SERVERSYSTEM

Abstract

A server system comprises a multitude of prediction servers and a mediator server that mediates data transfer between the prediction servers. A prediction server contains a capture unit that captures a communication service protocol regarding a sub-communication area, a prediction unit that predicts the future communication service quality of that sub-communication area based on the communication service protocol, and a communication unit that transmits a first prediction result from the prediction unit to the mediator server and selectively receives a second prediction result from another prediction server from the mediator server. Based on the first and second prediction results, the prediction unit forecasts the communication service quality between a node within one sub-communication area and another node within a different sub-communication area.

Description

BACKGROUND OF THE INVENTION 1. Field of the invention

[0001] The present invention relates to a technical area of ​​a server system that predicts communication quality. 2. Description of the related prior art

[0002] For example, one such system has been proposed that estimates the radio quality in a coverage area using a radio wave propagation simulator and predicts a deterioration of the radio quality at a future location of a wireless terminal device (see JP 2009 - 278 421 A). SUMMARY OF THE INVENTION

[0003] A digital twin has been proposed in which artificial intelligence (AI) performs analyses and simulations in a virtual space using data collected in a real-world environment, and provides feedback to the real world. Implementing a digital twin in the technical field of communication services, for example, requires collecting, accumulating, and analyzing an enormous amount of communication service protocols. This makes the overhead of data collection a significant problem.

[0004] The present invention was made with regard to the problem described above and is aimed at providing a server system that is able to reduce the effort required for data acquisition.

[0005] A server system according to one aspect of the present invention is a server system with a plurality of prediction servers that predict the communication service quality of partial communication areas for which each of the prediction servers is responsible, and a mediation server that mediates the data transfer between the prediction servers, wherein a prediction server among the prediction servers comprises a detection unit that detects a communication service protocol with respect to a partial communication area for which the one prediction server is responsible, a prediction unit that comprises a first communication service quality that predicts the future communication service quality of the one partial communication area based on the communication service protocol, and a prediction unit,which sends a first prediction result indicating the first communication service quality to the mediation server and selectively receives a second prediction result from the mediation server, predicted by another prediction server among the prediction servers; the prediction unit predicts a second communication service quality, which is a future communication service quality between a node within one sub-communication area and another node within another sub-communication area, for which the other prediction server is responsible, using at least one of the first prediction result or the second prediction result; and the server system further includes an instruction unit that instructs one node to change a communication setting if the second communication service quality is lower than a predetermined quality. BRIEF DESCRIPTION OF THE FIGURES

[0006] Features, advantages and technical and industrial significance of exemplary embodiments of the invention are described below with reference to the accompanying figures, in which the same symbols denote the same elements and in which: Fig. Figure 1 is a conceptual diagram illustrating the concept of a server system according to one embodiment; and Fig. Figure 2 is a block diagram showing a configuration of the server system according to the embodiment. DETAILED DESCRIPTION OF THE EXECUTION FORMS

[0007] One embodiment of a server system is described with reference to Fig. 1 and Fig. 2 described.

[0008] In Fig. Server system 1 comprises servers 10a, 10b, 10c, and server 20. Servers 10a, 10b, 10c, and server 20 are configured to communicate with each other. For example, Message Queue Telemetry Transport (MQTT) can be used as the communication protocol between servers 10a, 10b, 10c, and server 20. In other words, server system 1 can be a publish-subscribe system. The number of servers provided for in server system 1 is an example and not limited.

[0009] Server 10a can capture a communication service log from each node contained in a communication scope Ar1. Server 10a can reproduce a communication environment of communication scope Ar1 in a virtual space using the captured communication service log. Server 10b can capture a communication service log from each node contained in a communication scope Ar2. Server 10b can reproduce a communication environment of communication scope Ar2 in a virtual space using the captured communication service log. Server 10c can capture a communication service log from each node contained in a communication scope Ar3. Server 10c can reproduce a communication environment of communication scope Ar3 in a virtual space using the captured communication service log.Note that servers 10a, 10b, 10c can be described as "digital twin servers".

[0010] The node can be, for example, a vehicle with a communication function, a relay device, a server device, or similar. It should be noted that the vehicle with the communication function can be a connected car. For example, a communication service protocol that refers to the vehicle as a node can include at least the vehicle's position, a communication history, communication quality, or server response delay. It should be noted that the vehicle's position can be interpreted as the position of a communication device. It should be noted that each of the communication areas Ar1, Ar2, and Ar3 can be defined based on at least one cell covered by a wireless base station or an area controlled by an accommodation station upstream of each wireless base station.

[0011] Server 20 facilitates data transfer between servers 10a, 10b, and 10c. For example, server 20 can transfer data collected by server 10a to at least one server 10b or 10c. Server 20 can transfer data collected by server 10b to at least one server 10a or 10c. Server 20 can transfer data collected by server 10c to at least one server 10a or 10b. Note that server 20 can be described as a "digital twin broker."

[0012] The operation of server system 1 is described below with reference to Fig. 2 in addition to Fig. 1 described. In Fig. 2. Server 10a comprises a protocol processing unit 11, a protocol database 12, a communication environment prediction unit 13, a network configuration formulation unit 14, and a communication device 15. Server 20 contains a prediction result delivery unit 21. It should be noted that servers 10b and 10c may have a similar configuration to server 10a. A vehicle 100, as a node, comprises a protocol acquisition unit 110 and a network configuration update unit 120.

[0013] The log acquisition unit 110 of vehicle 100 can transmit a communication service protocol to server 10a, which is responsible for the communication area Ar1 in which vehicle 100 is located. The communication service protocol, which refers to vehicle 100 as a node, can include, for example, a history of position information for vehicle 100, the amount of communication data sent / received by vehicle 100, the throughput / latency at that time, the intensity and signal-to-noise ratio (S / N ratio) of a signal received by vehicle 100 from a wireless base station, data processing delay, server load level, and similar information. Note that the log acquisition unit 110 can transmit the communication service protocol to server 10a on a regular basis.

[0014] The protocol processing unit 11 of server 10a can capture the communication service protocol transmitted by vehicle 100. Protocol processing unit 11 can also capture communication service protocols transmitted by nodes other than vehicle 100 within the communication range Ar1. Protocol processing unit 11 can subject the communication service protocols captured by a multitude of nodes within the communication range Ar1 to predetermined processing or comparison. Protocol processing unit 11 can store the result of the comparison in the protocol database 12. It should be noted that protocol database 12 can also store previous results in addition to the current result.

[0015] The communication environment prediction unit 13 of server 10a can, for example, predict the future behavior of vehicle 100 as a node and the communication performance between nodes (e.g., vehicle 100 and vehicle 101) within communication range Ar1 based on the summed results of the communication service protocols stored in protocol database 12. In other words, the communication environment prediction unit 13 can predict the future communication service quality of communication range Ar1. In this case, the communication environment prediction unit 13 can use an AI that outputs the future behavior of vehicle 100 and the communication performance between nodes within communication range Ar1 by receiving inputs of the tally results of the communication service protocols.

[0016] It should be noted that the future behavior of vehicle 100 may include its motion trajectory, communication quantity, communication time, and similar factors. It should be noted that the position of vehicle 100 (e.g., its motion trajectory) can be predicted using a prediction algorithm described in Nachiket Deo, Mohan M. Trivedi; Convolutional Social Pooling for Vehicle Trajectory Prediction, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2018, pp. 1581-1589. It should be noted that the communication quality can be predicted using a radio propagation simulator, a communication simulator, or similar methods, or through regression analysis, machine learning, or similar techniques.

[0017] The communication unit 15 can transmit an initial prediction result, which is the prediction result of the communication environment prediction unit 13, to server 20. Server 20 can store the initial prediction result transmitted by server 10a in a storage device (not shown). The prediction result delivery unit 21 of server 20 can transmit the initial prediction result to at least one of servers 10b or 10c. For example, a topic related to the initial prediction result can be associated with it. The prediction result delivery unit 21 can transmit the initial prediction result to a server where the topic associated with the initial prediction result is registered between servers 10b and 10c based on the topic associated with the initial prediction result.It should be noted that the prediction result delivery unit 21 may discard the first prediction result if there is no server on which the topic associated with the first prediction result is registered.

[0018] Server 10b can predict the future communication service quality of communication area Ar2 based on the communication service protocols (e.g., the sum of the communication service protocols). Server 10b can transmit a second prediction result, indicating the predicted communication service quality, to server 20. The prediction result delivery unit 21 of server 20 can transmit the second prediction result to at least one of servers 10a or 10c.

[0019] Server 10c can predict the future communication service quality of communication area Ar3 based on the communication service protocols (e.g., the sum of the communication service protocol results). Server 10c can transmit a third prediction result, indicating the predicted communication service quality, to server 20. The prediction result delivery unit 21 of server 20 can transmit the third prediction result to at least one of servers 10a or 10b.

[0020] For example, the communication environment prediction unit 13 of server 10a can predict the future communication service quality between vehicle 100 as a node in communication area Ar1 and a server 200 (see Fig.1) as a node in communication area Ar2. In this case, the communication device 15 of server 10a can receive the prediction result (e.g., the second prediction result described above) of server 10b, which is responsible for communication area Ar2, from server 20. It should be noted that, in addition to the prediction result of server 10b, the communication device 15 can also receive the prediction result of the communication service quality of one or more servers responsible for respective nodes on a communication path between vehicle 100 and server 200 from server 20.

[0021] For example, the communication environment prediction unit 13 of server 10a can predict the future communication service quality between vehicle 100 and server 200 by using at least one of the following results: the future communication service quality (e.g., the first prediction result described above) of communication area Ar1, or the prediction result (e.g., the second prediction result described above) of server 10b. The communication service quality may include an end-to-end delay. The communication environment prediction unit 13 can input a fourth prediction result, specifying the future communication service quality between vehicle 100 and server 200, into the network configuration formulation unit 14.

[0022] The network configuration formulation unit 14 can determine whether the communication service quality indicated by the fourth prediction result meets a required predetermined quality. If the communication service quality indicated by the fourth prediction result is lower than the predetermined quality, the network configuration formulation unit 14 can transmit a network configuration instruction to the vehicle 100 via the communication device 15. It should be noted that the network configuration formulation unit 14 does not need to issue the network configuration instruction if the communication service quality indicated by the fourth prediction result meets the predetermined quality.The network configuration instruction can include, for example, updating a routing table, changing a communication target application server, changing the means of communication, changing the transmission frequency of data, or the level of detail and compression ratio of transmitted data.

[0023] For example, updating the routing table is expected to reroute communication traffic to avoid congested network segments. Similarly, changing the communication destination application server might result in selecting a server with low end-to-end latency. Switching the communication medium could involve, for instance, switching between different cellular networks or using a combination of different types of communication networks (such as Wi-Fi (registered trademark) and satellite communication).

[0024] The network configuration update unit 120 of the vehicle 100, as a node that has received the network configuration instruction, can reproduce the network configuration instruction in the communication control.

[0025] It should be noted that each time the communication environment prediction unit 13 predicts the future communication service quality of communication area Ar1, server 10a can transmit a prediction result indicating the newly predicted communication service quality to server 20 via communication device 15. When server 10a transmits the new prediction result to server 20, it can delete the previous prediction result relating to communication area Ar1 from server 20.

[0026] If, for example, server system 1 is a publish-subscribe system, server 10a can selectively receive only one prediction result to be used by server 10a from server 20, from a multitude of prediction results that indicate the communication service quality and were predicted by other servers (e.g., at least one of servers 10b or 10c). This can reduce the communication overhead. Technical effects

[0027] In server system 1, servers 10a, 10b, and 10c collect communication service logs relating to the communication domains Ar1, Ar2, and Ar3, for which servers 10a, 10b, and 10c are each responsible. Compared to a server system where only one server collects communication service logs, server system 1, according to the present embodiment, can reduce the data acquisition effort.

[0028] In other words, it is necessary to collect a vast amount of communication service protocols from the respective nodes in a network without delay in order to predict communication service quality, such as the reception quality of a radio wave, the degree of network congestion, and the processing load of a server, with high accuracy. In server system 1 according to the present embodiment, the plurality of servers 10a, 10b, 10c are distributed to perform predictive processing of communication service quality, making it possible to define an area within which each of the servers 10a, 10b, 10c collects communication service protocols. This allows server system 1 to reduce the overhead of data acquisition.

[0029] Aspects of the invention that result from the embodiment described above are described below.

[0030] A server system according to one aspect of the invention is a server system with a plurality of prediction servers that predict the communication service quality of sub-communication areas for which each of the prediction servers is responsible, and a mediation server that mediates the data transfer between the prediction servers, wherein a prediction server among the prediction servers comprises a detection unit that detects a communication service protocol with respect to a sub-communication area for which the one prediction server is responsible, a prediction unit that comprises a first communication service quality, which predicts the future communication service quality of the one sub-communication area based on the communication service protocol, and a communication unit that sends a first prediction result, indicating the first communication service quality, to the mediation server and selectively sends a second prediction result.which is predicted by another prediction server among the prediction servers, receives from the mediation server, the prediction unit predicts a second communication service quality, which is a future communication service quality between a node within one sub-communication area and another node within another sub-communication area, for which the other prediction server is responsible, using at least one of the first prediction result or the second prediction result, and the server system further includes an instruction unit that instructs one node to change a communication setting if the second communication service quality is lower than a predetermined quality.

[0031] In the embodiment described above, “Server 10a, 10b, 10c” corresponds to an example of the “Prediction Server”, “Server 10b” corresponds to an example of the “Mediation Server”, “Protocol Processing Unit 11” corresponds to an example of the “Capture Unit”, “Communication Environment Prediction Unit 13” corresponds to an example of the “Prediction Unit”, “Communication Device 15” corresponds to an example of the “Communication Unit”, and “Network Configuration Formulation Unit 14” corresponds to an example of the “Instruction Unit”.

[0032] In one example of the server system, one node could be a vehicle with a communication function.

[0033] The present invention is not limited to the embodiment described above and can be modified at will in an area that does not deviate from the core or idea of ​​the invention as it is apparent from the claims and the entire description, and a server system associated with such a modification is also included in the technical scope of the present invention. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2009 - 278 421 A

[0002] Cited non-patent literature

[0000] Deo, Mohan M. Trivedi; Convolutional Social Pooling for Vehicle Trajectory Prediction, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2018, S. 1581-1589

[0016]

Images