WO2020070861A1

WO2020070861A1 - Transmission system, transmission device, and program

Info

Publication number: WO2020070861A1
Application number: PCT/JP2018/037212
Authority: WO
Inventors: 悠紀永江
Original assignee: 株式会社アプトポッド
Priority date: 2018-10-04
Filing date: 2018-10-04
Publication date: 2020-04-09

Abstract

According to the present invention, a data acquisition unit acquires multiple pieces of sensor data outputted from multiple sensors. A storage unit stores therein multiple methods for reducing the data amount of at least one of the multiple pieces of sensor data. A situation acquisition unit acquires situation information indicating the situation of communication between a server device and a client device. A setting unit sets a target transmission rate in accordance with the situation information. A determination unit determines planned transmission rates for the multiple pieces of sensor data on the basis of the data amount per unit time of each of the multiple pieces of sensor data. An adjustment unit makes adjustment to the data amount of at least one of the multiple pieces of sensor data, by using at least one of the multiple methods, so as to bring the planned transmission rate to be equal to or lower than the target transmission rate. A transmission unit transmits, to the server device, the multiple pieces of sensor data that have undergone the adjustment.

Description

Transmission system, transmission device, and program

<< The present invention relates to a technique for transmitting a plurality of sensor data.

技術 Techniques for reducing the amount of data to be transmitted according to communication conditions are known. For example, Patent Literature 1 describes that a wireless transmission band on a wireless LAN (Local Area Network) is estimated, a video coding rate of a codec is compared with a wireless transmission band, and a data compression ratio is varied. Have been. Patent Literature 2 describes that a data compression ratio is controlled based on a transmission state in a communication unit.

JP 2010-258850 A JP 2007-158786 A

When collecting a plurality of sensor data output from a plurality of sensors in real time, these sensor data may be transmitted together from a client device to a server device. At this time, if the communication situation is not sufficient, all data will be equally delayed. In each of the techniques described in

Patent Documents

1 and 2, a method of changing the compression ratio of a single data in accordance with the communication situation is proposed. For example, the amount of data reduction may be different for each sensor data. In some cases, it cannot be used for the purpose of efficiently and flexibly reducing the data amount and delay of a plurality of sensor data.
An object of the present invention is to reduce data delay in transmission of a plurality of sensor data.

The present invention provides a data acquisition unit that acquires a plurality of sensor data output from a plurality of sensors, and a storage unit that stores a plurality of methods for reducing a data amount of at least one sensor data among the plurality of sensor data. A status obtaining unit that obtains status information indicating a status of communication between the server device and the client device, a setting unit that sets a target transmission rate according to the status information, and a unit for each of the plurality of sensor data. A determining unit that determines a scheduled transmission rate of the plurality of sensor data based on a data amount per time; and using at least one of the plurality of methods, the scheduled transmission rate is equal to or less than the target transmission rate. An adjusting unit that adjusts a data amount of the at least one sensor data among the plurality of sensor data; To provide a transmission system comprising a transmitter for transmitting the sensor data to the server device.

The adjusting unit may thin out the at least one sensor data when the scheduled transmission rate is higher than the target transmission rate.

In the plurality of methods, priorities are determined, and the adjusting unit, when the scheduled transmission rate is higher than the target transmission rate, until the scheduled transmission rate becomes equal to or less than the target transmission rate, The at least one sensor data may be thinned out using the plurality of methods in the order of priority.

The storage unit may store a copy of the at least one sensor data in a state before the thinning, and the transmission unit may transmit the copy stored in the storage unit at a predetermined timing.

The predetermined timing may include when the expected transmission rate is lower than the target transmission rate.

The adjusting unit may stop the thinning of the at least one sensor data when the scheduled transmission speed becomes lower than the target transmission speed.

The plurality of methods may include two or more methods of reducing the same sensor data included in the plurality of sensor data by different thinning methods.

優先 The plurality of methods may have a priority order, and in the two or more methods, the thinning amount may be larger as the priority order is lower.

The present invention also provides a data acquisition unit for acquiring a plurality of sensor data output from a plurality of sensors, and a storage for storing a plurality of methods for reducing a data amount of at least one sensor data among the plurality of sensor data. Unit, a status obtaining unit that obtains status information indicating a status of communication between the server device, a setting unit that sets a target transmission rate according to the status information, and a unit time for each of the plurality of sensor data. A determining unit that determines a scheduled transmission rate of the plurality of sensor data based on a data amount per hit; and the scheduled transmission rate becomes equal to or less than the target transmission rate using at least one of the plurality of methods. An adjusting unit that adjusts the data amount of the at least one sensor data among the plurality of sensor data, and the plurality of sensor data whose data amount is adjusted. To provide a transmission apparatus and a transmission unit that transmits the data to the server device.

Further, the present invention provides a computer comprising: obtaining a plurality of sensor data output from a plurality of sensors; obtaining status information indicating a status of communication between a server device and a client device; Setting a target transmission rate in accordance with information; determining a scheduled transmission rate of the plurality of sensor data based on a data amount per unit time of each of the plurality of sensor data; and The planned transmission rate is equal to or less than the target transmission rate by using at least one of the plurality of methods for reducing the data amount of at least one sensor data among the data, wherein the at least one of the plurality of methods is stored in a storage unit Adjusting the data amount of the at least one sensor data of the plurality of sensor data so that Serial data amount to provide a program for executing a step of transmitting the plurality of sensor data is adjusted to the server device.

According to the present invention, it is possible to reduce data delay in transmitting a plurality of sensor data.

FIG. 1 is a diagram illustrating an example of a configuration of a transmission system 100 according to an embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of a client device 120. FIG. 2 is a diagram illustrating an example of a functional configuration of a client device 120. FIG. 4 is a diagram illustrating an example of a priority table 211. It is a flow chart which shows an example of transmission processing of sensor data.

1. Configuration FIG. 1 is a diagram illustrating an example of a configuration of a transmission system 100 according to the present embodiment. The transmission system 100 is a system that collects a plurality of sensor data output from a plurality of sensors 110.

The transmission system 100 includes a plurality of sensors 110, a client device 120, a server device 130, and a terminal device 140. Note that the number of each device shown in FIG. 1 is an example, and the present invention is not limited to this. The plurality of sensors 110 and the client device 120 are connected via a communication line 150. The communication line 150 transmits communication between the plurality of sensors 110 and the client device 120. The client device 120, the server device 130, and the terminal device 140 are connected via a communication line 160. The communication line 160 transmits communication between the client device 120, the server device 130, and the terminal device 140. The communication line 160 may include, for example, a mobile communication network and the Internet.

The sensor 110 has, for example, a sensor element and an AD (Analog-to-Digital) conversion circuit, detects a physical quantity from a sensing target, and outputs sensor data corresponding to the detected physical quantity. The physical quantity may be, for example, displacement, pressure, temperature, humidity, sound, light, electricity, or magnetism. The sensing target may be, for example, a machine such as a car or a drone, a component of a machine such as a car tire or a drone motor, or a living thing such as a person or an animal. For example, when the sensing target is a car, the plurality of sensors 110 include a microphone that acquires the sound of the engine, a tachometer that measures the number of revolutions of the engine, a speedometer that measures the speed of the car, and an angle that measures the angle of the steering. Various sensing devices such as a meter, a fuel injection meter, a thermometer, a GPS (Global Positioning System) for measuring the position of the vehicle, and a camera for photographing an image inside or outside the vehicle may be included. The sensor 110 performs sampling periodically or aperiodically. The sampling period may be different for each sensor 110. The sensor data is time-series data indicating a physical quantity detected from a sensing target in a time-series manner.

FIG. 2 is a diagram illustrating an example of a hardware configuration of the client device 120. The client device 120 collects the sensor data output from the sensor 110 and transmits the sensor data to the server device 130. For example, when the sensing target is a car, the client device 120 may be mounted on the car. For the client device 120, for example, a general-purpose computer may be used. For example, the client device 120 includes a processor 121, a memory 122, a storage 123, and a communication interface 124. These components are connected via a bus 125. The processor 121 performs various processes by reading the program into the memory 122 and executing the program. As the processor 121, for example, a CPU (Central Processing Unit) may be used. The memory 122 is a computer-readable recording medium, and is also called a main storage device. In the memory 122, a program executed by the processor 121 is stored. As the memory 122, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), or a combination thereof may be used. The storage 123 is a computer-readable recording medium, and is also called an auxiliary storage device. The storage 123 stores various data and programs. For the storage 123, for example, a hard disk or a flash memory may be used. The communication interface 124 is connected to the

communication lines

150 and 160, and performs data communication via the

communication lines

150 or 160. For example, the client device 120 may be connected to the communication line 160 by wire according to a wired communication standard such as a TCP (Transmission Control Protocol) / IP (Internet Protocol) network or Ethernet (registered trademark), or may be 3G or LTE (Long). Term @ Evolution), Wi-Fi (registered trademark), WiMAX (registered trademark), Bluetooth (registered trademark), or other wireless communication standards, may be connected to the communication line 160 wirelessly. Further, communication between the client device 120 and the sensor 110 via the communication line 150 may be performed in accordance with a bus communication standard such as CAN (Controller Area Network) protocol (registered trademark) and PROFIBUS (registered trademark). .

Returning to FIG. 1, the server device 130 performs various processes using the sensor data collected from the client device 120. This processing includes non-real-time processing and real-time processing. The non-real-time processing is, for example, processing for accumulating data collected from the client device 120. Thereby, the data collected from the client device 120 is managed collectively. This data is used, for example, for various analyzes on the situation of the sensing target. Therefore, non-real-time processing requires data accuracy. On the other hand, the real-time processing is processing for transmitting, for example, data collected from the client device 120 to another device such as the terminal device 140 in real time. It should be noted that this real time does not need to be completely simultaneous, and there may be some time delay. This data is used, for example, by the user to grasp the current situation of the sensing target. Therefore, real-time processing requires immediacy. For the server device 130, for example, a general-purpose computer may be used. For example, the server device 130 has the same configuration as the client device 120 described above.

The terminal device 140 is used by a user to input and output data to and from the server device 130. As the terminal device 140, for example, a computer such as a client computer, a tablet terminal, or a smartphone may be used. For example, the terminal device 140 includes an input unit 141 and a display unit 142 in addition to the same configuration as the client device 120 described above. The input unit 141 is used for inputting various information. As the input unit 141, for example, a touch panel or operation keys may be used. The display unit 142 displays various images. As the display unit 142, for example, a liquid crystal display may be used.

FIG. 3 is a diagram illustrating an example of a functional configuration of the client device 120. The client device 120 functions as a data acquisition unit 201, a status acquisition unit 202, a setting unit 203, a determination unit 204, a determination unit 205, a selection unit 206, an adjustment unit 207, and a transmission unit 208. These functions are realized by the processor 121 performing an operation or controlling communication by the communication interface 124 in cooperation with a program stored in the memory 122 and the processor 121 executing the program.

The data acquisition unit 201 acquires a plurality of sensor data output from the plurality of sensors 110. These sensor data are respectively sampled at a predetermined sampling rate.

The status obtaining unit 202 obtains status information indicating the status of communication between the server device 130 and the client device 120. This communication situation refers to a situation that affects the amount of data that can be transmitted from the client device 120 to the server device 130 per unit time. In this communication situation, for example, the communication band of the communication line 160 may be used. Note that the communication band may be at least a part of the communication band of the communication line 160, such as a communication band that is a bottleneck of the communication line 160. The communication band may be estimated by the status acquisition unit 202. Known methods of estimating the communication band include known bands such as VPS (Variable Packet Size) probing, PPTD (Packet Pair / Train Dispersion) probing, SLoPS (Self-Loading Periodic Streams), and TOPP (Trains of Packet Packets). An estimation method may be used.

The setting unit 203 sets a target transmission rate according to the status information acquired by the status acquisition unit 202. The target transmission speed is a transmission speed suitable for a communication situation. For example, the transmission rate at which the amount of data that can be transmitted per unit time corresponding to the communication status is transmitted per unit time may be set as the target transmission rate. Here, the amount of data that can be transmitted per unit time corresponding to the communication situation is a theoretical value. If it is known that the amount of data that can be actually transmitted is smaller than the theoretical value, the target transmission speed is calculated by multiplying the amount of data that can be transmitted per unit time corresponding to the communication situation by a coefficient less than 1. A transmission rate may be set such that the obtained data amount is transmitted per unit time. Conversely, if it is known that the amount of data that can be actually transmitted is larger than the theoretical value, a predetermined coefficient is added to the amount of data that can be transmitted per unit time corresponding to the communication status to the target transmission speed. The transmission rate may be set such that the amount of data obtained by the transmission is transmitted per unit time. For example, a target throughput may be used as the target transmission rate.

The determination unit 204 determines the expected transmission speed of the plurality of sensor data based on the data amount per unit time of the plurality of sensor data acquired by the data acquisition unit 201. The scheduled transmission speed is a transmission speed that minimizes a delay in transmission of a plurality of sensor data. For example, when the data amount of a plurality of sensor data transmitted per unit time is equal to or larger than the data amount per unit time included in the plurality of sensor data, the transmission delay of these sensor data is minimized. Therefore, for example, the transmission rate at which the data amount per unit time included in the sensor data is transmitted per unit time may be determined as the scheduled transmission rate. For example, a scheduled throughput may be used as the scheduled transmission speed.

The determination unit 205 compares the target transmission speed set by the setting unit 203 with the planned transmission speed determined by the determination unit 204, and determines the magnitude relationship between the planned transmission speed and the target transmission speed. This magnitude relationship may include, for example, a relationship that the planned transmission rate is greater than the target transmission rate. If the expected transmission rate is higher than the target transmission rate, it indicates that a delay occurs in the transmission of a plurality of sensor data. On the other hand, when the expected transmission rate is equal to or lower than the target transmission rate, it indicates that the transmission delay of the plurality of sensor data is minimized.

The selection unit 206 selects at least one of a plurality of thinning methods for reducing the data amount in accordance with the priority. Priorities are determined in advance for a plurality of thinning methods. For example, when the scheduled transmission rate is higher than the target transmission rate, a plurality of thinning methods may be selected in descending order of priority. This selection may be made based on the priority table 211 stored in the storage 123, for example.

FIG. 4 is a diagram showing an example of the priority table 211. The priority table 211 includes a priority, a data identifier, and a thinning method in association with each other. The priority indicates the order in which the thinning method is applied. A thinning method with a higher priority is applied before a thinning method with a lower priority. The data identifier is information for uniquely identifying sensor data to be thinned out by applying the thinning method. One data identifier may be associated with the same priority, or a plurality of data identifiers may be associated. The thinning method is a method of thinning data. This thinning means that at least a part of the sensor data is deleted. The thinning-out method includes, for example, lowering a sampling rate, extracting and substituting a feature point, quantizing, performing reversible or irreversible compression, generating an approximate value and substituting, calculating a statistical value and substituting, For example, a known thinning-out method may be used, such as lowering the encoding rate or changing to an encoding method having a high compression ratio.

The priority table 211 may be generated in the server device 130 according to an operation performed by the user using the input unit 141 in, for example, an initial setting operation. At this time, for example, as the importance of the sensor data is higher, the user may lower the priority of the method of thinning out the sensor data. Further, when it is preferable that the plurality of sensor data be thinned out at the same time, the user may set the same priority for a plurality of methods of thinning out the sensor data. Furthermore, if there is one sensor data among the plurality of sensor data and the other sensor data is unnecessary, the user may associate the other sensor data with a thinning-out method of deleting all the sensor data. The priority table 211 generated in this manner is transmitted from the server device 130 to the client device 120 and stored in the storage 123. The priority table 211 may be changed or deleted by a similar method.

The adjustment unit 207 adjusts the data amount of at least one of the plurality of sensor data based on the thinning method selected by the selection unit 206 so that the planned transmission speed is equal to or less than the target transmission speed. For example, when the planned transmission speed is higher than the target transmission speed, it is better to reduce the data amount of at least one sensor data among the plurality of sensor data to minimize the transmission delay of the plurality of data. Therefore, for example, the target sensor data may be thinned using the thinning method selected by the selection unit 206. On the other hand, after the target sensor data has been thinned out, if the planned transmission speed is lower than the target transmission speed, the data amount of at least one of the plurality of sensor data is increased to increase the transmission speed. Better. Therefore, for example, the application of the thinning method selected by the selection unit 206 may be canceled, and the thinning using the thinning method for the target sensor data may be stopped.

The transmission unit 208 transmits a plurality of sensor data to the server device 130. This transmission may be performed in a streaming manner. The “streaming method” refers to a method of sequentially transmitting data according to data acquisition. For example, each sensor data is divided into a plurality of sections. A section of another sensor data is inserted between one sensor data section and transmitted so that the amount of data included per unit time of each sensor data is transmitted per unit time. As a result, a plurality of sensor data are transmitted together at the predetermined transmission speed.

2. Operation FIG. 5 is a flowchart illustrating an example of a sensor data transmission process. This transmission process is executed in the client device 120. The plurality of sensors 110 include, for example, a sensor 110A, a sensor 110B, a sensor 110C, and a sensor 110D. For example, the sensor 110A is a goniometer that measures the steering angle of a car. The sensor 110A outputs measurement data indicating a steering angle of the vehicle in a time series. The data identifier of this measurement data is “001”. The sensor 110B is a video camera that captures a moving image of a driver's seat and a passenger seat of a car in full HD (high definition). The sensor 110B outputs moving image data indicating the moving image. The data identifier of this moving image data is “002”. The sensor 110C is a video camera that shoots a moving image of a driver's seat of a car in HD. The sensor 110C outputs moving image data indicating the moving image. The data identifier of this moving image data is “003”. The sensor 110D is a video camera that captures a moving image in front of the car using QVGA (Quarter VGA). The sensor 110D outputs moving image data indicating the moving image. The data identifier of this moving image data is “004”.

In step S11, measurement of the sensor data received by the data acquisition unit 201 from the sensors 110A to 110D is started. The start of the measurement may be performed, for example, in response to an input of a signal instructing the start of the measurement of the sensor data in response to a user operation. This signal may be transmitted from the terminal device 140, for example, or may be transmitted from the sensing target when the sensing target and the client device 120 are connected. For example, measurement of sensor data with data identifiers “001” to “004” received from the sensors 110A to 110D is started.

では In step S12, the determining unit 204 calculates the expected throughput Sp. For example, the sampling rate of the measurement data with the data identifier “001” is 1000 [Hz], and the data amount per sample is X1 [bit]. The sampling rate of the moving image data with the data identifier “002” is 100 [Hz], and the data amount per sample is X2 [bit]. Further, the sampling rate of the moving image data with the data identifier “003” is 100 [Hz], and the data amount per sample is X3 [bit]. The sampling rate of the moving image data with the data identifier “004” is 50 [Hz], and the data amount per sample is X4 [bit]. In this case, the scheduled throughput Sp is obtained by the formula of 1000 × X1 + 100 × X2 + 100 × X3 + 50 × X4.

In step S13, the situation acquisition unit 202 estimates the communication band of the communication line 160. For example, a communication band of 50 bps (Bit @ Per @ Second) is estimated.

In step S14, the setting unit 203 sets the target throughput St according to the communication band estimated in step S13. For example, 50 bps equal to the estimated communication band is set as the target throughput St.

In step S15, the determination unit 205 determines whether or not the scheduled throughput Sp calculated in step S12 is larger than the target throughput St set in step S14. When it is determined that the scheduled throughput Sp is larger than the target throughput St (the determination in step S15 is YES), the process proceeds to step S16.

In step S16, the selecting unit 206 selects the thinning method having the highest priority among the unselected thinning methods included in the priority table 211 as the thinning method to be applied. For example, when all the thinning methods are not selected in the priority table 211 illustrated in FIG. 4, the thinning method associated with the priority “1” is selected. This thinning method is a method of thinning out the measurement data with the data identifier “001” at equal intervals so that the sampling rate becomes 100 [Hz].

In step S17, the determining unit 204 calculates the expected throughput Sp1 when the thinning method selected in step S17 is applied. For example, when the thinning method with the priority “1” shown in FIG. 4 is selected, for example, the sampling rate of the measurement data with the data identifier “001” becomes 100 [Hz]. In this case, the scheduled throughput Sp is obtained by the formula of 100 × X1 + 100 × X2 + 100 × X3 + 50 × X4.

In step S18, the determination unit 205 determines whether the planned throughput Sp1 calculated in step S17 is larger than the target throughput St set in step S14. If the planned throughput Sp1 is larger than the target throughput St, the determination in step S18 is YES, and the process returns to step S16 described above. In this case, in step S16, for example, in the priority table 211 shown in FIG. 4, the thinning method associated with the priority "2" is selected. This thinning-out method is a method of thinning out moving image data having a data identifier of “002” at equal intervals so that the sampling rate becomes 30 [Hz]. In this case, the sampling rate of the moving image data with the data identifier “002” is 30 [Hz]. Therefore, in step S18, the scheduled throughput Sp1 is calculated by the formula of 100 × X1 + 30 × X2 + 100 × X3 + 50 × X4.

Steps S16 to S18 are repeated until the planned throughput Sp1 becomes equal to or less than the target throughput St. In the meantime, a plurality of thinning methods included in the priority table 211 shown in FIG. 4 are applied in descending order of priority.

(4) When the scheduled throughput Sp1 becomes equal to or less than the target throughput St, the determination in step S18 becomes NO. In this case, the process proceeds to step S19.

In step S19, a copy of the state of the sensor data to be thinned out using one of the thinning methods selected in step S16 before thinning is stored in the retransmission queue of the memory 122 or the storage 123. For example, when the thinning methods with priority levels “1” to “7” are selected, sensor data with data identifiers “001” to “004” are thinned using these thinning methods. Therefore, a copy of the sensor data with the data identifiers “001” to “004” received in step S11 is stored in the retransmission queue. The sampling rate of the copy of the measurement data with the data identifier “001” is 1000 [Hz]. The sampling rate of the copy of the moving image data with the data identifier “002” is 100 [Hz]. The sampling rate of the copy of the moving image data whose data identifier is “003” is 100 [Hz]. The sampling rate for copying the moving image data with the data identifier “004” is 50 [Hz].

In step S20, the adjustment unit 207 thins out the first untransmitted data unit in the target sensor data using all the thinning methods selected in step S17. As this data unit, for example, the smallest processing unit may be used. When a plurality of thinning methods are applied to the same sensor data, only the thinning method having the lowest priority among the thinning methods may be applied. For example, when the thinning method with the priority order of “1” to “7” is selected, only the characteristic points are extracted from the measurement data with the data identifier “001”, and the data other than the characteristic points are deleted. The moving image data with the data identifier “002” is thinned out so that the sampling rate becomes 0 [Hz], that is, all the thinning-out is performed. The moving image data with the data identifier “003” is also decimated so that the sampling rate becomes 0 [Hz], that is, all are decimated. The moving image data with the data identifier “004” is thinned out at equal intervals so that the sampling rate becomes 30 [Hz].

In step S21, the first untransmitted data unit of the plurality of sensor data received by the data acquisition unit 201 is transmitted from the transmission unit 208 to the server device 130 with the target sensor data thinned out in step S20. . For example, as for the measurement data with the data identifier “001”, only the feature points are transmitted. The moving image data with the data identifier “002” and the moving image data with “003” are not transmitted. The moving image data with the data identifier “004” is transmitted with the sampling rate reduced to 30 [Hz].

On the other hand, when it is determined in step S15 that the planned throughput Sp is equal to or smaller than the target throughput St (NO in step S15), the processing in step S16 to step S20 is not performed, and the processing proceeds to step S21. Proceed to. In this case, in step S21, the first untransmitted data unit in the plurality of sensor data received by the data acquisition unit 201 is transmitted from the transmission unit 208 to the server device 130 without being thinned. For example, the measurement data with the data identifier “001” is transmitted at a sampling rate of 1000 [Hz]. The moving image data with the data identifier “002” is transmitted at a sampling rate of 100 [Hz]. The moving image data with the data identifier “003” is transmitted at a sampling rate of 100 [Hz]. The moving image data with the data identifier “004” is transmitted at a sampling rate of 50 [Hz].

In step S22, it is determined whether the measurement of the sensor data has been completed. This determination may be made, for example, according to whether or not a signal instructing the end of the sensor data measurement has been input in response to a user operation. This signal may be transmitted from the terminal device 140, for example, or may be transmitted from the sensing target when the sensing target and the client device 120 are connected. For example, when the signal instructing the end of the measurement of the sensor data is not input, the determination in step S22 is NO. In this case, the process returns to step S13 described above. Then, the processing after step S13 is performed for the next data unit in the plurality of sensor data received by the data acquisition unit 201. As described above, the processing in steps S13 to S22 is performed for each data unit of the sensor data.

Here, for example, the communication band of the communication line 160 may change due to the movement of the client device 120 or a change in the communication environment. In this case, in step S13 performed after the communication band changes, the communication band after the change is estimated. For example, a communication band of 80 bps is estimated. In the following step S14, a new target throughput St is set according to the changed communication band. For example, 80 bps equal to the changed communication band is set as the target throughput St.

In this case, when it is determined in step S15 that the planned throughput Sp is equal to or less than the new target throughput St (NO in step S15), the next data unit in the plurality of sensor data received by the data acquisition unit 201 Is sent without being decimated. At this time, the application of the thinning method performed in the previous processing is released. For example, when sensor data with data identifiers “001” to “004” are thinned out in the previous process, the thinning out of these sensor data is stopped.

On the other hand, when it is determined in step S15 that the planned throughput Sp is larger than the new target throughput St (YES in step S15), steps S16 to S18 are performed until the planned throughput Sp1 becomes equal to or smaller than the new target throughput St. Is repeated. In this case, a plurality of thinning methods included in the priority table 211 shown in FIG. 4 are applied in order from the thinning method having the highest priority. That is, even when some thinning methods are applied in the previous process, all of the applications of these thinning methods are temporarily canceled.

When the scheduled throughput Sp1 becomes equal to or less than the new target throughput St, in step S19, a copy of the current target sensor data is stored in the retransmission queue. Subsequently, in step S20, the next data unit in the target sensor data is thinned using all the thinning methods selected in step S16. At this time, all the thinning performed in the previous processing is temporarily stopped. For example, when sensor data with data identifiers “001” to “004” are thinned out in the previous process, the thinning of these sensor data is temporarily stopped. Then, the current target sensor data is thinned again using the thinning method selected this time. For example, when the thinning method of the priority order “1” or “2” is selected, the measurement data with the data identifier “001” is thinned out at equal intervals so that the sampling rate becomes 100 [Hz]. The moving image data with the data identifier “002” is thinned out at equal intervals so that the sampling rate becomes 30 [Hz]. Moving image data with data identifiers “003” and “004” are not thinned out. Then, in step S21, the next data unit in the plurality of sensor data is transmitted with the target sensor data thinned out. For example, the measurement data with the data identifier “001” is transmitted with the sampling rate reduced to 100 [Hz]. The moving image data with the data identifier “002” is transmitted with the sampling rate reduced to 30 [Hz]. The moving image data with the data identifier “003” is transmitted at a sampling rate of 100 [Hz]. The moving image data with the data identifier “004” is transmitted at a sampling rate of 50 [Hz].

On the other hand, for example, when a signal instructing the end of the sensor data measurement is input, the determination in step S22 is YES. In this case, the process proceeds to step S23.

In step S23, it is determined whether a copy of the sensor data is stored in the retransmission queue. For example, when a copy of the data with the data identifier “001” to “004” is stored in the retransmission queue, the determination in step S23 is YES. In this case, the process proceeds to step S24.

In step S24, a copy of the sensor data stored in the retransmission queue is retransmitted from the transmission unit 208 to the server device 130. For example, when duplicates of sensor data with data identifiers “001” to “004” are stored in the retransmission queue, duplicates of these sensor data are transmitted.

On the other hand, if no copy of the sensor data is stored in the retransmission queue, the determination in step S23 is NO. In this case, the process ends without performing the process of step S24.

In the above-described embodiment, the “memory 122” or the “storage 123” and the “client device 120” are used as the “storage unit” and the “transmission device” according to the present invention, respectively.

According to the above-described embodiment, since the sensor data is thinned out according to the communication situation, the transmission delay of a plurality of sensor data can be reduced. Further, when the planned throughput Sp becomes equal to or less than the target throughput St due to a change in the communication situation, at least part of the application of the thinning method may be canceled, so that the sensor data can be efficiently transmitted. it can. Further, since a plurality of thinning methods are applied in descending order of priority, it is possible to reduce the data amount by applying a thinning method having a higher priority before a thinning method having a lower priority. For example, when the user determines the priority according to the importance of the data, the amount of data can be reduced and the delay can be reduced according to the importance of the data required by the user. In this case, it is possible to reduce the delay of data having high importance. Furthermore, since the priority table 211 includes a plurality of different thinning methods for the same sensor data, the same sensor data can be thinned by different thinning methods. Further, in a plurality of thinning methods for the same sensor data, the sampling rate after the thinning is lower as the priority order is lower, that is, since the thinning amount is larger, the data amount of the same sensor data is gradually reduced. Can be reduced.

3. Modifications The embodiment described above is an example of the present invention. This embodiment may be modified as follows. Moreover, you may implement combining the following modification examples.

In the above embodiment, the client device 120 may adjust the amount of sensor data under the control of the server device 130. In this case, the priority table 211 is stored in the storage of the server device 130. Further, the sampling rate and the data amount per sample of the sensor data output from the sensor 110 are transmitted from the client device 120 to the server device 130. When the sensor 110 performs aperiodic sampling, the sampling rate may be replaced by using a statistical value such as an average sampling rate. The server device 130 includes a status acquisition unit 202, a setting unit 203, a determination unit 204, a determination unit 205, and a selection unit 206 illustrated in FIG. Further, the server device 130 instructs the client device 120 to execute a thinning process using the thinning method selected by the selecting unit 206. The client device 120 thins out the target sensor data using the selected thinning method according to the instruction of the server device 130. According to this modification, even if the client device 120 does not have the status acquisition unit 202, the setting unit 203, the determination unit 204, and the selection unit 206 shown in FIG. 3, the transmission delay of a plurality of sensor data is reduced. be able to.

予定 In the above-described embodiment, the planned throughput Sp may not always be calculated. For example, when compressing sensor data, the data amount after compression may not be known in advance. In this case, the scheduled throughput Sp may be determined based on the data amount after the sensor data is actually compressed.

In the embodiment described above, the sensor data to be thinned out may be determined using the group identifier. In this case, the priority table 211 includes a group identifier instead of the data identifier. This group identifier is information for uniquely identifying a group of sensor data. This group may be classified, for example, by the type of sensor data, or may be classified by a channel number for identifying a channel to which sensor data is input.

In the above-described embodiment, each thinning method may be associated with the thinning method and a data amount to be reduced when a thinning method having a higher priority than the thinning method is applied. The data amount to be reduced may be calculated in advance. In this case, a thinning method with the highest priority and a thinning method with a higher priority than the thinning method among the thinning methods in which the data amount equal to or more than the difference between the scheduled throughput Sp and the target throughput St may be applied. . According to this modification, sensor data can be thinned using a thinning method suitable for a communication situation without applying a plurality of thinning methods in order.

In the above-described embodiment, the method for reducing the data amount is not limited to the above-described thinning method. Any method may be used as long as the data amount can be reduced.

In the embodiment described above, another device may have at least a part of the function of the client device 120. For example, the server device 130 may have at least a part of the functions of the client device 120 illustrated in FIG.

Steps of processing performed in the transmission system 100 are not limited to the example described in the above-described embodiment. The steps of this process may be interchanged as long as there is no inconsistency. The present invention may be provided as a transmission method including steps of processing performed in the transmission system 100.

The present invention may be provided as a program executed in the sensor 110, the client device 120, the server device 130, or the terminal device 140. This program may be downloaded via a communication line such as the Internet. These programs are provided in a state recorded on a computer-readable recording medium such as a magnetic recording medium (magnetic tape, magnetic disk, etc.), an optical recording medium (optical disk, etc.), a magneto-optical recording medium, and a semiconductor memory. May be done.

In the above-described embodiment, the timing at which the copy of the sensor data stored in the retransmission queue is transmitted is not limited to after the measurement of the sensor data is completed. This transmission may be performed at a predetermined timing. The predetermined timing may be, for example, a timing that does not affect the transmission of the sensor data. This is because transmission of sensor data requires immediacy, whereas transmission of a copy of sensor data stored in the retransmission queue is retransmission, so immediacy is not required. For example, when the scheduled throughput Sp is smaller than the target throughput St, a copy of the sensor data stored in the retransmission queue may be transmitted. This is because, when the scheduled throughput Sp is smaller than the target throughput St, it is considered that there is a margin in the communication band. Therefore, even when the sensor data is transmitted and the duplicate of the sensor data stored in the retransmission queue is transmitted, the sensor data is transmitted. This is because it is considered that there is a low possibility of affecting transmission.

In the above-described embodiment, when the planned throughput Sp becomes smaller than the target throughput St after thinning out the target sensor data, the adjustment unit 207 gives a low priority to the application of the thinning method applied in the previous process. It may be canceled in order. In this case, the thinning method that has been applied is selected in descending order of priority, the application of the selected thinning method is released, and the scheduled throughput when the application of the thinning method having a higher priority than the selected thinning method is maintained. Sp2 is calculated. This process is repeated until the scheduled throughput Sp2 reaches the target throughput St or a predetermined range from the target throughput St. The predetermined range may be, for example, a range from a value obtained by multiplying the target throughput St by a coefficient less than 1 to the target throughput St. For example, 0.8 may be used for this coefficient.

For example, when the thinning method of priority “1” to “7” is applied in the priority table 211 shown in FIG. 4 in the previous process, the thinning method of priority “7” is selected first. This thinning method is a method of extracting only the characteristic points of the measurement data with the data identifier “001” and deleting the points other than the characteristic points. When the thinning method with the priority “7” shown in FIG. 4 is canceled, the extraction of the feature point from the measurement data with the data identifier “001” is stopped. In this case, the sampling rate of the measurement data with the data identifier “001” is in a state where the thinning method with the priority “5” is applied, that is, 10 [Hz]. Note that since the application of the thinning method with the priority order of “1” to “6” is maintained, the sampling rates of the moving image data with the data identifiers “002” and “003” are both 0 [Hz]. The sampling rate of the moving image data with the data identifier “004” is 30 [Hz]. In this case, the scheduled throughput Sp2 is obtained by the formula of 10 × X1 + 0 × X2 + 0 × X3 + 30 × X4.

If the estimated throughput Sp2 calculated in this way does not reach the target throughput St or within a predetermined range from the target throughput St, for example, it is associated with the priority “6” in the priority table 211 shown in FIG. Two thinning methods are selected. When the application of the first thinning method among these thinning methods is released, the thinning of the moving image data having the data identifier “004” to 30 [Hz] is stopped. In this case, the sampling rate of the moving image data with the data identifier “004” is in a state where no thinning is performed, that is, 50 [Hz]. When the application of the second thinning method is released, the thinning of the moving image data having the data identifier "003" to 0 "Hz" is stopped. In this case, the sampling rate of the moving image data with the data identifier “003” is in a state where the thinning method with the priority order “4” is applied, that is, 30 [Hz]. In addition, since the application of the thinning method with the priority order of “1” to “5” is maintained, the sampling rate of the moving image data with the data identifier “001” is 10 [Hz]. The sampling rate of the moving image data with the data identifier “002” is 0 [Hz]. In this case, the scheduled throughput Sp2 is obtained by the formula of 10 × X1 + 0 × X2 + 30 × X3 + 50 × X4.

Thus, the above-described processing is repeated until the scheduled throughput Sp2 reaches a predetermined range from the target throughput St or the new target throughput St. In the meantime, the application of the applied thinning-out method included in the priority table 211 shown in FIG. 4 is canceled in ascending order of priority. Then, when the scheduled throughput Sp2 reaches within a predetermined range from the target throughput St or the new target throughput St, the application of all the selected thinning methods among the applied thinning methods is released. As a result, the thinning of the sensor data of the object thinned using these thinning methods is stopped. For example, when the method of thinning out the priorities “6” and “7” is canceled, the measurement data with the data identifier “001” is thinned out at equal intervals so that the sampling rate becomes 10 [Hz]. The moving image data with the data identifier “003” is thinned out so that the sampling rate becomes 30 [Hz]. The thinning of the moving image data with the data identifier “004” is stopped. Note that all the moving image data with the data identifier “002” is continuously deleted.

Then, the data unit of the plurality of sensor data received by the data acquisition unit 201 is transmitted from the transmission unit 208 to the server device 130 in a state where the application of the thinning method to at least a part of the target sensor data is released. For example, the measurement data with the data identifier “001” is transmitted at a sampling rate of 10 [Hz]. The moving image data with the data identifier “002” is not transmitted. The moving image data with the data identifier “003” is transmitted at a sampling rate of 30 [Hz]. The moving image data with the data identifier “004” is transmitted at a sampling rate of 50 [Hz].

In short, the “adjustment unit” according to the present invention is configured such that, after thinning out at least one sensor data using at least one method, when the planned transmission speed becomes smaller than the target transmission speed, the planned transmission speed is reduced to the target transmission speed. Until the speed is reached, the application of at least one method may be canceled in the order of lower priority, and the thinning using the method whose application is canceled for the at least one sensor data may be stopped.

100: Transmission system, 110: Sensor, 120: Client device, 130: Server device, 140: Terminal device, 201: Data acquisition unit, 202: Status acquisition unit, 203: Setting unit, 204: Decision unit, 205: Judgment unit , 206: selection unit, 207: adjustment unit, 208: transmission unit

Claims

A data acquisition unit that acquires a plurality of sensor data output from a plurality of sensors,
A storage unit that stores a plurality of methods for reducing a data amount of at least one sensor data among the plurality of sensor data;
A status obtaining unit that obtains status information indicating a status of communication between the server device and the client device;
A setting unit for setting a target transmission rate according to the status information,
Based on the data amount per unit time of each of the plurality of sensor data, a determination unit that determines a scheduled transmission speed of the plurality of sensor data,
An adjusting unit that adjusts a data amount of the at least one sensor data among the plurality of sensor data using at least one of the plurality of methods so that the planned transmission speed is equal to or less than the target transmission speed. When,
A transmission unit that transmits the plurality of sensor data whose data amount has been adjusted to the server device.
The transmission system according to claim 1, wherein the adjustment unit thins out the at least one sensor data when the expected transmission rate is higher than the target transmission rate.
Priorities are defined for the plurality of methods,
The adjusting unit, when the scheduled transmission rate is higher than the target transmission rate, uses the plurality of methods in the descending order of the priority until the scheduled transmission rate becomes equal to or less than the target transmission rate. The transmission system according to claim 2, wherein the sensor data is thinned out.
The storage unit stores a copy of the at least one sensor data in the state before the thinning,
The transmission system according to claim 2, wherein the transmission unit transmits the copy stored in the storage unit at a predetermined timing.
The transmission system according to claim 4, wherein the predetermined timing includes a time when the expected transmission rate is smaller than the target transmission rate.
The transmission system according to any one of claims 2 to 5, wherein the adjustment unit stops the thinning of the at least one sensor data when the planned transmission speed becomes lower than the target transmission speed.
The transmission system according to any one of claims 1 to 6, wherein the plurality of methods include two or more methods of reducing the same sensor data included in the plurality of sensor data by different thinning methods.
Priorities are defined for the plurality of methods,
The transmission system according to claim 7, wherein in the two or more methods, the lower the priority order, the larger the thinning amount.
A data acquisition unit that acquires a plurality of sensor data output from a plurality of sensors,
A storage unit that stores a plurality of methods for reducing a data amount of at least one sensor data among the plurality of sensor data;
A status obtaining unit that obtains status information indicating a status of communication with the server device;
A setting unit for setting a target transmission rate according to the status information,
Based on the data amount per unit time of each of the plurality of sensor data, a determination unit that determines a scheduled transmission speed of the plurality of sensor data,
An adjusting unit that adjusts a data amount of the at least one sensor data among the plurality of sensor data using at least one of the plurality of methods so that the planned transmission speed is equal to or less than the target transmission speed. When,
A transmission unit that transmits the plurality of sensor data whose data amount has been adjusted to the server device.
On the computer,
Obtaining a plurality of sensor data output from the plurality of sensors;
Obtaining status information indicating the status of communication between the server device and the client device;
Setting a target transmission rate according to the status information;
Based on the data amount per unit time of each of the plurality of sensor data, determining a scheduled transmission speed of the plurality of sensor data,
A plurality of methods for reducing a data amount of at least one sensor data among the plurality of sensor data, wherein at least one of the plurality of methods stored in a storage unit is used, and the expected transmission speed is reduced by the Adjusting the data amount of the at least one sensor data of the plurality of sensor data so as to be equal to or less than a target transmission speed;
And transmitting the plurality of sensor data whose data amount has been adjusted to the server device.