Load balancing routing method for multi-service QoS (Quality of Service) demand in satellite-ground convergence network

Abstract

The invention discloses a multi-service QoS (Quality of Service) demand-oriented load balancing routing method in a satellite-ground convergence network. The method comprises the following steps: a ground control center obtains a network topology; calculating an alternative path set; selecting a transmission path based on a genetic algorithm by taking maximization of a load balancing factor as a target; according to the method, a complete full-path route is calculated for each satellite node, and the service transmission requirement is better guaranteed; according to the technical scheme, an alternative path set mode is adopted, organic connection is established between a multi-service QoS guarantee strategy and a network load balancing strategy, selectable paths in the load balancing strategy are all schemes meeting service transmission requirements, and network loads are balanced while service QoS is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of satellite-ground fusion network communication, and in particular relates to a load balancing routing method oriented to multi-service QoS requirements in a satellite-ground fusion network. Background Art

[0002] Satellite Internet is an Internet based on satellite communications. It forms a large-scale network through a certain number of satellites, thus radiating across the globe and building a satellite communication system with real-time information processing. It is a new type of network that can provide broadband Internet access and other communication services to ground and air terminals. It has the characteristics of wide coverage, low latency, broadband, and low cost. Satellite Internet has gone through the stage of competing with the ground network and the stage of supplementing the ground communication network. Now, satellite Internet and the ground communication system complement each other and cooperate and develop in an integrated manner, entering the stage of satellite-ground integrated network.

[0003] The purpose of satellite routing technology in the satellite-ground integrated network system is to establish inter-satellite and satellite-ground transmission channels based on inter-satellite links and satellite-ground feeder links to ensure reliable data transmission in the network. It is a basic technology to ensure the interconnection between nodes of the satellite Internet. The choice of routing will affect the network's data transmission efficiency and the user's quality of experience.

[0004] In the current research on satellite network routing algorithms, most of the research is only focused on the inter-satellite routing process, but the inter-satellite routing algorithm only focuses on the path calculation process between the source satellite and the destination satellite node pair in the satellite network. In the actual network, there are some ground-based bearer services or data backhaul services that ultimately need to be transmitted to the ground node for analysis and calculation, and access to the core network or private network. In this process, the destination satellite is not unique, such as on-board payload data, extremely low-frequency scientific research data, government department private networks, multinational enterprise private networks and other business private networks in various fields. This increases the diversity of business transmission routing options. The transmission of such ground-based backhaul services requires the study of satellite-ground integrated network routing algorithms.

[0005] The routing algorithm of the satellite-ground integrated network faces the challenges of highly dynamic network topology, difficulty in ensuring QoS for multiple services, and uneven load distribution of gateways.

[0006] Highly dynamic network topology: The topology of the satellite network is predictable, regular, and periodic. However, in the satellite-ground fusion network, the satellite-ground link is highly complex due to problems such as the rotation of the earth, the distance between the satellite and the ground gateway, and rain attenuation interference. In addition, the inter-satellite links in the satellite-ground fusion network will be affected by solar eclipse interference, link failures, and other problems, resulting in link establishment and disconnection. This process occurs continuously with the movement of the satellite, and the topology of the satellite-ground fusion network changes extremely frequently.

[0007] It is difficult to guarantee QoS for multiple services: The services carried by the satellite-ground integrated network are characterized by multiple demands and a mixture of multiple services. Different types of services have different service quality requirements for latency, latency jitter, transmission bandwidth, etc. The path calculation process needs to map the service quality of the service into network requirements such as link connection duration and link bandwidth, and calculate the path that meets the requirements based on different requirements.

[0008] Concentrated distribution of gateways and unbalanced load: Global user network traffic is unevenly distributed, and ground gateways are unevenly distributed. In particular, my country's ground stations are relatively concentrated. Satellite business data is sent down to the ground through nearby gateways, which will cause a large amount of data to converge on satellite nodes above the ground gateways, easily causing congestion in the space segment links and some satellite-to-ground links, thereby affecting service quality.

[0009] In order to solve the load balancing problem of multi-service QoS requirements in satellite-ground integrated networks, most current studies choose to divide light-load and heavy-load areas according to the distribution addresses of gateways, and select different routing algorithms for the two areas for routing calculation. Such algorithms divide the service transmission path into two independent processes: from the source satellite to the edge node of the heavy-load area, and from the edge node of the heavy-load area to the ground. This method makes it difficult to ensure the QoS of multiple services. The satellite routing algorithm for collaboratively ensuring QoS and load balancing based on machine learning relies on on-board computing capabilities. However, due to the current limitations of satellite quality control, production cost control, and technological development, the on-board processing capabilities and available computing resources of satellites are extremely limited. The use of distributed routing algorithms such as machine learning is inconsistent with current satellite capabilities.

[0010] Based on the above problems existing in the prior art, the present invention provides a load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network. Summary of the invention

[0011] The present invention provides a load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network.

[0012] The present invention adopts the following technical solution:

[0013] A load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network, comprising:

[0014] Step 1: The ground control center has learned the network topology;

[0015] Step 2, calculate the set of candidate paths;

[0016] Step 3, selecting a transmission path based on a genetic algorithm with the goal of maximizing the load balancing factor;

[0017] Step 4: Output the transmission path of the current service of each satellite node.

[0018] Further, step 2 includes:

[0019] Step 2.1: The ground gateway station selects a satellite to establish a satellite-to-ground feeder link based on the satellite's trajectory and the number of its antennas. After the link is established, the ground gateway station reports the connection status to the ground control center, and the ground control center obtains the candidate landing satellite set V l ,

[0020] in Indicates the candidate landing satellite numbered k;

[0021] Step 2.2, configuring values ​​of weight factors corresponding to different business types according to different business types, wherein the sum of the values ​​of all weight factors is 1;

[0022] Step 2.3: Calculate the weight w of each candidate landing satellite according to the selected weight factor k , select a suitable landing satellite for the current business according to the landing satellite weight;

[0023] Step 2.4, delineate the candidate path rectangle with the source satellite and the landing satellite as the endpoints, and adopt the shortest path algorithm based on the number of hops. Each hop can select the same-orbit inter-satellite link and the different-orbit inter-satellite link, and form a candidate path set through different selection methods;

[0024] Step 2.5: For each satellite node, select the landing satellite suitable for the current business and calculate the corresponding set of candidate paths. i Its class c service alternative path set in is the kth alternative path for the cth category service of satellite i,

[0025] Furthermore, in step 2.3, the state weight of the landing satellite numbered k is expressed as:

[0026] ω k =λ1t1+λ2t2+...+λ k t k ...(1),

[0027] In formula (1), λ1, λ2, λ k is a weight factor, satisfying 0<λ<1 and λ1+λ2+...+λ k =1,t1,t2,t k As an influencing factor.

[0028] Furthermore, in step 2.3, the state weight of the landing satellite numbered k is expressed as:

[0029] w k =λ1hk +λ2l k +λ3B k ……(2),

[0030] In formula (2), λ1, λ2, λ3 are weight factors, satisfying λ1+λ2+λ3=1; h k is the normalized duration of the feeder link of the landing satellite numbered k; l k B is the normalized routing hop count between the landing satellite numbered k and the source satellite; k The normalized carried services of the landed satellite numbered k.

[0031] Furthermore, k satisfy:

[0032]

[0033] In formula (3), The estimated remaining connection time of the feeder link of the landing satellite numbered k; The maximum value of the estimated remaining connection time of the feeder link of n landed satellites.

[0034] Furthermore, h k satisfy:

[0035]

[0036] In formula (4), is the number of hops of the shortest path from the landing satellite numbered k to the source satellite; are respectively the maximum and minimum values ​​of the shortest path hops from the n landed satellites to the source satellite.

[0037] Furthermore, B k satisfy:

[0038]

[0039] In formula (5), The landing satellite numbered k has carried services; It is the maximum value of the services carried by n landed satellites.

[0040] Furthermore, in step 3, the regional load balancing factor is used as the population fitness parameter to perform genetic algorithm calculation.

[0041] in:

[0042] In formula (6), α LB is the regional load balancing factor; μ1, μ2 are weight factors; is the set of grounded satellites; M is the total number of grounded satellites in the region; N is the number of non-grounded satellites in the region; B k For the current satellite-borne business volume, the genetic algorithm outputs the transmission path for each satellite node for the current business. After the calculation for one type of business is completed, the ground control center obtains the transmission path corresponding to each satellite node for each business.

[0043] Furthermore, step 3 includes delineating a load balancing index area, wherein the area includes all candidate landing satellites; the area also includes other satellites that are not landing satellites, and is divided into regular rectangles or into one-hop areas or multi-hop areas of landing satellites according to the distribution of landing satellites.

[0044] The beneficial effects of the present invention are:

[0045] 1. The load balancing routing method for multi-service QoS requirements in the satellite-ground fusion network of the present invention calculates a complete full-path route for each satellite node to better guarantee service transmission requirements;

[0046] 2. The load balancing routing method for multi-service QoS requirements in the satellite-ground fusion network of the present invention adopts the method of alternative path set to establish an organic connection between the multi-service QoS guarantee strategy and the network load balancing strategy. The optional paths in the load balancing strategy are all solutions that meet the service transmission requirements, which can ensure the service QoS while balancing the network load;

[0047] 3. The load balancing routing method for multi-service QoS requirements in the satellite-ground integrated network of the present invention adopts a centralized routing strategy. The ground control center performs centralized routing calculations and spreads the routing table to each node of the network through the landing satellite, which reduces the difficulty of on-board processing and can better balance the network load. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a scenario of a load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to an embodiment of the present invention;

[0049] Figure 2 It is a flow chart of a load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network in an embodiment of the present invention;

[0050] Figure 3 Schematic diagram of packet loss rate comparison at different data transmission rates in an embodiment of the present invention;

[0051] Figure 4 is a schematic diagram of throughput comparison at different data transmission rates in an embodiment of the present invention;

[0052] Figure 5A schematic diagram of comparing load balancing factors of different algorithms in an embodiment of the present invention;

[0053] Figure 6 Schematic diagram of average delay comparison under different data transmission rates in an embodiment of the present invention;

[0054] Figure 7 The figure is a schematic diagram of comparing service packet loss rates at different data transmission rates in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to more clearly understand the above-mentioned purposes, features and advantages of the present invention, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0056] Example

[0057] In the load balancing routing method for multi-service QoS requirements in the satellite-ground fusion network described in the embodiment of the present application, the ground control center calculates the routing table of the landing bearer service or data backhaul service for all satellite nodes, and notes it to the satellite node. The satellite node forwards the data packet according to the received routing table. When the network topology changes, the network node transmits the change information to the ground control center. The ground control center performs routing calculation based on the new topology and notes the new routing table for each satellite node. Figure 1 A scene diagram of the method described in the embodiment of the present application;

[0058] The method is a load balancing routing method for multi-service QoS requirements. The method needs to distinguish service types to perform hierarchical routing calculations, give priority to high-priority service routing calculations, and update satellite node states according to the calculation results. The updated node weights are used to perform next-priority service routing calculations until all service routing calculations are completed. According to service requirements, services are divided into delay-sensitive, delay-jitter-sensitive, bandwidth-sensitive, best-effort transmission, etc., wherein delay-sensitive services have high real-time requirements, and such services need to minimize network delays; delay-jitter-sensitive services need to continuously transmit multiple data packets within a period of time, and such services require small delay changes during transmission; bandwidth-sensitive service data packets are large, and such service transmission requires that the transmission path meet certain bandwidth requirements; best-effort delivery services, such services have no requirements for the network.

[0059] For example, for the Walker constellation, its constellation configuration satisfies N c / M c , where N c Expressed as the total number of satellite network orbits, M cThe total number of satellites in each orbit is N = N. c ×M c , the satellite network can be mapped into an undirected graph, represented as G(V,E), where V represents the set of satellite nodes of a large-scale satellite constellation V = {v i |1≤i≤N}, E is the set of links between satellites E={e ij |1≤i≤N,1≤j≤N,i≠j}, where:

[0060]

[0061] The landing satellite is a satellite that establishes a feeder link connection with the ground gateway station. The set of all landing satellites is the candidate landing satellite set. in represents the candidate landing satellite numbered k. All the landing satellites connected to a single gateway are called a landing satellite cluster. For each satellite landing Its weight w k Different influencing factors and influencing factors can be set according to different business types, such as:

[0062] ω k =λ1t1+λ2t2+...+λ k t k ……(2),

[0063] Among them, w k is the state weight of the landing satellite numbered k; λ1,λ2,λ k The weight factor must satisfy 0<λ<1 and λ1+λ2+...+λ k =1,t1,t2,t k As influencing factors, in actual calculation, they are set as required to include delay, delay jitter, remaining bandwidth, average load of satellite cluster, etc. according to business needs;

[0064] For example, in the embodiment of the present application, the weight of the landing satellite is represented by three influencing factors: link duration, average number of path hops, and carried traffic. For example:

[0065] w k =λ1h k +λ2l k +λ3B k ……(3),

[0066] Among them, w k is the state weight of the landing satellite numbered k; λ1, λ2, λ3 are weight factors, which must satisfy λ1+λ2+λ3=1; h kis the normalized duration of the grounded satellite feeder link numbered k, and its calculation satisfies equation (5); l k is the normalized routing hop count between the landing satellite numbered k and the source satellite, and its calculation satisfies formula (4); B k is the normalized carried traffic of the landed satellite numbered k, and its calculation satisfies formula (6).

[0067]

[0068]

[0069]

[0070] in, The estimated remaining connection time of the feeder link of the landing satellite numbered k; is the maximum value of the estimated remaining connection time of the feeder links of n landed satellites; is the number of hops of the shortest path from the landing satellite numbered k to the source satellite; are the maximum and minimum values ​​of the shortest path hops from the n landing satellites to the source satellite, The landing satellite numbered k has carried services; is the maximum value of services carried by n landed satellites;

[0071] like Figure 2 As shown, the load balancing routing method for multi-service QoS requirements in the satellite-ground fusion network includes:

[0072] Step 1: The ground control center has learned the network topology;

[0073] Step 2, calculate the set of candidate paths;

[0074] Step 3: Select the load balancing path;

[0075] Step 4, output the transmission path of the current service of each satellite node;

[0076] As a specific implementation, the calculation of the candidate path set for satellite differentiated services in step 2 includes:

[0077] Step 2.1: The ground gateway station selects a satellite to establish a satellite-to-ground feeder link based on the satellite's trajectory and the number of its antennas. After the link is established, the gateway station reports the connection status to the ground control center, and the ground control center obtains the candidate landing satellite set V l ;

[0078] Step 2.2: Differentiate the different business types for λ1,λ2,λ in formula (2). k The weight factor configuration value should reflect the different needs of different businesses.

[0079] For example, for delay-sensitive services, the number of hops on the path should be as small as possible, and the weight factor value should satisfy λ1>λ2>λ3. For delay-sensitive services, the duration of the grounded satellite feeder link connection should be as long as possible, and the weight factor value should satisfy λ2>λ1>λ3. For bandwidth-sensitive services, the amount of traffic carried by the grounded satellite should be as small as possible, and the weight factor value should satisfy λ3>λ1>λ2.

[0080] Step 2.3: Calculate the weight w of each candidate landing satellite according to the selected weight factor k , select a suitable landing satellite for the current business according to the landing satellite weight;

[0081] Step 2.4, delineate the alternative path rectangle with the source satellite and the landing satellite as the endpoints, and adopt the shortest path algorithm based on the number of hops. Each hop can choose the same-orbit inter-satellite link and the different-orbit inter-satellite link, and different choices form an alternative path set;

[0082] Step 2.5: According to the above order, select the landing satellite suitable for the current business for each satellite node, and calculate the corresponding set of candidate paths. i Its class c service alternative path set in is the kth alternative path for the cth category service of satellite i,

[0083] As a specific implementation, step 3 of selecting a transmission path from a load balancing perspective includes:

[0084] Step 3.1, delineate the load balancing index area, and the delineation principle is as follows: (1) the area includes all candidate landing satellites; (2) the area also includes other satellites that are not landing satellites. There is no restriction on the area delineation. For example, it can be divided into regular rectangles or into one-hop areas or multi-hop areas of landing satellites according to the distribution of landing satellites;

[0085] Step 3.2, the calculated candidate path set is used as the gene, and each chromosome is composed of an alternative path of all satellites in the constellation, where Indicates satellite v i The transmission path of the cth service;

[0086] Step 3.3 uses the regional load balancing factor as the population fitness parameter to perform genetic algorithm calculation, which is defined as follows:

[0087]

[0088] Among them, α LB is the regional load balancing factor; μ1, μ2 are weight factors; is the set of grounded satellites; M is the total number of grounded satellites in the region; N is the number of non-grounded satellites in the region; B k The current satellite traffic volume;

[0089] It should be noted that the genetic algorithm outputs the transmission path for each satellite node for the current service. After the calculation for a type of service is completed, the ground control center can obtain the transmission path corresponding to each satellite node for each service. After the calculation is completed, the ground control center will add the routing table to the satellite node, and the satellite node will distinguish the service type according to the routing table to forward data packets.

[0090] In order to verify the superior technical effect of the embodiment of the present application, in the embodiment of the present application, a satellite-ground fusion communication network is deployed, wherein the LEO satellite network includes a total of 108 satellites, which are placed in a 1,400-kilometer LEO orbit, using 6 orbital planes, 18 satellites are deployed on each orbital plane, and three gateway stations are placed on the ground, each of which can establish a feeder link with up to three LEO satellites at the same time;

[0091] like Figure 3-7 As shown, in the embodiment of the present application, the satellite link intersatellite bandwidth capacity, the satellite-to-ground feeder link bandwidth capacity, and the single-line generated service rate are set to the same unit, the satellite link intersatellite bandwidth capacity is set to 5 units, and the satellite-to-ground feeder link bandwidth capacity is set to 10 units. By changing the data packet generation rate of each satellite, different degrees of congestion in the LEO satellite network are simulated. The reliability of the method described in the embodiment of the present application is evaluated by the average end-to-end delay, packet loss rate, throughput, average routing hops, and load balancing factor. The buffer size of each link is set to 10 packets;

[0092] In summary, the method described in the embodiment of the present application effectively reduces the packet loss rate of the satellite-ground fusion network, improves the network throughput and the load balancing index of the gateway station overhead area. As the packet generation rate of the source satellite gradually increases, the network link status changes rapidly and the congestion level increases, the effect of the present invention becomes more significant.

[0093] The present invention is not limited by the above embodiments. The above embodiments and descriptions are only for explaining the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention may be subject to various changes and improvements, which are within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the appended claims.

Claims

1. A load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network, characterized in that: include: Step 1: The ground control center has learned the network topology; Step 2, calculate the set of alternative paths; Step 3, selecting a transmission path based on a genetic algorithm with the goal of maximizing the load balancing factor; Step 4: Output the transmission path of the current service of each satellite node.

2. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 1 is characterized in that: Step 2 includes: Step 2.1: The ground gateway station selects a satellite to establish a satellite-to-ground feeder link based on the satellite's trajectory and the number of its antennas. After the link is established, the ground gateway station reports the connection status to the ground control center, and the ground control center obtains the candidate landing satellite set V l , in Indicates the candidate landing satellite numbered k; Step 2.2, configuring values ​​of weight factors corresponding to different business types according to different business types, wherein the sum of the values ​​of all weight factors is 1; Step 2.3: Calculate the weight w of each candidate landing satellite according to the selected weight factor k , select a suitable landing satellite for the current business according to the landing satellite weight; Step 2.4, delineate the candidate path rectangle with the source satellite and the landing satellite as the endpoints, adopt the shortest path algorithm based on the number of hops, and select the same-orbit inter-satellite link and the different-orbit inter-satellite link for each hop; Step 2.5: For each satellite node, select the landing satellite suitable for the current business and calculate the corresponding set of candidate paths. i Its class c service alternative path set in is the kth alternative path for the cth category service of satellite i, 3. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 2 is characterized in that: In step 2.3, the state weight of the landing satellite numbered k is expressed as: oh k =λ1t1+λ2t2+...+λ k t k ……(1), In formula (1), λ1, λ2, λ k The weight factor must satisfy 0<λ<1 and λ1+λ2+...+λ k =1,t1,t2,t k As an influencing factor.

4. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 2 or 3, characterized in that: In step 2.3, the state weight of the landing satellite numbered k is expressed as: w k =λ1h k +λ2l k +λ3B k ……(2), In formula (2), λ1, λ2, λ3 are weight factors, satisfying λ1+λ2+λ3=1; h k is the normalized duration of the feeder link of the landing satellite numbered k; l k B is the normalized routing hop count between the landing satellite numbered k and the source satellite; k The normalized carried services of the landed satellite numbered k.

5. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 4 is characterized in that: l k satisfy: In formula (3), The estimated remaining connection time of the feeder link of the landing satellite numbered k; The maximum value of the estimated remaining connection time of the feeder link of n landed satellites.

6. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 4, characterized in that: h k satisfy: In formula (4), is the number of hops of the shortest path from the landing satellite numbered k to the source satellite; are respectively the maximum and minimum values ​​of the shortest path hops from the n landed satellites to the source satellite.

7. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 4, characterized in that: B k satisfy: In formula (5), The landing satellite numbered k has carried services; It is the maximum value of the services carried by n landed satellites.

8. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 1, characterized in that: In step 3, the regional load balancing factor is used as the population fitness parameter to perform genetic algorithm calculation, where: In formula (6), α LB is the regional load balancing factor; μ1, μ2 are weight factors; is the set of grounded satellites; M is the total number of grounded satellites in the region; N is the number of non-grounded satellites in the region; B k For the current satellite-borne business volume, the genetic algorithm outputs the transmission path for each satellite node for the current business. After the calculation for one type of business is completed, the ground control center obtains the transmission path corresponding to each satellite node for each business.

9. The load balancing routing method for multi-service QoS requirements in a satellite-ground fusion network according to claim 1, characterized in that: Step 3 includes defining a load balancing index area, wherein the area includes all candidate landing satellites; the area also includes other satellites that are not landing satellites, and is divided into regular rectangles or into one-hop areas or multi-hop areas of landing satellites according to the distribution of landing satellites.

Images