Distributed calculation simulation method and system based on gRPC radar

Abstract

The invention discloses a radar distributed computing simulation method and system based on gRPC, the system comprises a client and a server cluster, and the server cluster comprises a scheduling server and at least one distributed function server. The method applied to the system comprises the steps that a client generates subtasks to be executed by all servers according to simulation tasks and sends the subtasks to a scheduling server, the scheduling server distributes the subtasks to distributed function servers used for executing the subtasks, and the distributed function servers obtain execution results after executing the subtasks and send the execution results to the scheduling server; the scheduling server summarizes the execution results to obtain simulation results and sends the simulation results to the client; wherein communication channels between the client and the servers and between the servers are established in advance based on the gRPC framework. The communication efficiency of the system can be improved, and the communication stability is enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar simulation, and particularly relates to a radar distributed computing simulation method and system based on gRPC. Background Art

[0002] With the continuous development and application of radar technology, the distributed simulation of radar has become a field of great concern. The distributed simulation of radar has the characteristics of distributed deployment of multiple radar module nodes, and is one of the important means to evaluate the system performance, optimize the system design and verify the system reliability. It can achieve target monitoring and tracking with a wider range and higher accuracy, and is usually used in fields such as air traffic management and weather monitoring, and has strict requirements for the performance and reliability of the system. Traditional simulation methods are mainly based on single-machine computing, such as distributed simulation methods based on Socket communication, distributed simulation methods based on HTTP communication, and distributed simulation methods based on distributed computing frameworks.

[0003] The distributed simulation method based on Socket communication uses network sockets for communication between nodes, and realizes information exchange by sending and receiving data. Due to the performance limitations of Socket communication, the data transmission efficiency is low, and it is easily affected by problems such as network latency and packet loss.

[0004] The distributed simulation method based on HTTP communication uses the HTTP protocol for communication between nodes, and realizes data exchange through HTTP requests and responses. Due to the characteristics of HTTP communication, a connection needs to be established and disconnected for each communication, resulting in higher communication overhead and latency.

[0005] The distributed simulation method based on a distributed computing framework uses a distributed computing cluster for simulation calculation, and realizes communication and data exchange between nodes through a message passing mechanism. Although this method has certain improvement in performance, there are still problems such as low communication efficiency and susceptibility to network conditions.

[0006] Therefore, the current distributed radar simulation methods have problems such as low communication efficiency and poor communication stability, and cannot fully meet the system's requirements for efficient and stable communication. Summary of the Invention

[0007] The embodiments of the present invention provide a radar distributed computing simulation method and system based on gRPC, which can solve the problems of low communication efficiency and poor communication stability existing in the current distributed radar simulation methods.

[0008] In a first aspect, an embodiment of the present invention provides a gRPC-based radar distributed computing simulation method, which is applied to a gRPC-based radar distributed computing simulation system. The system includes a client and a server cluster, and the server cluster includes a scheduling server and at least one type of distributed function server. The method includes:

[0009] The client generates sub-tasks to be executed by each server according to the simulation task and the load status and priority relationship of each server in the server cluster.

[0010] The client sends a task request message to the scheduling server, where the task request message is used to indicate the sub-tasks to be executed by each server, and the communication channel between the client and the scheduling server is established in advance based on the gRPC framework.

[0011] The scheduling server distributes the sub-tasks to be executed by each distributed function server according to the task request message, where the communication channel between the distributed function server and the scheduling server is established in advance based on the gRPC framework.

[0012] The distributed function server executes the sub-tasks to be executed by itself to obtain an execution result and sends the execution result to the scheduling server.

[0013] The scheduling server aggregates the execution results to obtain a simulation result and sends the simulation result to the client.

[0014] In a second aspect, an embodiment of the present invention provides a gRPC-based radar distributed computing simulation system, which includes a client and a server cluster. The server cluster includes a scheduling server and at least one distributed function server.

[0015] The client includes a task scheduling module, which is used to generate sub-tasks to be executed by each server according to the simulation task and the load status and priority relationship of each server in the server cluster; send a task request message to the scheduling server, where the task request message is used to indicate the sub-tasks to be executed by each server, and the communication channel between the client and the scheduling server is established in advance based on the gRPC framework.

[0016] The scheduling server is used to distribute the sub-tasks to be executed by each distributed function server according to the task request message, where the communication channel between the distributed function server and the scheduling server is established in advance based on the gRPC framework.

[0017] The distributed function server is also used to execute the subtasks to be executed by itself to obtain an execution result and send the execution result to the scheduling server;

[0018] The scheduling server is used to aggregate the execution results to obtain a simulation result and send the simulation result to the client.

[0019] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: According to the simulation method provided by the present invention, the system communicates under the gRPC framework, which can achieve multiplexing and data compression, significantly reducing the communication delay caused by the radar node module when processing a large amount of data and improving the data transmission rate; at the same time, gRPC supports two-way communication, allowing the node to send its own calculation results while receiving data sent by other modules, further improving the communication efficiency. Moreover, the gRPC framework provides a powerful connection management and flow control mechanism, which can effectively cope with network fluctuations and communication interruptions, ensuring the stability and reliability of data transmission; the gRPC framework also supports real-time data transmission and processing, which can meet the high requirements of radar distributed simulation for real-time and synchronization, improving the response speed and accuracy of the simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic structural diagram of a radar distributed computing simulation system based on gRPC provided by an embodiment of the present invention;

[0021] Figure 2 It is a flowchart of an implementation of a method for establishing a communication channel based on gRPC provided by an embodiment of the present invention;

[0022] Figure 3 It is a flowchart of an implementation of a radar distributed computing simulation method based on gRPC provided by an embodiment of the present invention;

[0023] Figure 4 It is a schematic diagram of a scenario where a client communicates with a server provided by an embodiment of the present invention;

[0024] Figure 5 It is a schematic structural diagram of a specific radar distributed computing simulation system based on gRPC provided by an embodiment of the present invention;

[0025] Figure 6 It is a schematic diagram of a communication interface between a client and a server provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures, technologies, etc. are presented to provide a thorough understanding of the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0027] It should be understood that when used in the specification and claims of the present invention, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0028] It should also be understood that the term "and / or" as used in the specification and claims of the present invention refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0029] As used in the specification and claims of the present invention, the term "if" can be interpreted as "when" or "once" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrases "if determined" or "if detecting [the described condition or event]" can be interpreted as meaning "once determined" or "in response to determining" or "once detecting [the described condition or event]" or "in response to detecting [the described condition or event]" depending on the context.

[0030] In addition, in the description of the specification and claims of the present invention, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0031] The reference to "one embodiment" or "some embodiments" or the like described in the specification of the present invention means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the present invention. Thus, the statements "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having", and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0032] The present invention will be further described in detail below in conjunction with specific embodiments, but the embodiments of the present invention are not limited thereto.

[0033] Embodiment 1

[0034] Figure 1 The figure shows a schematic structural diagram of a radar distributed computing simulation system provided by an embodiment of the present invention. By way of example and not limitation, system 100 may include a client 110 and a server cluster 120, and the server cluster 120 may include a scheduling server 121 and at least one distributed function server 122.

[0035] In a possible implementation manner, before performing radar simulation, system 100 may first establish communication channels between various nodes based on the gRPC framework, such as the communication channels between the client 110 and the server cluster 120 and between the servers within the server cluster.

[0036] In an example, before establishing the communication channels, the user may define the service interfaces and message types of system 100, and generate the code stubs of the client and the server from the object interface definition language (IDL) file according to the definition results by using the protobuf compiler. The server side implements the interfaces defined in the IDL, and the client uses the stub to call the methods of the server side.

[0037] Exemplarily, the service interface definition of the system may include the remote procedure call protocol (RPC) methods of each node (server and client) in the radar simulation. RPC mainly may include InitializeSimulationService, RunSimulation, TerminateSimulation, etc., and RegisterNode, SendHeartbeat, etc. for node management and monitoring.

[0038] Exemplarily, the message type definition may include the data structures of requests and responses transmitted between the client and the server, such as radar parameter information, target information, scheduling instruction information, etc. The key message types include InitializeSimulation request / response, RunSimulation request / response, TerminateSimulation request / response, RegisterNode request / response, SendHeartbeat request / response, etc.

[0039] Exemplarily, the IDL of system 100 may be Protocol Buffers.

[0040] Defining the service interfaces and message types through IDL can ensure that the system has good cross-language and cross-platform compatibility and supports multiple programming languages and operating systems.

[0041] In a possible implementation, after receiving a simulation request, the client 110 may generate subtasks to be executed by each server according to the simulation task, the load conditions of each server, and the priority relationship, and then send them to the scheduling server 121. The scheduling server then distributes the relevant subtasks to be executed and the data required for the execution tasks to the distributed function server 122 to perform distributed radar simulation.

[0042] Embodiment 2

[0043] Based on the above Embodiment 1, Figure 2 The following shows a flowchart of an implementation of a method for establishing a communication channel based on gRPC provided by an embodiment of the present invention. By way of example and not limitation, this communication channel establishment method may be applied to the above system 100. The method may include steps S201 - S205, which are described below.

[0044] S201, each server in the server cluster and the client are initialized.

[0045] In an example, the initialization process of a node may include: loading a configuration file, setting communication parameters, initializing the gRPC library, and setting basic gRPC parameters.

[0046] Exemplarily, the communication parameters may include a port number, an authentication message.

[0047] Exemplarily, the gRPC parameters may include a maximum message size, a connection timeout, etc.

[0048] S202, the client creates a gRPC service instance and sets a listening port.

[0049] In an example, the gRPC service instances may correspond one - to - one with the servers in the server cluster. The client can manage the servers corresponding to the service instances by managing a node list composed of the service instances.

[0050] In an example, through the listening port, the client can receive a registration request from the server.

[0051] S203, the server sends a registration request to the client.

[0052] Exemplarily, the registration request may include a server ID, a server type, service capabilities, and current status information of the server.

[0053] S204, the client verifies the legitimacy of the server according to the registration request and returns a registration response to each server.

[0054] Exemplarily, the registration response can be used to indicate whether the server registration is successful.

[0055] Specifically, if the server verification is successful, the client can add the information of the server to its corresponding service instance, add it to the node list, and at the same time send a registration success response to the server. If the server verification fails, the client can return a registration failure response to the server and may also require the server to re-register.

[0056] S205, the client establishes a communication channel with the successfully registered server, and the successfully registered server establishes a communication channel with other successfully registered servers.

[0057] In one example, the client can establish a communication channel with each successfully registered server based on the gRPC framework. At the same time, inside the successfully registered server, a communication channel can be established between itself and other successfully registered servers based on the gRPC framework according to communication requirements, such as the communication channels between the scheduling server 121 and the distributed function server 122, and between the distributed function servers 122.

[0058] Exemplarily, these communication channels can be an efficient communication framework based on gRPC, and use two-way transport streams based on the HTTP / 2 protocol for information transmission. The gRPC framework allows nodes to send and receive data on the same connection, making the communication between each node module (server) of the radar more real-time.

[0059] Embodiment 3

[0060] Based on Embodiment 2, Figure 3 The following shows a flowchart of the implementation of a gRPC-based radar distributed computing simulation method provided by an embodiment of the present invention. By way of example and not limitation, the simulation method may include steps S301 - S306, which will be described below.

[0061] S301, the client generates sub-tasks to be executed by each server according to the simulation task, the load status and priority relationship of each server in the server cluster.

[0062] Exemplarily, after establishing the communication channel through the above method, each server in the server cluster has sent its registration request to the client and received the registration response from the client, and has been successfully registered. At this time, the client can store and maintain a node list through the server ID, and this list can record all registered servers and their status information.

[0063] S302, the client sends task request information to the scheduling server.

[0064] Exemplarily, the task request information can be used to indicate the subtasks to be executed by each server.

[0065] In one example, referring to Figure 4 , when the client sends task request information, it can use the dynamic proxy technology of JDK, write the codeword reflecting the task request information using the invoke method; then use the Filter method to implement functions such as monitoring, routing, degradation, and authentication; afterwards, it can obtain a thread from the client's thread pool (Channel Pool), and then write the data into the thread (Channel); finally, serialize and compress the data to reduce the amount of data transmitted over the network and send the task request to the task scheduling server.

[0066] S303. The scheduling server distributes the subtasks to be executed to each distributed function server according to the task request information.

[0067] In one example, similarly, referring to Figure 4 , after receiving the task request, the scheduling server can first decompress and deserialize it, and then hand the data to the thread pool for processing to obtain the task request information.

[0068] Exemplarily, the scheduling server can extract the waveforms and other setting parameters required for the current simulation task from its built-in database according to the task request information, and send them to the distributed function server together with each subtask.

[0069] The client remotely calls the distributed function server through the relay scheduling server in the gRPC communication framework. This framework supports the automatic generation of communication code for the client and the server. Developers only need to define the call and output interfaces and data structures of each radar node, which greatly reduces the development difficulty and time cost and simplifies the communication process between nodes.

[0070] S304. The distributed function server executes its own subtasks to obtain the execution results.

[0071] In a possible implementation, the types of distributed function servers can include antenna simulation servers, receivers, signal processing servers, and data processing servers. Servers of different types can implement different functions.

[0072] In one example, the antenna simulation server can perform scanning and resolution on the targets in the airspace according to the model parameters to obtain the antenna resolution results.

[0073] Exemplarily, the model parameters can include the waveforms and other setting parameters required for the current simulation task.

[0074] In one example, the receiver can generate signals such as target echo signals, interference signals, clutter signals, and noise based on the antenna solution results.

[0075] In one example, the signal processing server can perform anti-interference processing on the target echo signal, and perform target detection through processing such as pulse compression and accumulation, and then measure the position information of the target.

[0076] Exemplarily, the position information of the target may include parameter information such as the distance, speed, and pitch azimuth angle of the target.

[0077] In one example, the data processing server can perform filtering processing on the target according to the parameter measurement results (i.e., the position information of the target) sent back by the signal processing, form information such as target traces and tracks, and can perform track association and extrapolation to predict the target category; at the same time, an event request is generated, and the category information of the target is sent back to the scheduling server.

[0078] Exemplarily, the category information of the target may include target traces, tracks, target categories, etc.

[0079] S305, the distributed function server sends the execution result to the scheduling server.

[0080] Exemplarily, the execution result of the antenna simulation server may be the antenna solution result, the execution result of the receiver may be signals such as target echo signals, clutter, interference, and noise, the execution result of the signal processing server may be the position information of the target, and the execution result of the data processing server may be the category information of the target.

[0081] S306, the scheduling server aggregates the execution results to obtain the simulation result, and sends the simulation result to the client.

[0082] Exemplarily, the simulation result may include the antenna solution result, the target echo signal, the position information of the target, and the category information of the target.

[0083] According to the simulation method provided by the present invention, the system communicates under the gRPC framework, can achieve multiplexing and data compression, significantly reduces the communication delay caused by the radar node module when processing a large amount of data, and improves the data transmission rate; at the same time, gRPC supports two-way communication, allowing the node to send its own calculation results while receiving data sent by other modules, further improving the communication efficiency. And the gRPC framework provides a powerful connection management and flow control mechanism, which can effectively cope with network fluctuations and communication interruptions, ensuring the stability and reliability of data transmission; the gRPC framework also supports real-time data transmission and processing, can meet the high requirements of radar distributed simulation for real-time and synchronization, and improves the response speed and accuracy of the simulation.

[0084] Furthermore, the system uses the IDL of gRPC to define service interfaces and message types, enabling developers to develop and maintain using different programming languages and different operating systems on different servers, which is conducive to improving development efficiency. At the same time, the gRPC framework supports automatic generation of communication code for the client and server sides. With the unified gRPC communication protocol, developers only need to define service interfaces and data structures, greatly reducing the development difficulty and time cost, and simplifying the system development and maintenance work.

[0085] Embodiment 4

[0086] Based on Embodiment 3, Figure 5 The figure shows a specific structural schematic diagram of a gRPC radar distributed computing simulation system provided by an embodiment of the present invention.

[0087] In some embodiments, referring to Figure 5 , the client 110 may include a service registration module, a task scheduling module, a status monitoring module, and a fault handling module.

[0088] The service registration module may execute steps S201, S202, and S205 in the above channel establishment method during server registration. The status monitoring module may receive a registration request from the server in step S203. In addition, the status monitoring module may also monitor the load status of each server. The task scheduling module may be used to execute step S301 in the above simulation method. The fault handling module may handle error messages from the server and the client itself.

[0089] Optionally, referring to Figure 6 , the client 110 may use RegistResponse RequestInfo as its communication interface.

[0090] In a possible implementation manner, referring to Figure 5 , the distributed function server 122 may include an antenna simulation server 1221, a receiver 1222, a signal processing server 1223, and a data processing server 1224.

[0091] In an example, referring to Figure 5 , the scheduling server 121 may include a waveform library component, an event library component, and a scheduling component.

[0092] Exemplarily, the scheduling server 121 may extract the waveforms required for the current simulation task from the waveform library component, extract other setting parameters required for the current simulation task from the event library component, and under the action of the scheduling component, send this information and each subtask to the distributed function server 122 that can execute the subtask.

[0093] Optionally, refer to Figure 6 , both the distributed function server 122 and the scheduling server 121 can use RegistResponse RequestInfo as their communication interfaces.

[0094] In one example, refer to Figure 5 , the antenna simulation server 1221 can include an antenna pattern module, a wave position arrangement module, and a scenario resolution module.

[0095] The antenna pattern module can be used to generate an antenna pattern. The wave position arrangement module can be used to define the search mode of the antenna, thereby generating a detection scenario for the antenna. The scenario resolution module can be used to simulate the antenna to scan and resolve the targets in the airspace in this scenario to obtain the antenna resolution result.

[0096] In one example, the receiver 1222 can include a target echo generation module, an interference signal generation module, a clutter signal generation module, and a noise signal generation module. These modules can be used to generate target echo signals, interference signals, clutter signals, and noise respectively.

[0097] Exemplarily, the target echo signal can be composed of a target signal, an interference signal, a clutter signal, and noise.

[0098] In one example, the signal processing server 1223 can include an anti-interference module, a pulse compression and accumulation module, a target detection module, and a parameter measurement module. The anti-interference module can perform anti-interference processing on the target echo signal. The pulse compression and accumulation module can perform pulse compression and accumulation processing on the anti-interference processed target echo signal. The target detection module can perform target detection. The parameter measurement module can generate the position information of the target according to the result of the target detection.

[0099] In one example, the data processing server 1224 can include a track association module, a track filtering module, and a target recognition module.

[0100] Exemplarily, the track association module can eliminate the data and error data belonging to false targets in the position information of the target to obtain the associated position information of the target. The track filtering module can perform filtering processing on the associated position information of the target. The target recognition module can identify the type of the target according to the filtered data to obtain the category information of the target.

[0101] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

Claims

1. A radar distributed computing simulation method based on gRPC, characterized in that: The method is applied to a gRPC-based radar distributed computing simulation system, the system comprising a client and a server cluster, the server cluster comprising a scheduling server and at least one distributed function server; the method comprises: The client generates a subtask to be executed by each server according to the simulation task and the load status and priority relationship of each server in the server cluster; The client sends task request information to the scheduling server, wherein the task request information is used to indicate the subtask to be executed by each server, and the communication channel between the client and the scheduling server is established in advance based on the gRPC framework; The scheduling server sends the subtask to be executed to each of the distributed function servers according to the task request information, wherein the communication channel between the distributed function server and the scheduling server is established in advance based on the gRPC framework; The distributed function server executes its own to-be-executed subtask to obtain an execution result and sends the execution result to the scheduling server; The scheduling server aggregates the execution results to obtain simulation results, and sends the simulation results to the client.

2. The method according to claim 1, characterized in that: Before executing the first simulation request, the method further includes: The server cluster and the client are initialized; The client creates a gRPC service instance corresponding to each server in the server cluster to manage the server corresponding to the gRPC service instance, and sets a listening port to receive a registration request from the server; The client verifies the legitimacy of the server according to the registration request, and returns a registration response to each server, wherein the registration response is used to indicate whether the server is successfully registered, and the server whose legitimacy verification succeeds can be successfully registered; The client establishes a communication channel with each successfully registered server based on the gRPC framework, and the successfully registered server establishes a communication channel with other successfully registered servers based on the gRPC framework.

3. The method according to claim 2, characterized in that The initialization process of the server cluster and the client includes: loading configuration files, setting communication parameters, initializing the gRPC library, and setting basic gRPC parameters; Among them, the communication parameters include port number and authentication message; the gRPC parameters include maximum message size and connection timeout.

4. The method according to claim 2, characterized in that: The registration request includes server ID, server type, service capability and current status information of the server.

5. The method according to claim 1, characterized in that The service interface and message type of the system are defined by object interface definition language.

6. The method according to claim 5, characterized in that The object interface definition language is ProtocolBuffers.

7. A gRPC-based radar distributed computing simulation system, characterized in that: The system includes a client and a server cluster, wherein the server cluster includes a scheduling server and at least one distributed function server; The client comprises a task scheduling module, which is used to generate a subtask to be executed by each server according to the simulation task and the load status and priority relationship of each server in the server cluster; Sending task request information to the scheduling server, wherein the task request information is used to indicate the subtask to be executed by each server, and the communication channel between the client and the scheduling server is established in advance based on the gRPC framework; The scheduling server is used to send the subtask to be executed to each of the distributed function servers according to the task request information, wherein the communication channel between the distributed function server and the scheduling server is established in advance based on the gRPC framework; The distributed function server is also used to execute its own to-be-executed subtask to obtain an execution result and send the execution result to the scheduling server; The scheduling server is used to summarize the execution results to obtain simulation results, and send the simulation results to the client.

8. The system according to claim 7, characterized in that The client also includes a server registration module and a status monitoring module; After the servers in the server cluster are initialized, the server registration module is used to create a gRPC service instance corresponding to each server one by one to manage the server corresponding to the gRPC service instance; The server registration module is further used to verify the legitimacy of the server according to the registration request received by the status monitoring module from each server, and return a registration response to each server, wherein the registration response is used to indicate whether the server is successfully registered, and the server with successful legitimacy verification can be successfully registered; After each of the servers receives the registration response, the server registration module is further used to establish a communication channel between the client and each successfully registered server based on the gRPC framework; the successfully registered server is also used to establish a communication channel with other successfully registered servers based on the gRPC framework.

9. The system according to claim 7, characterized in that The simulation results include antenna solution results, target echo signals, target location information and target category information; Wherein, the distributed function server includes: antenna simulation server, receiver, signal processing server, data processing server; The antenna simulation server is used to scan and solve the target in the airspace according to the model parameters to obtain the antenna solution result; The receiver is used to generate the target echo signal according to the antenna solution result; The signal processing server is used to perform anti-interference processing and pulse pressure accumulation processing on the target echo signal, and perform target detection on the processed target echo signal to obtain the position information of the target; The data processing server is used to filter the target according to the location information of the target to obtain the category information of the target.

Images