CN110166296A - A method of information centre's network architecture towards manned deep space exploration - Google Patents

Abstract

An information center network architecture method for manned deep space exploration. According to the space area and node communication range, according to the connection plan, the aggregation node in each domain can be selected as the domain server of the domain. It is responsible for internal information management in the domain and external Responsible for building inter-domain transmission links to realize data cross-domain transmission; through this domain server, each domain forms a top-level domain topology network for networking communication, and data and information are synchronized and transmitted between these top-level domains; each The data transmission and networking management in the top-level domain are realized through the concept of hierarchical domains; according to the longitude, latitude and altitude geographic information spatial position, the top-level domain or domain space is subdivided into multiple second-level domains, and each second-level domain also has a domain Servers, all domain servers are interconnected to form the backbone transmission network of the entire network and tree-like network topology management, which solves the problems of flexible networking and delay tolerance between domains for manned deep space exploration in the future.

Description

A Method of Information Center Network Architecture for Manned Deep Space Exploration

technical field

The invention relates to the field of space information network, in particular to an information center network architecture method for manned deep space exploration.

Background technique

With the development of aerospace technology, human exploration of deep space has never stopped. my country's deep space exploration activities started in 2004. After more than ten years of rapid development, it started from the moon and moved towards a new step of comprehensive development. It is expected to complete the "circling", "falling" and "returning" lunar exploration projects before 2020 In three steps, the Mars exploration mission was launched in the same year, opening a new chapter in deep space exploration.

The current deep space network structure presents the characteristics of few nodes and single communication services. In this case, delay tolerant networks (Delay Tolerant Networks, DTN) can better meet communication needs and solve long-delay and frequent interruption environments. The next transmission problem. However, with the promotion of the deep space exploration programs of the world's major aerospace powers, the future deep space network structure will also undergo great changes. First, the number and types of nodes will increase significantly, such as mobile habitats, communication base stations, unmanned exploration vehicles, relay satellites, and astronauts; exploration tasks still include machine telemetry, but these operations can be performed by astronauts and their vehicles. Execution; frequent data interactions between nodes will greatly increase the amount of data transmission; communication services will be more diversified, in addition to traditional data push services, it will also include multimedia services such as email, voice, and video. There will also be higher requirements in terms of priority and priority. The communication service not only includes the push of periodic detection data, but each node also has a demand for active acquisition of specific data. In this case, DTN's single push mode and lack of an efficient content distribution strategy will seriously hinder its application in future manned deep space exploration scenarios. Therefore, it is extremely important to design an appropriate network protocol to meet the current and future needs of deep space exploration.

Information Centric Networking (ICN) abandons the traditional host-centric thinking and focuses the network on data acquisition, that is, when users request data, they do not need to know the source address of the content, but only need to send the required information Just send it to the network. Compared with DTN, ICN has more advantages in networking and content distribution. First of all, ICN provides more diversified transmission methods to meet the business needs of different communication scenarios in the future, such as periodic detection data push, specific content active acquisition and command data push, etc.; in addition, the cache mechanism allows intermediate nodes to back up popular data , so that when other nodes make repeated requests for this data, the intermediate node can reply directly, thereby reducing the transmission path and reducing the round-trip delay; the multi-source transmission mechanism can enable multiple data sources to reply to the same request, thereby improving Data transmission reliability, but also enables different data sources to transmit different content blocks of the same data, thereby reducing the load pressure on content providers and improving transmission efficiency. In addition, ICN also has certain support for security and mobility.

Therefore, although DTN can better solve the transmission problem in the current scene, with the complexity of the future deep space network structure and communication requirements, the present invention believes that the advantages of the information center network ICN in efficient networking and content distribution will make it It is more suitable for future manned deep space exploration scenarios.

Contents of the invention

technical problem:

The present invention aims to provide a method for the network architecture of manned deep space exploration in the future, to solve the characteristic problems of network node types diversification, network scale, and network heterogeneity in the manned exploration network, and the resulting problems Networking management, domain division and other networking issues; solve the diversification of business and service requirements in the future manned detection network, as well as the service discovery and service management problems brought about by diversification; solve the long delay in deep space, Problems such as low data transmission efficiency caused by high bit errors, interruptions, and asymmetrical link characteristics make this network architecture design method applicable to complex manned deep space exploration scenarios in the future.

Technical solutions:

In order to achieve the above purpose, the present invention proposes a method based on the network architecture of the information center network, which utilizes the subscription publishing mechanism of the information center network, and combines technical ideas such as information multi-source transmission, networking, and network topology to solve the problem of manned deep-sea problems in the future. Multi-service/service management of air detection, network management, flexible inter-domain networking, tolerance of delay and other issues.

Combining with Fig. 1, the present invention proposes a conception of a future manned deep space scene, and describes in detail the whole network architecture, topological relationship, and domain division in the scene. The entire network architecture can be divided into several top-level domains, such as the Earth domain, Mars domain, deep space domain, and other planetary domains.

Including the following steps:

Step 1. Construction of the control plane. This step divides the top-level domain into different domain spaces according to certain rules (such as coordinates, latitude and longitude, etc.) according to the spatial area and the communication range of nodes. All nodes in the network can belong to different domain spaces; According to the connection plan, the aggregation node in each domain space is selected as the domain server of the domain, which is responsible for information management within the domain internally, and is responsible for building inter-domain transmission links externally to realize data cross-domain transmission. At the same time, the domain servers are connected to each other to form an upper-layer control topology network, and various types of information are transmitted and synchronized in each domain space;

Step 2, service discovery, this step includes service discovery, service management, publication, subscription, etc.; each node in the domain space registers on the domain server, and publishes or subscribes to a certain type of service or data;

Step 3, the data transmission step, data transmission is performed between the subscriber and the domain server or the publisher, wherein there may be multiple publishers with the same data, that is, multi-source transmission of data.

Step 4. Service cancellation. In this step, the user cancels publishing or subscribing to a certain service or data, and logs out on the server.

The implementation process of steps 1-4 is implemented in the following ways:

1) Construction of the control plane, including booting and initialization of domain servers, and interconnection and communication between domain servers.

2) Service discovery, including the following steps: user registration; server's reply (ACK) to user registration request; user's release data/service request; server's reply (ACK) to release request; user subscription data/service request; Reply (ACK) to a subscription request.

3) Data transmission, including data transmission between subscribers and publishers, subscribers and servers. There may be multiple publishers with the same data, that is, multi-source transmission of data.

4) Service cancellation, including cancellation of publication, cancellation of subscription, and user's own logout. It is divided into the following steps:

A. Unpublishing: the user cancels the publishing of the data; the server responds (ACK) to the user canceling the publishing of the data;

B. Unsubscribe: the user cancels the subscription to the data; the server responds (ACK) to the user's unsubscribed data;

C. User self-logout: the user logs out of the network; the server replies (ACK) to the logout request.

There is no sequence between the above-mentioned subscription and publication.

During the node registration process in the service discovery step, the user's registration request information should cover all information about the user. Including information type (Message_TYPE): used to describe the type of information; user ID (USER_ID): the unique identifier of the user; data type (DATA_TYPE): indicating the type of information subscribed or published by the user, including static data, dynamic data, instruction data, etc.; User domain (scope): indicates the geographic location of the user; heartbeat time (KEEP_ALIVE): the heartbeat time negotiated between the user and the server; service interface reserved bit (Reserved): used to provide service/business interface elements.

During the information publishing process, the static metadata released should include the publisher ID, the name of the file, the source address path, the size of the file block, the hash value of the file block and other information.

During the information publishing process of the service discovery step, there are differences in the format of the metadata of the instruction data, static data and dynamic data. The metadata of the command data includes the type of command required (Command) and the source of the specified command (Command-source), that is, only commands from the specified source will be accepted, which improves security. Static data is uniquely identified by the file name (name). The file name can be named according to the publisher's domain and specific geographical location. Dynamic data is uniquely identified by timestamp (time stamp) + file name (name). The timestamp is used to identify the dynamic data, and the file name can be named according to the publisher's domain and specific geographical location.

During the information publishing process of the service discovery step, the publisher can allow the server to cache popular data, which is helpful for multi-source distribution of data. For the determination of popular data, the server will record each subscription information, count the content with a large number of subscriptions within a period of time, and cache the content locally. The potential legal issue is the owner of the data content cached by the server. Since the publisher informs the server that the data can be cached, the ownership should still belong to the original publisher.

In the information subscription process of the service discovery step, the subscription should be divided into single subscription and continuous subscription. Most of the single subscription occurs in the subscription of static data, and the continuous subscription may occur in the subscription of instructions and dynamic data, and the continuous subscription will not be deleted by the server. This option should be included in the subscription infoframe.

In the data transmission process, a multi-source data transmission scheme. Multi-source transmission divides files into data blocks of uniform size. Each publisher only transmits a part of data blocks to subscribers. After receiving all publishers' data blocks, subscribers can reassemble them into files. Subscribers need to communicate with all Publishers keep information synchronized, notifying existing blocks and missing blocks, so as to determine the data blocks sent by each publisher. In addition, after the subscriber has completely received the data, it can notify the server as the publisher of the data through a publishing request, and realize the source expansion of multi-source transmission through this logic.

In the above-mentioned service cancellation step, both the subscriber and the publisher need to log out when shutting down or leaving the current domain. When the publisher logs out, the server needs to be notified whether to delete the cached data of the server. In addition, if the user fails to actively send logout information due to unexpected reasons such as sudden crash or sudden movement, the heartbeat mechanism of the domain server will detect the user's disconnection, and then actively logout the user.

According to the space area and node communication range, it can be divided into different domains according to certain rules (such as coordinates, latitude and longitude, etc.), and according to the connection plan, the aggregation node in each domain can be selected as the domain server of the domain (the top-level domain may have several internally responsible for intra-domain information management, and externally responsible for building inter-domain transmission links to realize data cross-domain transmission. Through this domain server, each domain can form a top-level domain topology network for networking communication, and data and information can be synchronized and transmitted between these top-level domains. Data transmission and networking management in each top-level domain are realized through the concept of hierarchical domains;

Top-level domains or domain space can be subdivided into multiple second-level domains according to spatial location (such as geographic information such as latitude and longitude, altitude, etc.), each second-level domain also has a domain server, and all second-level domains belonging to the same top-level domain can be combined The network constitutes a two-level domain topology network, and data and information can be synchronized and transmitted between these two-level domains. And each second-level domain can also subdivide the third-level domain according to the situation. By analogy, a hierarchical tree-like network topology is finally formed, and all domain servers are interconnected, which can not only be used as a control network, but also constitute the backbone transmission network of the entire network. This tree-like network topology management not only improves the efficiency of information retrieval and information management, but also improves the security of information. Any information outside the domain needs to be detected by multi-layer domain servers.

The communication between these domains and between domains and within domains forms a control plane, on which the networking mode of the network is formed, and the networking structure of the entire network is more hierarchical.

Combined with the concept of domain, the overall network architecture can be divided into two functional planes in terms of functional division: control plane and data plane.

Control plane: It is formed by connecting servers in various domains to form an information control network. Domain servers are responsible for data information management and network topology management. User registration, publication, subscription, and logout request information are transmitted and synchronized on the control plane.

Data plane: It is composed of various communication nodes to realize data transmission and loading services after data and service discovery, including multi-source transmission and routing selection, etc. Each node acts as a subscriber and publisher. Each node is located in its own domain and will communicate with the domain server of the domain through one or more hops to send publishing information or subscription request information.

Specifically, referring to FIG. 1 , the node types of the control plane are mainly servers of various domains. The link element types in the control plane mainly include the following types:

1) Inter-domain relay link. It is mainly an interstellar relay satellite network, which can be a laser link and an RF link.

2) Inter-domain direct link. Satellites and ground stations are directly connected to the network, which can be laser links and RF links.

3) Intra-domain communication links. It is mainly the communication link between each domain server and the data transmission link between the domain server and the data center.

4) Cross link. Communication link between geostationary and low-orbit satellites.

5) Adjacent links. Communication links between low-orbit satellites or geostationary satellites and ground stations, communication base stations, probe vehicles, and astronauts.

6) W-LAN. It is mainly the short-distance communication links between astronauts, users, probe vehicles, communication base stations, and servers (not limited to W-LAN, but also other communication methods such as Bluetooth).

The node types of the data plane mainly include the following types:

1) Probe car. Probes on the planet's surface. Detect planetary temperature, humidity, atmospheric composition and other information. Can act as both a subscriber and a publisher.

2) Space probe satellites. Detector satellites wandering in the universe, detecting cosmic rays, temperature and other information.

Can act as both a subscriber and a publisher.

3) Astronauts. Can act as both a subscriber and a publisher.

4) Data center. store data. Take on the role of Publisher.

5) Relay satellites, including synchronous relay satellites, low-orbit relay satellites, etc. Play the role of the backbone transmission network for communication.

6) Satellite ground station. Acts as an access unit for communications.

7) Signal station. Network signal relay base stations on Earth and other planets.

8) Polar platform. Base platform for rover and astronauts.

Among them, node types 1-4 are all user types, and can act as subscribers or publishers; node types 5-8 are all domain server types, and they are responsible for storing and synchronizing information such as subscription and publication within the domain, and as domains outside the domain. The backbone node of the transmission link between them.

There are many types of data in the data plane, which can be mainly divided into the following types according to the type:

1) Command data. Instructions sent by the control center of the earth domain to the probes and astronauts in the various planetary domains.

2) Static data. Data that will not change, such as text data, images, offline videos, etc.

3) Dynamic data. Data that changes or updates over time. For example, voice calls and data detected by detectors.

Finally, considering the diversification of businesses and services that may appear in the future manned deep space exploration scenarios, such as email services, and even real-time voice or video services. In order to ensure the friendliness of the architecture for business/service expansion, it is necessary to reserve bits and interfaces for business and services when transmitting information. It should be pointed out that the underlying data type of business/service still belongs to one of static data, dynamic data and instruction data, so the transmission method will not change. The specific implementation will be further described below.

The present invention utilizes the basic subscription and publishing mechanism of the information center network, and combines technical ideas such as information multi-source transmission, networking, and network topology to design a complete set of network architecture solutions for manned deep space detection scenarios, and The multiple process steps and data frame format are described in detail. The architecture design can be divided into control plane and data plane. According to the space area and node communication range, it can be divided into different domains according to certain rules (such as coordinates, latitude and longitude, etc.). According to the connection plan, the aggregation node in each domain can be selected as the domain server of the domain, and is responsible for information management within the domain. , externally responsible for building inter-domain transmission links to realize data cross-domain transmission. The control plane is connected by servers in various domains to form an information control network. User registration, publishing, subscription and logout request information are transmitted and synchronized on the control plane. The data plane is composed of various communication nodes to realize file data transmission, including multi-source transmission and routing selection, etc., and each node acts as a subscriber and a publisher. Each node is located in its own domain and will communicate with the domain server of the domain through one or more hops to send publishing information or subscription request information. The present invention provides a solution to problems such as multi-service/service management, network management, inter-domain flexible networking, and link characteristics (channel bandwidth, bit error rate, propagation delay, etc.) of manned deep space exploration in the future.

Beneficial effects: the present invention designs a network architecture based on the Information Center Network (ICN), which can meet the needs of future manned deep space exploration for multi-service/service management, network management, inter-domain flexible networking, and tolerance for delays. Time and other functional requirements; for different data types: dynamic data, static data and instruction data adopt a relatively consistent publish and subscribe process, which is more unified in form; combined with the principle of multi-source transmission, it can effectively reduce the cost of a single link in the deep space link. Traffic load, improve data transmission efficiency.

Description of drawings

Fig. 1 is an architecture diagram of an information center network oriented to manned deep space exploration of the present invention.

Fig. 2 is a flow chart of network node function operation in the present invention.

Fig. 3 is a schematic diagram of node publishing and subscribing data in the present invention.

Fig. 4 is a schematic diagram of multi-source data transmission by nodes in the present invention.

Fig. 5 is a data frame format diagram of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

As shown in Figures 1 to 5, the present invention proposes an information center network architecture design method suitable for manned deep space exploration networks, that is, an information center network architecture design method for manned deep space exploration, which is divided into control planes and the data plane.

Control plane: It is formed by connecting servers in various domains to form an information control network. Domain servers are responsible for data information management and network topology management. User registration, publication, subscription, and logout request information are transmitted and synchronized on the control plane.

Data plane: It is composed of various communication nodes to realize file data transmission, including multi-source transmission and routing selection, etc., and each node acts as a subscriber and publisher. Each node has its own domain and will communicate with the domain server of the domain to send publishing information or subscription request information.

For domain division, a hierarchical structure division method is adopted. The topmost domain is the broadest range, such as the deep space domain, Mars domain, etc. The lower subdomains of each top-level domain are more subdivided geographical areas, and each domain can be further subdivided. The finer the domain division, the higher the efficiency of information management and information retrieval. Each domain will have a domain server, which is used to manage the publish and subscribe information in the domain. The servers in the subdomains belonging to the same parent domain can communicate with each other to keep the synchronization and update of the publish and subscribe information. Servers between subdomains distributed between two parent domains cannot communicate directly, and need to communicate through servers in the upper domain, that is, the parent domain. For published data, the domain + data name is used as the unique identifier of the data. In this domain, the parent domain name of the upper layer can be ignored. When the domain servers of the same layer synchronize the published content in this domain, the domain name of this domain needs to be added. mark. This tree-like layered domain division method not only improves the efficiency of information retrieval, but also improves the security of information. Any information outside the domain needs to be detected by multi-layer domain servers.

As shown in Figure 2, its main operation steps and configuration of transmission information are as follows:

Step 1. Build a domain networking architecture. This step is divided into different domains according to certain rules (such as coordinates, latitude and longitude, etc.) according to the space area and node communication range, and all nodes in the network can belong to different domain spaces; and select the convergence in each domain according to the connection plan As the domain server of the domain, the node is responsible for internal information management in the domain, and externally responsible for building inter-domain transmission links to realize data cross-domain transmission.

Step 2. Each node in the deep space network registers on the domain server to construct the architecture and topology connection of the entire network. The domain servers are connected to each other to form an upper-level control plane, which mainly includes service discovery, service management, publishing, Subscriptions and more. In both cases, the user is registered on the domain server;

(1) Register when the user just starts up or goes online;

(2) When a user moves to a new domain, he will log out in the old domain and register in the new domain. This kind of domain switching is relatively common.

When registering, the server uniquely identifies the registered user as the user ID. The registration information should include the domain information of the user, which is beneficial to the management of information, such as improving the efficiency of information retrieval; the registered user is connected to the domain server by one or more hops from other nodes in the domain, and the server and the registered user need to maintain Heartbeat contact to ensure that registered users are always connected to the server.

Step 3. The server responds (ACK) to the user registration request. The server needs to inform the user whether the registration is successful, and if the registration fails, it needs to inform the reason; if the request reply is not received within the set time after publishing, it is considered that the publishing has failed and needs to be republished.

Step 4. For the publisher, the generated data forms a publishing request and sends it to the server. The server performs a series of verifications. After passing the verification, the publishing request is recorded in the publishing list, and the information is synchronized to the upper domain server at the same time.

The published data types are mainly divided into static data, dynamic data and instructions. Among them, the command release is quite special, and the command release is the demand for the command issued by the receiver of the command. The other two types of data are statements issued by data generators (producers) for existing data. The published information statement should contain the metadata of the published data, which is used to uniquely identify the data. In addition, the server does not unconditionally accept all publishing requests for publishing data. In some cases, publishing is not allowed, such as domain mismatch, insufficient server storage space, reaching the upper limit of server publishing data, and mismatching timestamps of dynamic files. Instruction issuance has a higher priority than other data. For the publisher, the instruction issuance request is transmitted first. For the server, when the upper limit of the server’s published data is reached, the instruction issuance request can replace the existing published ordinary data. For the release of dynamic data, the method of adding time stamps is adopted, and the dynamic data is uniquely identified in the form of data name + time stamp, so as to distinguish dynamic data with the same name but different versions.

For the published information, efficient provision of popular data (or common information) should also be considered. In fact, the server will record each subscription information, and count the content with a large number of subscriptions within a period of time, and cache the content locally, so that the user can download it directly from the server when subscribing next time, without having to communicate with the publisher to get in touch. The ownership of the cached data still belongs to the publisher, and the publisher negotiates with the server in the publishing step to inform the server whether to allow caching of the published content and whether the publisher can actively request to delete the server cache. When the publisher deletes content locally, it notifies the server of the corresponding delete cache.

Step 5. The server replies (ACK) to the published information. The server needs to inform the publisher whether the publication is successful, and if the publication fails, it needs to inform the reason, such as timestamp mismatch, domain mismatch, insufficient storage space, etc.

Step 6. The user may cancel/delete the release of the data. A user may cancel/delete a publication of data, using the same metadata as the publication request to ensure that the corresponding publication is cancelled.

Step 7, the server responds (ACK) to the user's cancellation of publishing data. The server needs to inform the user whether unpublishing is successful.

Step 8. For the subscriber, the generated request forms the subscription information and sends it to the server. The server performs a series of verifications. After passing the verification, it will check whether there is a corresponding publication record in the publication list. If it exists, the server will notify the publisher and subscription Otherwise, the two start data transmission; if there is no publication record, the subscription request will be recorded in the subscription list and passed to the domain server at the upper layer for search.

Similarly, the subscribed data types are mainly divided into static, dynamic and instruction. For different types of subscribed data, the metadata information in the subscription request is different. Among them, the command subscription is quite special, and the command subscription is the demand for commands subscribed by the ground control center. The other two kinds of data are the required data subscribed by users (astronauts, earth users, etc.). Command subscriptions have a higher priority than other data. For subscribers, command subscription requests are transmitted first. For servers, when the upper limit of server subscription users is reached, command subscription requests can replace existing ordinary data subscriptions. For the subscription of dynamic data, use the method of data name + timestamp to accurately subscribe. If the timestamp is empty, it means specifying the current data. Finally, note that subscriptions should be divided into one-time subscriptions and continuous subscriptions. Most of the single subscription occurs in the subscription of static data, and the continuous subscription may occur in the subscription of instructions and dynamic data, and the continuous subscription will not be deleted by the server.

Step 9, the server replies (ACK) to the subscription request. The server needs to inform the subscriber whether the subscription is successful. If the subscription fails, the server needs to inform the reason, such as wrong timestamp, domain does not exist, command receiver does not exist, etc. If no reply to the request is received within the set time after subscription, the subscription is considered lost and needs to be re-subscribed.

Step 10, the user (astronaut, earth user, etc.) may unsubscribe from the data. Make sure to cancel the corresponding subscription using the same metadata as the subscription request.

Step 11, the server responds (ACK) to the user (astronaut, earth user, etc.) unsubscribing data. The server needs to inform the user whether the unsubscription is successful.

Step 12. The server communicates with the publisher (rovers, satellites, etc.) to ensure that the publisher is suitable for publishing data, and then the server sends the published meta-information to the subscribers to inform the publisher of the information. Data transfer can start.

Step 13, the subscriber communicates with the publisher, mainly to solve the problem of data transmission. There may be multiple publishers, so data transmission efficiency and fault tolerance can be improved through multi-source transmission, and the traffic burden of a single communication link can be reduced at the same time.

In order to adapt to the special transmission conditions in deep space, it is necessary to improve the data transmission method, such as using the managed transmission mechanism and reliable retransmission mechanism in DTN. In addition, a multi-source transmission mechanism for files can also be implemented to ensure transmission quality.

The multi-source transmission of files divides files into data blocks of uniform size. Each publisher only transmits a part of data blocks to subscribers. Subscribers can reassemble files after receiving all publishers’ data blocks. Among them, subscribers It is necessary to keep information synchronized with all publishers, inform existing blocks and missing blocks, so as to determine the data blocks sent by each publisher. In addition, after the subscriber has completely received the data, it can notify the server as the publisher of the data through a publishing request, and realize the source expansion of multi-source transmission through this logic.

Step 14. After the information transmission is completed, if the subscriber or publisher needs to log out of the network, it needs to send a logout request to the server to delete the relevant data and entries on the server.

It should be noted that the data cached by the server also needs to be processed as the publisher logs out. Tells the server whether to delete the server's data when the publisher logs out.

Step 15. The server responds (ACK) to the logout request. The server needs to inform the user whether the logout request is successful. If the logout fails, it needs to inform the reason. The reason may be domain mismatch, no such subscriber or publisher, data deletion failure, etc. .

Among them, there is no sequence between subscription and publication. During the implementation of the above steps 1-15, the data frame format shown in Figure 5 as an example can be used for data interaction.

The type of data frame is distinguished by Message_TYPE, including 7 types such as registration, publishing, and subscription; the length field only appears in the publishing/subscribing/unpublishing related information, indicating the length of the published message to ensure the complete reception of information; The USER_ID field is the unique identifier of the user; the DATA_TYPE field indicates the data type, and there may be many types of data: static data (STATIC_DATA), dynamic data (Dynamic_DATA), control instruction data (Command_DATA); the DATA_INFO field is only used for publishing/subscribing/unpublishing It appears in relevant information and is used to uniquely identify data. The specific content of DATA_INFO is shown in Figure 5. The first field is Info_length, the length of the entire DATA info structure. The second field is name, which is called name here, and actually refers to The name attribute related to the file content. For static and dynamic data, name means the name of the data. For instruction data, name refers to the name related to the instruction. The instruction is included in the naming. The particularity of dynamic data is that there are many A Time Stamp (time stamp) bit, which indicates the generation time of the file, and a Source field, which indicates the source of the release information, subscription information or instruction information. For the instruction information, only the instructions from the specified source will be accepted. Receive; the Scope field indicates the geographical location of the user, which is beneficial to information management, such as improving the efficiency of information retrieval; followed by optional fields, which correspond to different types of data frames with different values, heartbeat (KEEP_ALIVE) It only appears in the registration information to ensure that the registered user is always connected to the server. Cache_data only appears in the publishing information to tell the server whether to allow the content of the publication to be cached. Continuity only appears in the subscription information to distinguish between a single subscription and Continuous subscription, this field is only valid for dynamic and command data, Del_data only appears in the logout information, indicating whether the cached data on the server needs to be erased when the user logs out; the Success field only appears in the reply message, indicating publish/subscribe/cancel Whether the release is successful; the Reason field only appears in the reply message. When the user’s corresponding operation fails, explain the cause of the error, ERR_Time_stamp (timestamp error), ERR_Scope (domain error), insufficient storage space (ERR_Mem); the last part of the data frame is There is a reserved bit (reserved), which is used as an interface reserved bit for future multi-services/services.

In summary, the present invention designs a network architecture suitable for manned deep space exploration by using the publish-subscribe model of the information center network and multi-source transmission; meanwhile, it describes in detail the steps of publishing, subscribing, and information transmission The process and the data frame format within it. This makes the method clearer and more detailed, and easier to implement.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those skilled in the art of the present invention can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be defined by the claims.

Claims (10)

1. a kind of method of information centre's network architecture towards manned deep space exploration, which is characterized in that according to area of space and Node communication range is divided into different domains according to certain rules, can choose out the remittance in each domain according to connection plan Domain server of the poly- node as the domain, internally information management in responsible domain, is externally responsible for transmission link between building domain, realizes number According to cross-domain transmission；By the domain server, top level domain topological network is constituted between each domain, carries out group-net communication, data and letter Breath is synchronized and is transmitted between these top level domain；Data transmission and networking management in each top level domain pass through hierarchical domains Concept is realized；

Top level domain or domain space are subdivided into multiple second-level domains according to spatial position, each second-level domain also has domain server, belongs to In same top level domain all second-level domains can networking constitute second-level domain's topological network, data and information can be between these second-level domains It synchronizes and transmits；And each second-level domain can also according to circumstances segment third level domain；And so on, ultimately form hierarchical tree Network topology, interconnect between all domain servers, constitute the whole network backbone transport network；Overseas any information require by The detection of multilayer domain server；

According to longitude and latitude, height above sea level geographical information space position, top level domain or domain space are subdivided into multiple second-level domains, each Second-level domain also has domain server, belong to same top level domain all second-level domains can networking constitute second-level domain's topological network, data And information can be synchronized and be transmitted between these second-level domains；And each second-level domain can also according to circumstances segment third level domain； And so on, the network topology of hierarchical tree is ultimately formed, is interconnected between all domain servers, the whole network backbone transport network is constituted With tree-shaped network topology management；

Using the subscription issue mechanism of information centre's network, the transmission of combining information multi-source, networking, network topologies are solved not Come flexible networking between multi-service/service management of manned deep space exploration, network management, domain, tolerance latency issue；

Include the following steps:

Step 1, control plane building, for the step according to area of space and node communication range, top level domain is interior according to Spatial Rules It is divided into different domain spaces, all nodes can all belong to different domain spaces in network；And it is selected according to connection plan Domain server of the aggregation node as the domain in each domain space, internally information management in responsible domain, is externally responsible for building domain Between transmission link, realize data cross-domain transmission；Domain server is connected with each other simultaneously, forms top level control topological network, while each Category information transmitted in each domain space with it is synchronous；

Step 2, service discovery, the step include service discovery, service management, publication, subscription etc.；Each node of domain space exists It is registered on domain server, and issues or subscribe to certain type of service or data；

Step 3, data transmission step, carry out data transmission between subscriber and domain server or publisher, wherein there may be Multiple publisher's data having the same, i.e. the multi-source transmission of data；

Step 4, service are cancelled, the step, and user cancels publication or subscribes to certain service or data, and carries out on the server It nullifies；

The wherein realization process of step 1-4, is implemented by following manner:

1) building of control plane, the opening initialization including domain server, domain server are connected with each other communication；

2) service discovery, including the following steps: user's registration；The reply (ACK) that server requests user's registration；With Issue data/service request in family；Reply (ACK) of the server for posting request；User subscribes to data/service request；Service Device is for subscribing to the reply (ACK) requested；

3) data are transmitted, including the data transmission between subscriber and publisher, subscriber and server；Wherein there may be more A publisher's data having the same, i.e. the multi-source transmission of data；

4) service is cancelled, including cancels and issue, unsubscribe, user itself cancellation；It is respectively divided into following steps:

A. cancel publication: user cancels the publication to data；Server cancels user the reply (ACK) of publication data；

B. unsubscribe: user cancels the subscription to data；Server unsubscribes user the reply (ACK) of data；

C. user itself nullifies: user log off exits network；Reply (ACK) of the server for de-registration request.

2. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that described Service discovery step Node registry during, the registration information of user should cover all information of user；Including letter It ceases type (Message_TYPE): for illustrating the type of information；User ID (USER_ID): user's unique identification；Data type (DATA_TYPE): illustrating the information category that user subscribes to or issues, there is static data, dynamic data, director data etc.；User Domain (scope): illustrate the geographical location where user；Heart time (KEEP_ALIVE): the heartbeat of user and server negotiation Time；Service interface reserved bit (Reserved): for providing service/business interface element.

3. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that described Information issuing process in, the static metadata of publication should include publisher ID, the title of file, source address path, file point Block size, the Hash value information of file block；

In the information issuing process of the service discovery step, the metamessage lattice of director data, static data and dynamic data Formula has differences；The classification (Command) that the metamessage of director data instructs needed for including, and specify the source of the instruction (Command-source), i.e., the instruction only from specified source can just be received；Static data uses filename (name) Carry out unique identification；Wherein filename can be named according to the place domain of publisher and particular geographic location；Dynamic data is adopted Unique identification is carried out with timestamp (time stamp)+filename (name)；Wherein timestamp is for marking dynamic data Know, filename can be named according to the place domain of publisher and particular geographic location.

4. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that described Service discovery step information issuing process in, publisher allow server buffer hot topic data, facilitate data Multi-source distribution；Judgement for hot data, server can record each subscription information, count and order in a period of time The more content of number is read, local cache is carried out to the content；

During the information subscribing of the service discovery step, subscription should divide into single subscription and ongoing subscription；Single is ordered The subscription for mostly occurring in static data is read, ongoing subscription is likely to occur in the subscription to instruction and dynamic data, ongoing subscription It will not be deleted by server；It should include the option in subscription information frame.

5. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that described Data transmission procedure in, the multi-source transmission plan of data；File is divided into the unified data block of size, Mei Gefa by multi-source transmission Cloth person only transmits a part of data block to subscriber, and subscriber receives reconfigurable written after the data block of all publishers Part needs to keep the synchronous of information with all publishers in subscriber among these, informs existing block and lacks block, is determined often with this The data block that a publisher sends；In addition, passing through posting request, tell the server conduct after subscriber is completely received data The publisher of the data realizes that the source of multi-source transmission is expanded by the logic.

6. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that described Service cancellation step, either subscriber or publisher require to be unregistered, send out when shutting down or being detached from the current field When cloth person nullifies, need whether tell the server deletes the data cached of server；In addition, if user is crashed suddenly or movement suddenly Equal accidental causes lead to not active transmission log-off message, and the heartbeat mechanism of domain server can detect the lost contact of user, then Actively nullify the user.

7. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that

Between domain and domain, in domain between communication form a control plane, the networking mode of network formed on this, entirely The networking structure stratification of network；

The concept of binding domain, overall network framework are divided into two big flat functionals in function division: control plane, data plane；

Control plane: being formed by connecting by the server in each domain, and configuration information controls network, and domain server is responsible for data information pipe Reason and network topology management；User's registration, publication, subscription and de-registration request information control plane transmitted with it is synchronous；

Data plane: being made of each communication node, realizes data transmission and load service after data, service discovery, packet It includes multi-source transmission and Route Selection, each node serves as subscriber and issuer role；Each node is all located at oneself affiliated domain and meeting It is jumped with the domain server in affiliated domain by one or multi-hop realization communicates, sent release information or subscribe to solicited message.

8. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that control The node type of plane is the server in each domain；And it includes following several for controlling the link element type in plane:

1) repeated link between domain；Mainly interplanetary repeater satellite network can be laser link and RF link；

2) direct connected link between domain；Satellite and earth station's directly-connected network, including laser link and RF link；

3) intra-area communication link；The data between communication link and domain server and data center between each domain server Transmission link；

4) cross link；Communication link between synchronous satellite and low orbit satellite；

5) link is closed on；The direct-connected communication chain of low orbit satellite or synchronous satellite and earth station, communication base station, probe vehicles, astronaut Road；

6)W-LAN；Close range communication links between astronaut, user and probe vehicles and communication base station, server (are not limited to W-LAN can also be other communication modes such as bluetooth).

9. the method for information centre's network architecture according to claim 1 towards manned deep space exploration, which is characterized in that

And the node type of data plane includes following several:

1) probe vehicles；The detector on celestial body surface；Detect temperature, the humidity of planet, the information such as Atmospheric components；Subscriber can be served as With the role of publisher；

2) space probe satellite；It goes around the detector satellite in universe, the information such as detection cosmic ray, temperature；It can serve as The role of subscriber and publisher；

3) astronaut；The role of subscriber and publisher can be served as；

4) data center；Store data；Serve as issuer role；

5) repeater satellite, including synchronous relay satellite, low rail repeater satellite etc.；Serve as the backbone transport networks role of communication；

6) ground satellite station；Serve as the access unit of communication；

7) signal station；Network signal relay base station on the earth and other celestial bodies；

8) polar region platform；The base platform of probe vehicles and astronaut.

10. the method for information centre's network architecture according to claim 9 towards manned deep space exploration, which is characterized in that

Wherein 1-4 node type belongs to user type, can serve as the role of subscriber or publisher；5-8 node type all belongs to It in domain server type, is served as in domain and subscribes to the storage of the information such as publication, synchronous role, overseas, serve as chain between domain The backbone node on road；

There are a plurality of types of data in data plane, these data are divided into following several by type:

1) director data；The control centre in earth domain is sent to the detector in each planet domain and the instruction of astronaut；

2) static data；The data that will not be changed, such as text data, image, offline video etc.；

3) dynamic data；The data for changing over time or updating；Such as the data of call voice, detector detection.