Line control system, line control method, and program
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- NEC CORP
- Filing Date
- 2025-12-26
- Publication Date
- 2026-08-06
Smart Images
Figure JP2025045883_06082026_PF_FP_ABST
Abstract
Description
Circuit control system, circuit control method, and program
[0001] This disclosure relates to a circuit control system, a circuit control method, and a program.
[0002] As a related technology, Patent Document 1 discloses a mobile communication system. The mobile communication system described in Patent Document 1 includes a content server, a wireless base station, terminal equipment, and a control station. The control station calculates the Quality of Experience (QoE) based on the latency and traffic volume of past content distributions. The control station calculates the Quality of Experience for each area and each time period, and stores the calculated Quality of Experience in a knowledge database.
[0003] The control station receives content delivery requests from the terminal devices of users riding on a mobile vehicle, such as a train, that is traveling along a known route. For each of the multiple areas included in the travel route, the control station obtains the user's perceived quality at the time of arrival in each area. The control station identifies the locations and times along the travel route where the user's perceived quality meets predetermined criteria. The control station notifies the user of the terminal device of the time or location where the user's perceived quality meets the predetermined criteria.
[0004] As another related technology, Patent Document 2 discloses a media distribution system. In the media distribution system described in Patent Document 2, a transmitting device distributes media data such as video data to a receiving device via multiple lines. The transmitting device estimates the communication quality, such as available bandwidth or throughput, for each of the multiple lines. From the estimated communication quality, the transmitting device calculates the user-perceived quality when each line is used for distributing media data. Based on the user-perceived quality, the transmitting device determines the bitrate of the media data. Based on the estimated communication quality and the determined bitrate, the transmitting device distributes the amount of media data to the multiple lines.
[0005] Japanese Patent Publication No. 2014-183424, Japanese Patent No. 7136084
[0006] However, in Patent Document 1, although the user is notified of the time or place where the user's perceived quality meets a predetermined standard, the user cannot receive content delivery with a user-perceived quality that meets the predetermined standard until the notified time or place. Furthermore, Patent Document 2 does not consider the movement of the receiving device. In Patent Document 2, if the radio wave conditions deteriorate on any of the multiple lines to which media data is distributed, the user's perceived quality deteriorates.
[0007] One exemplary object of this disclosure is to provide a circuit control system, a circuit control method, and a program that can provide circuit switching to a mobile communication device while maintaining perceived quality.
[0008] A line control method according to a first aspect of the present disclosure includes, based on measurement data of communication quality including perceived quality information of a plurality of communication lines available to a communication device moving along a travel path, predicting for each of the plurality of communication lines a section in the travel path in which the perceived quality does not meet a predetermined standard as a low-quality section, predicting perceived quality information based on communication quality information measured by the communication device, and switching the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.
[0009] A program according to a second aspect of this disclosure predicts, based on communication quality measurement data including perceived quality information of multiple communication lines available to a communication device moving along a travel path, that for each of the multiple communication lines, a section in the travel path where the perceived quality does not meet a predetermined standard will be designated as a low-quality section, that perceived quality information will be predicted based on the communication quality information measured by the communication device, and that the computer will perform a process to switch the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.
[0010] A line control system according to a third aspect of the present disclosure includes: a first prediction means that predicts, for each of the plurality of communication lines, as a low-quality section in the travel path, a section in the travel path in which the perceived quality does not meet a predetermined standard, based on measurement data of communication quality including perceived quality information of a plurality of communication lines available to a communication device moving along a travel path; a second prediction means that predicts perceived quality information based on communication quality information measured by the communication device; and a line switching means that switches the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.
[0011] The line control system, line control method, and program related to this disclosure can provide line switching to a mobile communication device that maintains perceived quality.
[0012] This is a block diagram showing an example of the schematic configuration of the line control system relating to this disclosure. This is a flowchart showing the operating procedure of the line control system. This is a block diagram showing an example of the configuration of a network system including the line control system relating to this disclosure. This is a block diagram showing an example of the configuration of the line control system. This is a diagram showing an example of log data stored in the data storage unit. This is a diagram showing an example of the travel path of the mobile body 140. This is a graph showing the QoE of each point on the travel path of line A and line B. This is a diagram showing an example of the prediction result of the QoE degrade section. This is a diagram showing a comparison of the predicted QoE and the requested QoE for each line. This is a schematic diagram schematically showing the switching of communication lines. This is a flowchart showing the operating procedure of the line control system. This is a flowchart showing an example of the operating procedure of the line switching unit in step B4. This is a flowchart showing the operating procedure of the spatial QoE prediction unit. This is a block diagram showing the operating procedure of QoE prediction in the time-series QoE prediction unit. This is a block diagram showing an example of the configuration of a computer device.
[0013] Prior to describing embodiments of this disclosure, an overview of this disclosure will be provided. Figure 1 is a block diagram showing an example of a schematic configuration of a line control system according to this disclosure. The line control system 10 includes a first prediction unit 11, a second prediction unit 12, and a line switching unit 13. In this disclosure, "-unit" may also be read as "-means".
[0014] The first prediction unit 11 acquires communication quality measurement data, including perceived quality information of multiple communication lines available to the communication device moving along the travel path. Based on the measurement data, the first prediction unit 11 predicts for each of the multiple communication lines that, along the travel path, sections where the perceived quality does not meet a predetermined standard will be designated as low-quality sections.
[0015] The second prediction unit 12 predicts perceived quality information based on communication quality information measured by the communication device. The line switching unit 13 switches the communication line used by the communication device for communication based on the low-quality section predicted by the first prediction unit 11 and the perceived quality information predicted by the second prediction unit 12.
[0016] The operation procedure will now be explained. Figure 2 is a flowchart showing the operation procedure of the line control system 10. The operation procedure of the line control system 10 corresponds to the line control method. The first prediction unit 11 predicts, for each of the multiple communication lines, that a section in the travel path where the perceived quality does not meet a predetermined standard will be a low-quality section, based on measurement data of communication quality including perceived quality information of multiple communication lines (step A1). The second prediction unit 12 predicts the perceived quality information based on communication quality information measured by the communication device moving along the travel path (step A2). The line switching unit 13 switches the communication line used by the communication device for communication based on the low-quality section predicted by the first prediction unit 11 and the perceived quality information predicted by the second prediction unit 12 (step A3).
[0017] In this disclosure, the first prediction unit 11 predicts, from measurement data including perceived quality information, sections in the communication device's travel path where the perceived quality does not meet a predetermined standard as low-quality sections. The second prediction unit 12 predicts perceived quality information based on communication quality information measured in the communication device. The line switching unit 13 switches the communication line used by the communication device for communication based on the low-quality sections predicted by the first prediction unit 11 and the perceived quality information predicted by the second prediction unit 12. In this disclosure, communication lines can be switched using predictions of perceived quality information along the path and predictions of perceived quality information in the communication device. By using not only the low-quality sections predicted along the path but also the perceived quality information predicted by the second prediction unit 12, it is possible to achieve line switching that maintains perceived quality that meets a predetermined standard for a moving communication device.
[0018] The embodiments of this disclosure will be described in detail below with reference to the drawings. Note that the following description and drawings have been omitted and simplified as appropriate for clarity of explanation. Furthermore, in the following drawings, the same elements and similar elements are denoted by the same reference numerals, and redundant explanations have been omitted where necessary.
[0019] Figure 3 is a block diagram showing an example configuration of a network system including a line control system according to the present disclosure. The first embodiment will be described using Figure 3. The network system 100 shown in Figure 3 includes a line control system 110, a plurality of communication lines 120A and 120B, a communication device 130, and a remote device 150.
[0020] Although Figure 3 shows two communication lines 120A and 120B, the number of communication lines available in the network system 100 is not limited to two. The network system 100 may include three or more communication lines. In the following description, communication line 120A will also be called line A, and communication line 120B will also be called line B. Furthermore, unless otherwise necessary, communication lines 120A and 120B will also be called communication line 120.
[0021] The following embodiments describe an example in which the communication control system 110 is applied to the remote monitoring or remote control application of a mobile body 140. The mobile body 140 includes aerial mobility such as a drone, a railway train, or a road vehicle. The mobile body 140 may be configured to operate autonomously. The mobile body 140 may also be a transport robot that moves within a factory, such as an Automatic Guided Vehicle (AGV). The mobile body 140 moves along a travel path.
[0022] The remote control device 150 is a device used for remote monitoring or remote control of the mobile body 140. The communication device 130 is a wireless communication device mounted on the mobile body 140. Alternatively, the communication device 130 may be a wireless communication device carried by a person riding in the mobile body 140.
[0023] Each of the multiple communication lines 120 is a wireless communication network that the communication device 130 can use for communication. The multiple communication lines 120 include, for example, multiple wireless communication networks of different operators or types. Each communication line 120 includes, for example, a mobile communication network, a satellite radio network, and a radio network other than the mobile communication network and the satellite radio network. Communication line 120A and communication line 120B may be the same type of wireless communication network or different types of wireless communication networks. For example, communication line 120A may be a mobile communication network operated by communication carrier A, and communication line 120B may be a mobile communication network operated by another communication carrier B. Alternatively, communication line 120A may be a mobile communication network, and communication line 120B may be a satellite communication network.
[0024] The communication device 130 transmits information about the mobile unit 140 to the remote device 150 via at least one of the communication lines 120A and 120B. The communication device 130 transmits information such as video information and vehicle sensor information acquired by the mobile unit 140 to the remote device 150. The communication device 130 can also receive information such as control information from the remote device 150 via at least one of the communication lines 120A and 120B.
[0025] The line control system 110 selects the communication line to be used for communication between the communication device 130 and the remote device 150. In one embodiment, the line control system 110 is configured to switch the communication line used by the communication device 130 for communication between communication line 120A and communication line B while the mobile body 140 is moving along its travel path.
[0026] Figure 4 is a block diagram showing an example configuration of the line control system 110. The line control system 110 includes a context data filter unit 111, a spatial QoE prediction unit 112, a time-series QoE prediction unit 113, and a line switching unit 114. Physically, the line control system 110 can be configured as a computer device having, for example, one or more processors and one or more memories. At least a portion of the functions of each part in the line control system 110 can be realized by the processor executing processing according to instructions read from memory. The line control system 110 corresponds to the line control system 10 shown in Figure 1.
[0027] The data storage unit 250 stores various information, including log data or measurement data from communication lines 120A and 120B. Figure 5 shows an example of log data stored in the data storage unit 250. In the example in Figure 5, the log data includes time information, movement information, wireless type, cell information, communication information, QoE information, and application data. The time information indicates the date and time of communication. The movement information includes location information, including the latitude and longitude of the moving object, and information regarding the speed of movement. The wireless type indicates the type of wireless communication used for communication, such as Wi-Fi, 4th generation (4G), and satellite.
[0028] Communication information includes data size, session time, throughput, address information, and port number. QoE information, i.e., QoE data, indicates QE such as Web QoE, video QoE, video call QoE, online meeting QoE, and Tele-operated driving QoE. QoE is calculated based on communication quality. The technology for converting communication quality to QoE is not limited to any particular technology. For example, QoE is calculated using a standardized model. QoE may also be a Mean Opinion Score (MOS), a five-point rating scale from 1 to 5. Application data indicates information transmitted and received over the wireless network, such as sensor information or video information. The data storage unit 250 stores log data for each of the communication lines 120A and 120B.
[0029] The travel route information 220 includes a list of location information for each point included in the travel route of the mobile body 140. Figure 6 shows an example of the travel route of the mobile body 140. In Figure 6, the vertical axis represents latitude and the horizontal axis represents longitude. The numbers shown in Figure 6 represent the position on the travel route. The starting point of the travel route is represented by 0 and the ending point is represented by 1. In the example in Figure 6, an example is shown where the starting point and ending point of the travel route are the same location. For example, the travel route information 220 includes a list of latitude and longitude information for each point included in the travel route.
[0030] The context data filter unit 111 uses the travel route information 220 to filter the log data or measurement data according to filter conditions corresponding to the movement of the mobile body 140, and generates context data from the log data or measurement data. In other words, the context data filter unit 111 uses the travel route information 220 to extract log data or measurement data on the travel route of the mobile body 140 from the log data or measurement data stored in the data storage unit 250.
[0031] The context data filter unit 111 may, for example, filter log data based on conditions such as travel time, location information such as latitude and longitude, and travel speed, and extract log data that meets the conditions as context data. The context data filter unit 111 may also filter log data based on surrounding environment information such as congested traffic, many trees, or being in an urban area, and extract log data in the same or similar surrounding environment as context data. The context data filter unit 111 may also determine the surrounding environment based on video information or sensor information included in the log data.
[0032] The spatial QoE prediction unit 112 predicts QoE degradation sections based on QoE statistics at each point along the travel path of the mobile body 140. Here, a QoE degradation section refers to, for example, a section along the travel path of the mobile body 140 where there is a high probability that a QoE above a predetermined standard cannot be obtained. The spatial QoE prediction unit 112 calculates QoE statistics included in the log data extracted by the context data filter unit 111, for example, and predicts QoE degradation sections based on the calculated statistics. The spatial QoE prediction unit 112 corresponds to the first prediction unit 11 shown in Figure 1.
[0033] Figure 7 is a graph showing the QoE at each point along the travel paths of lines A and B. In Figure 7, the vertical axis represents QoE, and the horizontal axis represents the position on the travel path. For line A, the spatial QoE prediction unit 112 predicts the QoE degrade section based on the QoE statistics of line A at each point along the travel path. Similarly, for line B, the spatial QoE prediction unit 112 predicts the QoE degrade section based on the QoE statistics of line B at each point along the travel path.
[0034] Figure 8 shows an example of the prediction results for QoE degrade intervals. In Figure 8, the horizontal direction indicates the position on the travel path. The spatial QoE prediction unit 112 predicts, for each of line A and line B, that intervals where the QoE statistics do not meet a predetermined standard are QoE degrade intervals. In Figure 8, intervals indicated as "Low" correspond to QoE degrade intervals. Intervals indicated as "High" correspond to intervals that are not QoE degrade intervals, for example, intervals where the QoE statistics are above a predetermined standard.
[0035] The time-series QoE prediction unit 113 predicts the QoE after a predetermined time based on the communication quality in the communication device 130 for each communication line. For example, the time-series QoE prediction unit 113 calculates the predicted QoE after a predetermined time in real time from the real-time measurement results of the communication quality. The time-series QoE prediction unit 113 acquires, for example, the measurement results of the communication quality within a predetermined time period in the past. The time-series QoE prediction unit 113 predicts the future communication quality from the acquired measurement results. Known techniques can be used to predict the communication quality short time in the future from past communication quality measurement results. The time-series QoE prediction unit 113 calculates the predicted QoE after a predetermined time from the predicted communication quality.
[0036] The time-series QoE prediction unit 113 compares the predicted QoE with the required QoE for each line. The required QoE indicates, for example, the QoE required for remote monitoring or remote control of the mobile unit 140 in the network system 100. If the predicted QoE for a communication line is lower than the required QoE, the time-series QoE prediction unit 113 determines that a QoE decrease will occur in that communication line. If the time-series QoE prediction unit 113 determines that a QoE decrease will occur in a communication line, it predicts the section from the position of the mobile unit 140 after a predetermined time to a predetermined section length as the QoE decrease section for that line. The time-series QoE prediction unit 113 corresponds to the second prediction unit 12 shown in Figure 1.
[0037] FIG. 9 is a diagram showing a comparison between the predicted QoE for each line and the required QoE. The time-series QoE prediction unit 113 predicts the QoE, for example, T seconds after the current time for each of line A and line B. In the example of FIG. 9, the QoE predicted by the time-series QoE prediction unit 113 for line A is lower than the required QoE. On the other hand, the QoE predicted by the time-series QoE prediction unit 113 for line B is higher than the required QoE. In the case of the example of FIG. 9, the time-series QoE prediction unit 113 determines that the QoE of line A will decrease after T seconds. The time-series QoE prediction unit 113 predicts a section from the position of the mobile body 140 after T seconds to a position, for example, 500 m ahead as a QoE decrease section.
[0038] The line switching unit 114 determines whether to switch the communication line used by the communication device 130 for communication based on the prediction result of the spatial QoE prediction unit 112 and the prediction result of the time-series QoE prediction unit 113. When a QoE decrease section is predicted in the spatial QoE prediction unit 112 or the time-series QoE prediction unit 113, the line switching unit 114 determines to switch the communication line used by the communication device 130.
[0039] When the line switching unit 114 determines to switch the communication line, it switches the communication line used by the communication device 130 for communication with the remote device 150 in the network system 100. Examples of functions for switching the communication line include, for example, Multipath Transmission Control Protocol (MPTCP) or Quick User Datagram Protocol (UDP) Internet Connections (MPQUIC) in the transport layer / session layer. Also, as an example of a function for switching the communication line, there is IP routing in the Internet Protocol (IP) layer. The function used for switching the communication line is not limited to the above-described functions, and techniques that utilize multipath or multiple lines other than the above-described functions may be used. The line switching unit 114 corresponds to the line switching unit 13 shown in FIG. 1.
[0040] FIG. 10 is a schematic diagram schematically showing the switching of a communication line. The geographical prediction shown in FIG. 10 corresponds to the prediction in the spatial QoE prediction unit 112. The prediction after a predetermined time corresponds to the prediction in the time-series QoE prediction unit 113. At the start point of the travel route, it is assumed that the line switching unit 114 has selected line A.
[0041] The time-series QoE prediction unit 113 predicts the QoE after a predetermined time from the communication quality in the communication device 130 while the mobile body 140 is moving. The time-series QoE prediction unit 113 predicts that the QoE will decrease on line A after T seconds at the current position. Also, the time-series QoE prediction unit 113 predicts that the QoE on line B after T seconds is above a predetermined standard. In response to the prediction of the QoE decrease section on line A, the line switching unit 114 determines to switch the communication line used by the communication device 130 from line A to line B. The line switching unit 114 switches the communication line used by the communication device 130 from line A to line B.
[0042] If only geographical prediction is used for line switching, the switching from line A to line B is not performed until the mobile body enters the QoE decrease section predicted by the spatial QoE prediction unit 112 on line A. In one embodiment, when a QoE decrease section is predicted after T seconds in the prediction after a predetermined time, the line switching unit 114 determines to switch the communication line before the mobile body 140 enters the QoE decrease section predicted in the geographical prediction. In one embodiment, by combining the geographical prediction and the prediction after a predetermined time, when congestion occurs on line A, it is possible to avoid a decrease in QoE due to the congestion.
[0043] The operation procedure will be described. FIG. 11 is a flowchart showing the operation procedure of the line control system 110. The context data filter unit 111 generates context data from the log data or measurement data stored in the data storage unit 250 using the travel route information 220 (step B1). The spatial QoE prediction unit 112 predicts the spatial QoE based on the QoE statistic at each point on the travel route of the mobile body 140 (step B2). In step B2, the spatial QoE prediction unit 112 predicts, for example, a QoE decrease section.
[0044] The time-series QoE prediction unit 113 predicts the QoE after a predetermined time based on the communication quality at the communication device 130 for each communication line (step B3). For example, in step B3, the time-series QoE prediction unit 113 calculates the predicted QoE after a predetermined time in real time from the real-time communication quality measurement results. The line switching unit 114 determines whether or not to perform line switching based on the spatial QoE predicted in step B2 and the QoE based on the communication quality measurement predicted in step B3 (step B4).
[0045] Figure 12 is a flowchart showing an example of the operation procedure of the line switching unit 114 in step B4. The line switching unit 114 determines in spatial QoE prediction whether the position of the moving object a predetermined time, for example, T seconds after the current time, is predicted to be in a QoE degrade zone on the currently used communication line (step C1). If the line switching unit 114 determines in step C1 that the position after the predetermined time is not in a QoE degrade zone, it determines in the time-series QoE prediction unit 113 whether a QoE degrade is predicted for the currently used communication line (step C2). For example, if the QoE predicted by the time-series QoE prediction unit 113 on the currently used line A is lower than the requested QoE, the line switching unit 114 determines that a QoE degrade is predicted. If the line switching unit 114 determines in step C2 that a QoE degrade is not predicted, it terminates the process without performing a line switch.
[0046] If the line switching unit 114 determines in step C1 that the position of the moving object after a predetermined time is not in a QoE degrade zone, or if it determines in step C2 that no QoE degrade is predicted, it determines the line to switch to (step C3). In step C3, the line switching unit 114 determines the communication line to switch to, for example, the communication line with the longest distance from the current position that does not fall within a QoE degrade zone, based on the spatial QoE predicted by the spatial QoE prediction unit 112. The line switching unit 114 transmits a switching instruction to the function unit in the network system 100 that switches communication lines (step C4).
[0047] In the first embodiment, the spatial QoE prediction unit 112 predicts QoE degrade sections along the travel path based on QoE statistics. The time-series QoE prediction unit 113 predicts the QoE after a predetermined time in real time, for example, by real-time communication quality measurement. The line switching unit 114 combines the prediction of QoE degrade sections with the real-time QoE prediction to determine whether or not to switch the communication line used by the communication device 130 for communication. The line switching unit 114 can switch the communication line at an appropriate timing according to the prediction results of the two QoE degrade sections along the travel path of the communication device 130. Therefore, it is possible to perform line switching for the moving communication device 130 that maintains a perceived quality that meets a predetermined standard.
[0048] Next, a second embodiment will be described. In the second embodiment, the spatial QoE prediction unit 112 divides the travel path in stages and predicts a QoE degrade section for each divided section. In predicting the QoE degrade section, the spatial QoE prediction unit 112 generates a probability distribution of QoE for each section. Here, a probability distribution refers to a mathematical representation of the probability of a specific event (for example, a certain QoE value) occurring. The probability distribution of QoE for each section is generated from the QoE information for each section. The probability distribution of QoE for each section indicates what kind of QoE is possible to obtain in a given section. The probability distribution generated from the QoE information for each section makes it possible to statistically compare and evaluate the performance of the communication line in each section. In each section, the probability distribution of QoE can be represented using a predetermined distribution, for example, a mixed normal distribution.
[0049] The spatial QoE prediction unit 112 divides the target section into multiple sections by gradually dividing the target section until the fit of a predetermined probability distribution to the probability distribution of QoE included in the measurement data of the target section no longer improves. For each divided section, the spatial QoE prediction unit 112 compares the QoE in that section with a predetermined standard QoE and determines whether the section is experiencing a QoE decrease.
[0050] Figure 13 is a flowchart showing the operation procedure of the spatial QoE prediction unit 112. The spatial QoE prediction unit 112 first sets the entire route as the analysis target section (step D1). The spatial QoE prediction unit 112 determines whether there are any unprocessed analysis target sections (step D2). If the spatial QoE prediction unit 112 determines that there are any unprocessed analysis target sections, it selects one analysis target section.
[0051] The selected interval is represented by interval a. The spatial QoE prediction unit 112 approximates the probability distribution of QoE in interval a with, for example, a mixture of normal distributions (step D3). In step D3, the probability distribution of x = QoE is defined by, for example, a mixture of normal distributions represented by the following formula. Let G = 9 for the mixture of normal distributions, and the mean of each normal distribution be μ ｉ The value of the parameter θ is fixed at = 1 + 0.5(i-1). Here, i = 1, 2, ... G (= 9). In step D3, the spatial QoE prediction unit 112 updates the value of the parameter θ to maximize the log-likelihood logL(θ|x) = f(x|θ).
[0052] The spatial QoE prediction unit 112 classifies the analysis target interval into two intervals (step D4). In step D4, the spatial QoE prediction unit 112 divides interval a into infinitesimal intervals of a predetermined length and calculates feature quantities for each infinitesimal interval. For example, if the mean QoE in an infinitesimal interval is less than 3, the spatial QoE prediction unit 112 sets the feature quantity to 0. If the mean QoE in an infinitesimal interval is 3 or greater, the spatial QoE prediction unit 112 sets the feature quantity to 1. The feature quantities may be calculated based on the minimum value, standard deviation, and maximum value of QoE in an infinitesimal interval. Based on the feature quantities of the infinitesimal intervals, the spatial QoE prediction unit 112 classifies interval a into interval a1 and interval a2 using a classification algorithm such as the K-means method.
[0053] The spatial QoE prediction unit 112 approximates the probability distribution of QoE in section a1 with, for example, a mixture normal distribution (step D5). In step D5, the spatial QoE prediction unit 112 updates the value of the parameter θ of the mixture normal distribution so as to maximize the log-likelihood. Also, the spatial QoE prediction unit 112 approximates the probability distribution of QoE in section a2 with, for example, a mixture normal distribution (step D6). In step D6, the spatial QoE prediction unit 112 updates the value of the parameter θ of the mixture normal distribution so as to maximize the log-likelihood.
[0054] The spatial QoE prediction unit 112 calculates the fitting degrees of the mixture normal distributions approximated in steps D3, D5, and D6, respectively, and determines whether the fitting degrees of spaces a1 and a2 have improved with respect to the fitting degree of space a (step D7). For the fitting degree, for example, Akaike Information Criteria (AIC) represented by the following formula can be used. AIC is an index indicating that the smaller the value, the better the probability model fits the original data.
[0055] In step D7, the spatial QoE prediction unit 112 evaluates how much (how much smaller) the fitting degree (AIC ａ ) of section a1 has improved with respect to the fitting degree (AIC ａ１ ) of section a. Also, the spatial QoE prediction unit 112 evaluates how much the fitting degree (AIC ａ ) of section a2 has improved with respect to the fitting degree (AIC ａ２ ) of section a. The spatial QoE prediction unit 112 determines that the fitting degree has improved when AIC ａ１ is better than AIC ａ by a predetermined threshold or more, and AIC ａ２ is better than AIC ａ by a predetermined threshold or more.
[0056] If the spatial QoE prediction unit 112 determines in step D7 that the fit has improved, it divides the segment to be divided into segment a1 and segment a2 (step D8). The spatial QoE prediction unit 112 adds the divided segments a1 and a2 to the segment to be divided. The spatial QoE prediction unit 112 returns to step D2 and determines whether there are any unprocessed segments to be divided. The spatial QoE prediction unit 112 selects either segment a1 or segment a2, which are unprocessed segments to be divided, and determines whether to further divide the selected segment to be divided.
[0057] If it is determined in step D7 that the fit does not improve, section a, which is the section to be divided, is not divided. If it is determined in step D7 that the fit does not improve, the spatial QoE prediction unit 112 registers section a as the section after division. The spatial QoE prediction unit 112 returns to step D2 and determines whether there are any unprocessed sections to be divided. The spatial QoE prediction unit 112 repeatedly performs the processes from step D3 to step D8 until it is determined in step D2 that there are no unprocessed sections to be divided, thereby dividing the travel route into multiple sections in stages.
[0058] The spatial QoE prediction unit 112 divides the travel route into multiple sections, and then determines whether the QoE of each of the divided sections meets a predetermined standard. The spatial QoE prediction unit 112 predicts the sections among the divided sections in which the QoE does not meet the predetermined standard as QoE degradation sections.
[0059] If the spatial QoE prediction unit 112 predicts a QoE degrade interval using a fixed interval, the prediction accuracy of the QoE degrade interval is likely to change depending on the length of the fixed interval. In the second embodiment, the spatial QoE prediction unit 112 can adaptively determine the interval length. Therefore, the prediction accuracy of the QoE degrade interval can be improved compared to the case where the QoE degrade interval is predicted using a fixed interval length.
[0060] Next, a third embodiment will be described. In the third embodiment, the time-series QoE prediction unit 113 changes the communication quality information calculated from the communication data according to the characteristics of the communication data of the communication device 130 used for calculating or measuring communication quality. For example, the time-series QoE prediction unit 113 decides, according to the data characteristics, whether to calculate first communication quality information related to data transfer rate from the communication data, or second communication quality information related to communication delay from the communication data.
[0061] Figure 14 is a flowchart showing the operation procedure of QoE prediction in the time-series QoE prediction unit 113. The time-series QoE prediction unit 113 identifies the communication line used for the communication data of the communication device 130 (step E1). Based on address information such as an IP address, the time-series QoE prediction unit 113 identifies which communication line the communication data was transmitted and received through.
[0062] The time-series QoE prediction unit 113 determines whether the data size is greater than or equal to a predetermined data size A (step E2). If the data size is greater than or equal to A, the time-series QoE prediction unit 113 decides to calculate first communication quality information related to the data transfer rate, such as throughput. If the time-series QoE prediction unit 113 determines in step E2 that the data size is greater than or equal to A, it calculates the throughput in the communication line identified in step E1 from the communication data of data size A or greater (step E3).
[0063] If the data size is determined to be less than A in step E2, the time-series QoE prediction unit 113 determines whether the number of active sessions is greater than 1 (step E4). If the time-series QoE prediction unit 113 determines in step E3 that the number of active sessions is greater than 1, it merges multiple active sessions (step E5). After merging the active sessions, the time-series QoE prediction unit 113 proceeds to step E3 and calculates the throughput on the communication line identified in step E1 from the communication data of the merged active sessions.
[0064] If the number of active sessions is determined to be 1 or less in step E4, the time-series QoE prediction unit 113 calculates a second communication quality information related to delay, such as Turn Around Time (TAT) (step E6). In step E6, the time-series QoE prediction unit 113 calculates the TAT for the communication line identified in step E1 from the communication data.
[0065] The time-series QoE prediction unit 113 calculates at least one of throughput and TAT for each communication line. The time-series QoE prediction unit 113 predicts the QoE after a predetermined time using at least one of the calculated throughput and TAT for each communication line (step E7). The time-series QoE prediction unit 113 calculates, for example, past T ｂｅｆｏｒｅ Communication quality information measured per second Using a time-series prediction model that predicts the QoE after a predetermined time from past communication quality information, the QoE after T seconds ｔ＋Ｔ To predict. Time series forecasting models may include autoregressive forecasting models, state-space models, or machine learning models. Standard conversions may be used to convert communication quality information to QoE.
[0066] For example, if the communication line currently in use is line A, the time-series QoE prediction unit 113 can calculate communication quality information for line A from the communication data transmitted and received by the communication device 130 via line A. If line A is selected as the communication line to be used, the communication device 130 may transmit measurement data to other lines that are not selected. If another communication line, line B, is available at a fixed rate that does not depend on the amount of data, there will be no increase in usage charges for line B due to the transmission of measurement data. For communication lines that are not selected, the time-series QoE prediction unit 113 can calculate communication quality information based on the measurement data.
[0067] If the communication line is billed on a pay-per-use basis, and if measurements are to be taken as close to real-time as possible, such as in delay-sensitive applications like remote robot control, the communication device 130 may transmit a portion of the data using other communication lines that have not been selected. For example, if line A is selected as the communication line to be used, the communication device 130 may transmit a portion of the data via line B. The time-series QoE prediction unit 113 can calculate communication quality information for the unselected communication lines based on the portion of data transmitted.
[0068] If the communication line is billed on a pay-per-use basis and the application is relatively tolerant of latency, such as video, the time-series QoE prediction unit 113 does not need to calculate communication quality information in real time for communication lines that are not selected. For example, the time-series QoE prediction unit 113 calculates the average value of communication quality information for each communication line. For example, if line A is in use, the time-series QoE prediction unit 113 may use the average value of communication quality information calculated for line B to predict the QoE after a predetermined time. When the average value of communication quality information is used, the communication quality information for communication lines not in use becomes a fixed value.
[0069] In the third embodiment, the time-series QoE prediction unit 113 changes the quality information to be calculated according to the characteristics of the data used to calculate the communication quality information. The time-series QoE prediction unit 113 can calculate communication quality information according to the data characteristics. In the third embodiment, it is considered that the accuracy of QoE prediction based on communication quality information can be improved compared to when the communication quality information calculated according to the data characteristics is not changed.
[0070] In the above embodiment, an example was described in which the spatial QoE predicted by the spatial QoE prediction unit 112 and the QoE predicted by the time-series QoE prediction unit 113 based on communication quality are used for circuit switching. However, this disclosure is not limited to this. Spatial QoE and QoE based on communication quality can be used for simulation of base station design. Currently, base station design is carried out on a radio wave intensity basis. By utilizing spatial QoE and QoE based on communication quality in base station design and simulating from a communication quality perspective, it becomes possible to determine the sufficiency of communication quality geographically. Therefore, it becomes possible to optimize the investment costs of base stations for existing telecommunications carriers or private network operators such as local 5G (5th generation).
[0071] Next, the physical configuration of the line control system 110 will be described. Figure 15 is a block diagram showing an example configuration of a computer device that can be used as a line control system 110. The computer device 500 has a processor 510 such as a Central Processing Unit (CPU), a storage unit 520, a Read Only Memory (ROM) 530, a Random Access Memory (RAM) 540, a communication interface (Interface: IF) 550, and a user interface 560.
[0072] The communication interface 550 is an interface for connecting the computer device 500 to a communication network via wired communication means or wireless communication means. The user interface 560 includes a display unit, such as a display. The user interface 560 also includes input units such as a keyboard, mouse, and touch panel.
[0073] The memory unit 520 is an auxiliary storage device capable of holding various types of data. The memory unit 520 does not necessarily have to be part of the computer device 500; it may be an external storage device or cloud storage connected to the computer device 500 via a network.
[0074] ROM 530 is a non-volatile memory device. For example, a semiconductor memory device such as a relatively small-capacity flash memory is used for ROM 530. The program executed by the processor 510 can be stored in the storage unit 520 or ROM 530. The storage unit 520 or ROM 530 stores various programs that realize the functions of each part of the line control system 110.
[0075] The above program, when loaded into a computer, includes a set of instructions (or software code) for causing the computer to perform one or more of the functions described in the embodiments. The program may be stored in a non-temporary computer-readable medium or a physical storage medium. Examples, but not limited to, include memory technologies such as RAM, ROM, flash memory, or solid-state drives (SSDs), optical disc storage such as Compact Disc (CD), digital versatile disc (DVD), or Blu-ray®, and magnetic storage devices such as magnetic cassettes, magnetic tapes, or magnetic disk storage. The program may be transmitted over a temporary computer-readable medium or a communication medium. Examples, but not limited to, include transmitting signals in electrical, optical, acoustic, or non-electrical, optical, and non-acoustic forms.
[0076] The RAM 540 is a volatile memory device. Various semiconductor memory devices such as Dynamic Random Access Memory (DRAM) or Static Random Access Memory (SRAM) can be used for the RAM 540. The RAM 540 may be used as an internal buffer for temporarily storing data, etc. The processor 510 loads a program stored in the memory unit 520 or ROM 530 into the RAM 540 and executes it. By executing the program, the functions of each part of the line control system 110 can be realized. The processor 510 may have an internal buffer that can temporarily store data, etc.
[0077] In this disclosure, the line control system 110 does not necessarily have to be configured as a single computer device. The line control system 110 may be configured using multiple physically separated devices. The functions of the line control system 110 may be distributed across an application server, a Mobile / Multi-access Edge Computing (MEC) server, or an in-vehicle device. At least a portion of the functions of the line control system 110 may be located in a mobile communication network or a mobile device 140.
[0078] Although the present disclosure has been described above with reference to embodiments, the present disclosure is not limited to the embodiments described above. Various modifications to the structure and details of the present disclosure can be made as can be understood by those skilled in the art within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.
[0079] Each drawing is merely illustrative to illustrate one or more embodiments. Each drawing may be associated with one or more other embodiments, rather than being associated with only one specific embodiment. As those skilled in the art will understand, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings, for example, to create embodiments not explicitly shown or described. Not all features or steps shown in any one drawing to illustrate an exemplary embodiment are necessarily required, and some features or steps may be omitted. The order of steps described in any of the drawings may be changed as appropriate.
[0080] Some or all of the above embodiments may also be described as follows, but are not limited to the following:
[0081] [Note 1] A communication device moving along a travel path predicts, based on measurement data of communication quality including perceived quality information of multiple available communication lines, that for each of the multiple communication lines, a section in the travel path where the perceived quality does not meet a predetermined standard will be designated as a low-quality section; predicts perceived quality information based on communication quality information measured by the communication device; and switches the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.
[0082] [Note 2] Predicting the perceived quality information based on the communication quality information includes predicting the perceived quality information after a predetermined time period for each communication line, based on past communication quality information, as described in Note 1.
[0083] [Note 3] The line control method according to Note 1 or 2, wherein predicting the low-quality section includes dividing the section to be divided in stages until the fit of a predetermined probability distribution to the probability distribution of the perceived quality information contained in the measurement data in the section to be divided in the travel path no longer improves, and calculating the perceived quality information for each divided section.
[0084] [Appendix 4] The line control method according to Appendix 3, wherein predicting the low-quality section includes comparing the calculated perceived quality information with the predetermined standard for each divided section.
[0085] [Note 5] The communication quality information includes first communication quality information relating to data transfer speed and second communication quality information relating to communication delay, and predicting the perceived quality information based on the communication quality information includes calculating the first communication quality information or the second communication quality information from the communication data based on the size of the communication data in the communication device, as described in any one of Notes 1 to 4, for the line control method.
[0086] [Note 6] The line control method described in Note 5, which calculates the first communication quality information when the size of the communication data is greater than or equal to a predetermined size.
[0087] [Note 7] The line control method according to Note 5 or 6, which calculates the second communication quality information when the size of the communication data is less than a predetermined size and the number of active sessions is one.
[0088] [Note 8] A line control method according to any one of Notes 5 to 7, wherein if the size of the communication data is less than a predetermined size and the number of active sessions is greater than one, the communication data of multiple sessions is integrated, and the first communication quality information is calculated from the integrated communication data of multiple sessions.
[0089] [Note 9] A program that causes a computer to perform a process of switching the communication line used by the communication device for communication based on measurement data of communication quality including perceived quality information of multiple communication lines available to a communication device moving along a travel path, predicting for each of the multiple communication lines that the perceived quality does not meet a predetermined standard in the travel path as a low-quality section, predicting perceived quality information based on communication quality information measured by the communication device, and based on the predicted low-quality section and the predicted perceived quality information.
[0090] [Note 10] A line control system comprising: a first prediction means that predicts, for each of the multiple communication lines, a section in the travel path in which the perceived quality does not meet a predetermined standard will be a low-quality section, based on measurement data of communication quality including perceived quality information of multiple communication lines available to a communication device moving along a travel path; a second prediction means that predicts perceived quality information based on communication quality information measured by the communication device; and a line switching means that switches the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.
[0091] Some or all of the elements (e.g., configuration and function) described in Appendices 2 to 8 that are dependent on Appendice 1 may also be dependent on Appendices 9 and 10 in the same way as those described in Appendices 2 to 8. Some or all of the elements described in any appendice may be applied to various hardware, software, recording means, systems, and methods for recording software.
[0092] This application claims priority based on Japanese Patent Application No. 2025-015271, filed on 31 January 2025, and incorporates all of its disclosures herein.
[0093] 10: Line control system 11: First prediction unit 12: Second prediction unit 13: Line switching unit 100: Network system 110: Line control system 111: Context data filter unit 112: Spatial QoE prediction unit 113: Time series QoE prediction unit 114: Line switching unit 120A, 120B: Communication line 130: Communication device 140: Mobile device 150: Remote device 220: Travel route information 500: Computer device 510: Processor 520: Memory unit 530: ROM 540: RAM 550: Communication IF 560: User IF
Claims
1. A communication device moving along a travel path predicts, based on communication quality measurement data including perceived quality information of multiple available communication lines, that for each of the multiple communication lines, a section in the travel path where the perceived quality does not meet a predetermined standard will be designated as a low-quality section; predicts perceived quality information based on communication quality information measured by the communication device; and switches the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.
2. The line control method according to claim 1, wherein predicting the perceived quality information based on the communication quality information includes predicting the perceived quality information after a predetermined time period based on past communication quality information for each communication line.
3. The circuit control method according to claim 1 or 2, wherein predicting the low-quality section includes dividing the section to be divided in stages until the fit of a predetermined probability distribution to the probability distribution of the perceived quality information contained in the measurement data in the section to be divided in the travel path no longer improves, and calculating the perceived quality information for each divided section.
4. The circuit control method according to claim 3, wherein predicting the low-quality section includes comparing the calculated perceived quality information with a predetermined standard for each divided section.
5. The communication quality information includes first communication quality information relating to data transfer speed and second communication quality information relating to communication delay, and predicting the perceived quality information based on the communication quality information includes calculating the first communication quality information or the second communication quality information from the communication data based on the size of the communication data in the communication device, according to any one of claims 1 to 4.
6. The line control method according to claim 5, wherein the size of the communication data is greater than or equal to a predetermined size, and the first communication quality information is calculated.
7. The line control method according to claim 5 or 6, wherein the size of the communication data is less than a predetermined size and the number of active sessions is one, and the second communication quality information is calculated.
8. A line control method according to any one of claims 5 to 7, wherein if the size of the communication data is less than a predetermined size and the number of active sessions is greater than one, the communication data of multiple sessions is integrated, and the first communication quality information is calculated from the integrated communication data of multiple sessions.
9. A program that causes a computer to perform a process of switching the communication line used by the communication device for communication based on communication quality measurement data including perceived quality information of multiple communication lines available to a communication device moving along a travel path, predicting for each of the multiple communication lines that does not meet a predetermined standard in the travel path as a low-quality section, predicting perceived quality information based on communication quality information measured by the communication device, and based on the predicted low-quality section and the predicted perceived quality information.
10. A line control system comprising: a first prediction means that predicts, for each of the multiple communication lines, a section in the travel path in which the perceived quality does not meet a predetermined standard, as a low-quality section, based on measurement data of communication quality including perceived quality information of multiple communication lines available to a communication device moving along a travel path; a second prediction means that predicts perceived quality information based on communication quality information measured by the communication device; and a line switching means that switches the communication line used by the communication device for communication based on the predicted low-quality section and the predicted perceived quality information.