management configuration value; />

6. For set I ^- Is allocated to other priority residual bandwidths, and is calculatedThe formula is as follows:

G _i ＝G _i +min(φ _i ·Z ₊ ，-Z _i )；

Z ₊ ＝Z ₊ -min(φ _i ·Z ₊ ，-Z _i )；

Z _i ＝Z _i +min(φ _i ·Z ₊ ，-Z _i )；

if Z _i Not less than 0, remove I ^-

7. If Z ₊ >0, meaning that the last round of adjustment does not completely allocate the residual bandwidth, returning to step 5 and repeatedly executing until Z ₊＝0 or I^- Step 8 is entered for the empty set.

8. Bandwidth is allocated to each priority of each IPGW, and the calculation formula is as follows

Step S114: when the total bandwidth applied by the three-layer gateways is greater than the total bandwidth provided by the forward link gateway and greater than the expansion ratio value of the total bandwidth provided by the forward link gateway, the implementation condition is that: the total bandwidth of all IPGW applications is much greater than the total bandwidth offered by the MCS, i.e. greater than the scale-up value of the offered bandwidth,

for priorities 1 to M-1, G is calculated using the algorithm of step S113 _ij It should be noted that the calculation and comparison ranges for all priorities are 1 to M-1 instead of 1 to M.

The following algorithm is used for priority M:

the bandwidth value that can be allocated by the calculation priority M is:

the service weight of each IP gateway is calculated according to the number of service subscribers of the IP gateway and the service plan of the subscriber (since this value remains unchanged for the case of one IPGW based on the number of service plan, it may not be necessary to calculate each time, but it needs to be recalculated when the input changes or configuration changes), as:

the sum of the service weights of all IPGWs of one MCS service is calculated (as long as there is no new IPGW addition and a single IPGW change, this value remains unchanged all the time, not necessarily every calculation), as:

allocating bandwidth for the priority M of each IPGW as:

in addition, there will typically be multiple three-layer gateway IPGWs in the satellite network system, as well as multiple forward link gateways MCSs. One MCS may serve multiple tri-layer gateway IPGWs, and one tri-layer gateway IPGW may also have multiple forward link gateways MCS to serve it. The bandwidth needs to be dynamically adjusted between a plurality of forward link gateways MCS and a plurality of three-layer gateways IPGW, so as to realize load balancing. For this purpose, a bandwidth manager BM (Bandwidth Manager) network element is introduced in the present invention to realize resource management, i.e. bandwidth allocation. The IPGW is used as a bandwidth applicant and provides a bandwidth Demand (Demand); MCS is taken as a server to provide service bandwidth, and a bandwidth allocation manager of the MCS is responsible for bandwidth allocation managed by the MCS; the bandwidth manager BM is used as a distribution center to coordinate and schedule the bandwidth distribution plans provided by a plurality of MCSs, and dynamically adjust the bandwidth distribution plans. Bandwidth request and allocation are shown in fig. 3.

Within a bandwidth allocation management autonomous domain, a bandwidth manager BM, N MCSs (forward link gateways) and J IPGWs (tri-layer gateways) are typically deployed. The IPGW receives data from the core network and sends it to the end user via the MCS. To achieve QoS (Quality ofService ), priority-based buffer queues are introduced in both IPGW and MCS, with priority number M. The IPGW applies bandwidth to the MCS to which it is connected based on the scheduling queue length. The MCS allocates bandwidth to the IPGW for which it provides an application according to its own encapsulation transmission capability.

In general, one MCS and a plurality of IPGWs connected thereto can implement self-management through bandwidth application and bandwidth allocation.

But if each MCS is allocated according to its own capability, it may cause that some IPGWs are allocated too much bandwidth and some IPGWs are allocated little bandwidth. But a decentralized bandwidth allocation algorithm has no way to coordinate between multiple MCSs. To this end, a bandwidth allocation adjustment algorithm is introduced at the bandwidth manager. All MCSs send the self-generated bandwidth allocation schemes to the BM uniformly, the BM dynamically adjusts among a plurality of MCSs by looking up the upper limit of the promised service of the IPGW, and then returns the adjusted schemes to the MCS, and the MCS sends the adjusted bandwidth allocation schemes to the IPGW.

Thus, the bandwidth manager needs to know the three layers of service gateways and forward link gateways that exist in the system, as well as their connection topology. The adoption of a static configuration method is not beneficial to the dynamic capacity expansion of the system. The invention supports dynamic awareness of connection topology. As shown in fig. 4, the BM performs topology aware sampling at fixed time intervals. The number of consecutive samples is determined by the network management configuration parameters and then the topology is calculated. This topology configuration value is used until the next update period t_update has not expired.

And stopping BM calculation when the update period is not up but the topology map detects that the update exists, and calculating bandwidth allocation by the MCS until the new topology update is completed.

The following cases are considered as topology updates:

1) With new MCS addition, there is more MCS report data backlog (backlog) than the original topology after the BM sends a heartbeat (heartbeat);

2) The MCS is offline, namely, no MCS reply heartbeat report backlog is received for K+1 times continuously;

3) A new IPGW is added, and when the MCS replies the backlog of the heartbeat report, the new IPGW is added;

4) The bandwidth request backlog of the IPGW for MCS reporting has an IPGW loss (IPGW loss algorithm and MCS loss, calculated by MCS).

In one embodiment, the bandwidth manager may automatically learn to obtain the topology corresponding to the forward link gateway and the three-layer gateway, which is specifically as follows:

First, the definition of the letter representation used in the step is given.

T ₀ : the BM periodically sends a heartbeat time in ms, 100 by default. BM collects backlog of IPGW reporting MCS through heartbeat for topology automatic learning

T _update : BM update topology time, unit ms, default 50T ₀ I.e. 5000

K _sample : topology continuous sampling times, default to 3

e _th : the threshold number of the topology calculation is online, when the sampling is carried out, the backspace number is greater than or equal to the threshold and is active, otherwise, the threshold number is inactive, and the default is 1

K: the maximum number of Update requests allowed to be continuously lost by bandwidth calculation is 3 by default. And using the last available value within the allowable range, and stopping calculation when the last available value exceeds the allowable range.

The sampling calculation topology algorithm is as follows:

starting sampling and resetting the sampling times.

Transmitting a heartbeat, calculating a connection matrix A according to backlog ₁ A backlog of 0 or more is considered connected, and a 1 is counted, and no backlog is considered disconnected, and a 0 is counted. The last backlog report cannot be used at the time of sampling.

The sampling times are increased by 1, and the sampling is repeatedly executed until the sampling times reach K _sample . At this time we get K _sample And a plurality of connection matrices.

Handle K _sample Adding up the connection matrixTogether, a new matrix a is obtained. For each element in matrix A, a check is made that only the value is equal to or greater than e _th We consider the row MCS to be connected to the column IPGW and disconnected otherwise.

By default K _sample For example, =3, the explanation algorithm is as follows. Assuming 5 MCSs and 6 IPGWs, the connection matrix is 5 rows and 6 columns. The three samples are respectively:

the connection topology obtained after comparison with the threshold is as follows

The simplified topological diagram is as follows

Based on the current connection topology, there are three MCSs (2, 3,4, respectively) and four IPGWs (1, 3,4,5, respectively) in an on-line state.

For backlog of MCS report, no connection may be indicated by-1, 0 indicates connected but no backlog. Such as

From the definition, Q (1) and A _topo The match state is the case, and there is no increase or decrease in MCS and IPGW.

From the definition, Q (2) and A _topo Is a mismatch state, Q _3,4 =20, should have no backlog, indicating that there is an IPGW5 on MCS4 newly.

The following algorithm may be used for whether the backlog matches the topology matrix:

generating a backlog connection matrix E from the backlog matrix Q _topo ＝{e _i，j }

If E _topo -A _topo = {0}, then match, otherwise do not.

In one implementation manner of the embodiment of the present invention, when a plurality of forward link gateways allocate bandwidths for a plurality of tri-layer gateways, according to a bandwidth requirement and a priority order of each tri-layer gateway, a bandwidth allocation value of the tri-layer gateway to which each forward link gateway is connected is calculated, and a bandwidth manager dynamically adjusts the bandwidth allocation values of the plurality of forward link gateways according to a bandwidth capacity of each forward link gateway, so as to obtain adjusted bandwidth allocation values and return the adjusted bandwidth allocation values to each forward link gateway, where each forward link gateway sends the adjusted bandwidth allocation values to the plurality of tri-layer gateways, including:

In one specific example, within one BM management domain, there are multiple MCSs and multiple IPGWs. The data acquisition points are distributed in the MCS and IPGW, and a set of message synchronization mechanism is required to be defined for realizing bandwidth request and bandwidth allocation.

Between the IPGW and the MCS, a heartbeat message is sent by the MCS, synchronizing the bandwidth request and allocation. And transmitting a heartbeat message by the BM between the BM and the MCS, synchronizing the network topology connection, transmitting a bandwidth update request by the MCS, and replying the bandwidth update by the BM. If the two heatbeams can be synchronized, the optimal operation of the system can be ensured. Even if two headbeans cannot be synchronized, since the IPGW bandwidth request is a smooth value, the system can be guaranteed to operate stably. The following scheme is based on two heatbeams that do not have to be designed synchronously.

As shown in fig. 5, the messages represented by the two-layer gateway to three-layer gateway bars are synchronized by MCS through the heartbeat, and the bandwidth manager to two-layer gateway bars are synchronized by BM through the heartbeat. The dashed line indicates a possible occurrence, i.e. the condition for the occurrence of the solid line is not satisfied.

Three-layer gateway workflow:

the IPGW collects the length of the queue according to the scheduling period of the IPGW and prepares the requirement value of the bandwidth application. If multiple MCS services exist, it is necessary to ensure that the bandwidth requirement value acquisition points are at the same time.

The IPGW receives the heartbeat message, encapsulates the acquired bandwidth requirement value into a BW Solicit message, and sends the BW Solicit message to the MCS for bandwidth application.

And the IPGW receives the BWAadvertisement message, updates the bandwidth output upper limit of the IPGW, and performs scheduling forwarding on a data plane according to the new bandwidth allocation value.

Two-layer gateway workflow:

MCS periodically transmits a heartbeat message according to own clock, and records the current time as t _{MCS，heartbeat} The connected IPGW is required to report bandwidth requirements. The receipt of the IPGW's bandwidth request before the next heartbeat considers the IPGW to be online. If the continuous K (network management configuration value) periods do not receive the bandwidth request, the IPGW is considered to be offline, and the offline IPGW is allocated only according to the minimum request bandwidth (the network management configuration value can be 0) configured by the system.

And the MCS receives the BW Solict message and updates the bandwidth request of the IPGW.

MCS at t _{MCS，heartbeat} +Δt _MCS,C At this time, the bandwidth allocation calculation is started regardless of whether all IPGW BW Solicit messages are received. The MCS uses the bandwidth requirements as follows: if BW Solict is received and the bandwidth requirement of the period is used, if BW Solict is not received and the bandwidth requirement of the last period is used, if BW Solict is not used in the continuous K periods, the configured minimum value is used.

After the MCS calculates the bandwidth allocation plan, if the BM immediately reports BW Update Request message to the BM on line, the Timer is started at the same time. If BM is not on-line, the BWAadvertisement message is sent to IPGW immediately.

The MCS receives BW Update Response of the BM, and if the Timer time is not reached, immediately forwards the bandwidth of the BM to the IPGW, and stops the Timer. If the Timer time has arrived, it is indicated that the bandwidth allocation for the present period has been sent to the IPGW, and is not necessarily forwarded to the IPGW, but the latest bandwidth allocation update value needs to be recorded for the next use.

MCS at t _{MCS，heartbeat} +Δt _MCS,R At that moment, a bwoadvertisement message is sent to the connected IPGW. Bwoadvertisement values are organized according to the following rules: if no BW Update is returned to the allocation value of the MCS in the continuous K periods, otherwise, the last BW Update is returned.

Bandwidth manager workflow:

BM periodically transmits a heartbeat message to all MCSs according to own clock, and records the current time as t _{BM，heartbeat} 。

And the BM receives the TopologyUpdate message sent by the MCS and updates the connection matrix.

The BM receives all BWUpdate Request messages sent by the MCS and stores the latest Request and allocation.

BM at t _{BM，heartbeat} +Δt _BM,C At this point, the bandwidth adjustment calculation is started regardless of whether bwopdate Request updates for all MCSs are received. The BM processes the input values according to the following rules: the Request of the present cycle is used if BW Update Request is received, the Request of the last cycle is used if no bwopdate Request is received, and the configured minimum value is used if no bwopdate Request is received for K consecutive cycles.

The BM encapsulates the calculated update value in a BWUpdate Response message, immediately sends to all MCSs, and forwards to the IPGW from the MCS.

In one implementation manner of the embodiment of the present invention, the bandwidth manager acquires connection topologies of the forward link gateway and the three-layer gateway in a third period, including:

In one implementation of the embodiment of the present invention, according to the connection topology of the forward link gateway and the tri-layer gateway, the forward link gateway transmits the calculated forward bandwidth allocation plan to the bandwidth manager, and the bandwidth manager dynamically adjusts between the plurality of forward link gateways by checking the guaranteed service upper limit of the tri-layer gateway, so that the steps of obtaining the adjusted forward bandwidth allocation plan include steps S121 to S124.

First, the definition of the letter representation used in the step is given.

M: representing M priorities, wherein the default is 4, and the system is not dynamically changed in the running process (M represents the priority below);

n: indicating that there are N forward link layer gateway MCS services, dynamic increase and decrease can occur in the system operation (the lower corner mark i indicates MCS);

the j-th IPGW promises service upper limit, the unit kbps, the network management configuration value, default 500Mbps;

H _ij : the matrix of N rows and J columns represents the connection condition of MCS and IPGW. If the ith MCS is connected with the jth IPGW, it is 1, otherwise it is 0.

Δd: the step size is adjusted, the unit kbps, the default value is 100, and the network management configuration value.

a: adjusting judging ratio, default value 0.5, network management configuration value.

Q _ij (m): the jth IPGW reports the backlog of the mth priority queue of the ith MCS, the unit is kbps, the IPGW generates, and the backlog is forwarded to the BM by the MCS as an input parameter;

C _ij (m) bandwidth allocated to the mth priority queue of the jth IP gateway for the ith MCS in kbps, MCS generated, and transmitted to the BM by the MCS as an input parameter;

G _i : the transmission bandwidth capability value of the ith MCS, if transmitted by the MCS to the BM as an input parameter, is only used to verify whether the input parameter is correct, and is not generally transmitted, and the calculation can be obtained:

the BM-adjusted IP gateway j prioritizes the bandwidth allocated from the ith MCS in kbps (algorithm output value).

Step S121: summarizing backlog and bandwidth allocation;

1. calculating the total backlog reported to the ith MCS by the jth IPGW, and summarizing all priorities, wherein the total backlog is as follows:

2. calculating the total bandwidth allocated to the jth IPGW by the ith MCS, and summarizing all priorities, wherein the total bandwidth is as follows:

3. calculating bandwidth capacity of ith MCS, collecting all IPGWs connected with the bandwidth capacity, and transmitting the bandwidth capacity and the MCS to G of BM _i Comparing to confirm that the bandwidth of the MCS allocation is correct, i.e. C _MCS，i ≤G _i If MCS does not send G _i This step may be skipped.

4. Calculating the total bandwidth allocated to the jth IPGW, and summarizing all MCSs connected with the total bandwidth;

step S122: determining an IPGW to be adjusted;

1. definition of temporary variables

Saving all bandwidths allocated to the jth IP gateway by the latest i-th MCS after adjustment, and taking the input bandwidth allocation value as an adjustment initial value;

2. finding out an IPGW set with the allocation bandwidth larger than the upper limit of the promised bandwidth;

3. finding out an IPGW set with the allocated bandwidth smaller than the upper limit of the promised bandwidth;

wherein ,/>

The upper limit of promised service for the jth three-layer gateway;

4. if Z ⁺ And the method is an empty set, namely, all bandwidth allocation is smaller than or equal to the upper limit of the promised bandwidth, and adjustment is not needed, and the return output parameters are as follows:

5. If Z ⁺ If the set is not empty, the next step is carried out for adjustment.

Step S123: reducing the allocation value of the IPGW with the allocation bandwidth larger than the upper limit of the promised bandwidth;

1. initializing each MCS bandwidth allocation reduction value to 0;

D _ij ＝0，j∈Z ⁺ ；

2. for j E Z ⁺ The following loop A step is performed in a traversing manner:

(1) The upper limit of the IPGW promised bandwidth is fairly distributed to the MCS connected with the IPGW according to the backlog queue lengths of different MCSs;

(satisfy H _kj =1, i.e. MCS has a connection with IPGW);

(2) Initializing a current bandwidth update value of the IP gateway j;

(satisfy H _kj ＝

1, i.e. MCS is connected to IPGW);

(3) When the conditions are

The following loop B step is performed in a loop:

for meeting H _kj K=1, the following loop C step is performed in a loop:

if it is

Meaning that the bandwidth of the MCS allocation may be reduced; />

If all k have been satisfied

Or->

Meeting, jumping out of the cycle B;

(4) Recording the reduction value of all MCS bandwidth allocation connected with the IP gateway j;

3. if Z ^- If the bandwidth is empty, the adjustment is finished, new adjusted bandwidth is redistributed according to the proportion of original priority distribution, other parameters are unchanged, and the output parameters are:

4. if Z ^- Instead of an empty set, a summary of each MCS bandwidth allocation reduction value is calculated, and the summary value can be adjusted to set Z ^- The process proceeds to step S124.

Step S124: increasing an allocation value of the IPGW that allocates bandwidth less than an upper bound of the committed bandwidth;

1. for meeting in MCS

The index numbers of (2) are ordered to obtain an ordered list set N _sort Wherein the number of IP gateways connected to the MCS is low.

2. For i epsilon N _sort The following loop A step is performed in a traversing manner:

(1) When (when)

The following loop B step is performed under the condition:

handle

The unallocated bandwidth is preferentially allocated to the IP gateway which does not satisfy the bandwidth application, i.e., backlog. For all j E Z ^- Loop execution the following loop step CThe steps are as follows:

if H _ij＝1 and

if j satisfies

Then from set Z ^- Removing j;

if all j E Z ^- All have satisfied

Then jump out of cycle C;

if set Z ^- If empty, cycle B is skipped.

If it is

Jumping out of cycle B.

(2)

There is a remainder in which unallocated bandwidth is reallocated to IP gateways that do not reach the upper bound of committed bandwidth for the IP gateway. For all j E Z ^- The loop performs the following loop D: />

If H _ij ＝1；

If j satisfies

Then from set Z ^- Removing j;

if set Z ^- If empty, cycle B is skipped.

If it is

Jumping out of cycle B.

3. And reallocating the new adjusted bandwidth according to the original priority allocation proportion, and returning to output parameters:

The invention has the following advantages:

1) The architecture for separating and deploying the three-layer gateway and the two-layer gateway supports one three-layer gateway to be connected with a plurality of two-layer gateways and supports one two-layer gateway to be connected with a plurality of three-layer gateways. The multi-to-multi interconnection architecture of the two-layer gateway supports the dynamic increase and decrease of the gateway, and the service gateway can be dynamically started and stopped according to the number of users and the load of the users; 2) The bandwidth manager collects the two-layer gateways reported by all the two-layer gateways and three-layer gateway data connected with the two-layer gateways through a heartbeat mechanism. The bandwidth manager dynamically learns the interconnection topology of all the gateways according to the bandwidth application of the three-layer gateway and the bandwidth capacity of the two-layer gateway; 3) The two-layer gateway distributes bandwidth for the three-layer gateway connected with the two-layer gateway, and the distribution principle is based on dynamic calculation of the bandwidth capacity of the two-layer gateway, the bandwidth application of the three-layer gateway based on priority, the service commitment upper limit of the three-layer gateway and the like; 4) And the three-layer gateway applies bandwidth to the two-layer gateways according to the priority in a unified way according to the two-layer gateways associated with the user registration, and supports one three-layer gateway to connect a plurality of the two-layer gateways. The user registration association determines the distribution proportion of the load on the two-layer gateway; 5) The forward total bandwidth is distributed among a plurality of two-layer gateways, and the situation that the bandwidth distribution values received by the three-layer gateways are unbalanced may exist. The dynamic adjustment is carried out through the bandwidth manager, so that the load balancing can be realized; 6) The bandwidth manager does not become a single point of failure for the network to operate. The bandwidth manager is offline, and the two-layer gateway can normally work for the bandwidth allocation of the three-layer gateways, but only the load balance cannot be realized, and the allocation value cannot be dynamically adjusted among the gateways. The bandwidth allocation of the two-layer gateway can ensure that the three-layer gateway obtains bandwidth and forward data service is normally transmitted.

Based on steps S11-S12, a forward bandwidth allocation method for a satellite network is provided, and the method provides service quality guarantee for user terminals in the network under the condition of limited bandwidth, and basically starts to flexibly allocate and dynamically adjust available resources of a wireless channel and the network under the conditions of uneven network data volume distribution, fluctuation and change of channel characteristics due to channel weakness and interference and the like, so that the utilization rate of inter-satellite links and inter-satellite links is improved to the greatest extent, network congestion is prevented, and the signaling load is kept as small as possible; a bandwidth manager is necessary on the forward link to perform overall planning and dynamic adjustment for bandwidth application and allocation of multiple two-layer gateways and multiple three-layer gateways. The load balancing of the link bandwidth in the three-layer gateway and the two-layer gateway can be realized through the dynamic adjustment of the bandwidth manager. And under the condition of the same physical link resources, the plurality of groups of users are maximally served.

Those skilled in the art will appreciate that the various modules in the apparatus may be adaptively split or combined. Such splitting or combining of specific modules does not cause the technical solution to deviate from the principle of the present invention, and therefore, the technical solution after splitting or combining falls within the protection scope of the present invention.

Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will fall within the scope of the present invention.

Claims

1. A forward bandwidth allocation method for a satellite network, comprising:

2. The forward bandwidth allocation method for a satellite network according to claim 1, wherein when one forward link gateway allocates bandwidth to a plurality of three-layer gateways, the forward link gateway calculates bandwidth allocation values according to a certain ratio of bandwidth capacities according to bandwidth requirements and priorities of each three-layer gateway, and allocates the bandwidth allocation values to the plurality of three-layer gateways, comprising:

calculating the forward link gateway to allocate bandwidth to the plurality of tri-layer gateways as: g _ij ＝η·R _ij ；

will Z _i <0 i additionSet I ^- Will Z _i >I of 0 joins set I ⁺ ；

For set I ⁺ Is: g _i ＝R _i ；

And calculates the sum of all the residual bandwidths as:

wherein ,N₊ For set I ⁺ The number of elements;

or->

wherein ,

for the ith priority in set I ^- The assigned weights of (a);

if Z _i Not less than 0, remove I ^- ；

when the total bandwidth of the plurality of three-layer gateway applications is greater than the total bandwidth provided by the forward link gateway and greater than the scaling-up value of the total bandwidth provided by the forward link gateway

for priority M, the calculation method is:

calculating the bandwidth value that priority M can allocate:

wherein ,u_hj The number of service users planned in the h service for the j-th IP gateway; r is (r) _h ^max Maximum transmission bandwidth for the service plan; g _h Satisfaction weight for the service plan;

3. the forward bandwidth allocation method for a satellite network according to claim 1, wherein when the plurality of forward link gateways allocate bandwidths for the plurality of tri-layer gateways, a bandwidth allocation value of each of the tri-layer gateways to which the forward link gateway is connected is calculated according to a bandwidth requirement and a priority order of each of the tri-layer gateways, the bandwidth manager dynamically adjusts the bandwidth allocation values of the plurality of forward link gateways according to a bandwidth capacity of each of the forward link gateways, obtains the adjusted bandwidth allocation values and returns the adjusted bandwidth allocation values to each of the forward link gateways, and each of the forward link gateways transmits the adjusted bandwidth allocation values to the plurality of tri-layer gateways, comprising:

4. A forward bandwidth allocation method for a satellite network according to claim 3, wherein the bandwidth manager acquires the connection topology of the forward link gateway and the three-layer gateway in a third period, comprising:

5. The forward bandwidth allocation method according to claim 4, wherein the bandwidth manager samples connection topologies of the forward link gateway and the tri-layer gateway in a third period, calculates a topology configuration value according to the sampled data, and updates the connection topologies of the forward link gateway and the tri-layer gateway if the topology configuration value is updated, comprising:

starting sampling, and resetting the sampling times of the last third period;

6. A forward bandwidth allocation method for a satellite network according to claim 3, wherein the forward link gateway transmits the calculated forward bandwidth allocation plan to a bandwidth manager according to a connection topology of the forward link gateway and the tri-layer gateway, the bandwidth manager dynamically adjusting between the plurality of forward link gateways by looking at a guaranteed service upper limit of the tri-layer gateway to obtain the adjusted forward bandwidth allocation plan, comprising:

determining three layers of gateways to be adjusted;

7. The forward bandwidth allocation method for a satellite network according to claim 6, wherein said aggregating data backlog and bandwidth allocation for each forward link gateway comprises:

8. the forward bandwidth allocation method for a satellite network according to claim 7, wherein said determining a three-tier gateway to adjust comprises:

The method comprises the following steps: />

Three layers with allocated bandwidth greater than the upper bound of committed bandwidthGateway set, is: z is Z ⁺ = { j, j satisfies

}；

The three-layer gateway set with the allocated bandwidth smaller than the upper limit of the promised bandwidth is as follows: z is Z ^- = { j, j satisfies

}, wherein->

The upper limit of promised service for the jth three-layer gateway;

if Z ⁺ And not the empty set, then the adjustment is made.

9. The forward bandwidth allocation method for a satellite network according to claim 8, wherein reducing the allocation value of the three-tier gateway for which the allocation bandwidth is greater than the upper bound of the committed bandwidth comprises:

initializing the current band of three-layer gateway jThe value of the wide update is a value,

(1) Obtaining all forward link gateway bandwidth allocation reduction values connected with three layers of gateways j when the conditions are met

Satisfying, the following loop steps are circularly executed;

for meeting H _kj K=1, the following loop steps are performed in a loop; if it is

The bandwidth allocated by the forward link gateway may be reduced as:

if all k have been satisfied

Or->

The method can be used for solving the problems that,

jumping out of the circulation step;

if Z ^- Is an empty set of the sets,then the adjustment is finished, the new adjusted bandwidth is redistributed according to the proportion of the original priority distribution, the other bandwidth is unchanged, and the parameters are output

10. the forward bandwidth allocation method for a satellite network according to claim 9, wherein increasing the allocation value of the three-layer gateway for which the allocated bandwidth is smaller than the upper limit of the committed bandwidth comprises:

for meeting in MCS