Timestamp correction system, sensor system, and timestamp correction method

Abstract

A timestamp correcting system (121) according to the present invention is a system for correcting a timestamp attached to a packet when receiving packets transmitted at a predetermined transmission time interval, and comprises: a delay circuit (1211) that delays a timestamp; a subtraction unit (1212) that calculates a difference in timestamp between a delayed packet and a subsequently received packet as an arrival time interval; a packet number estimating unit (1213) that calculates an average value of quantized numbers of arrival time intervals as a packet number estimation value; a computing unit (1214) that multiplies the transmission time interval by the packet number estimation value to calculate an arrival time interval; a moving average filter (1215) that executes moving averaging of the arrival time intervals; and an addition unit (1216) that adds an initial time to a moving average value.　In this way, the present invention can provide a timestamp correcting system that reduces an error of a timestamp attached to a packet.

Description

Technical Field

[0001] The present invention relates to a timestamp correction system, a sensor system, and a timestamp correction method in data transmission and reception.

Background Art

[0002] In an IoT (Internet of Things) network, various sensors are connected, and it is expected to collect a large amount of various data and extract useful information by analyzing the data. Therefore, it is required that the terminal accommodating the sensor corresponds to various use case needs, and power consumption reduction is required in long-term measurement (Non-Patent Document 1).

[0003] In particular, in a sensor, by appropriately putting the MPU (Micro-Processing Unit), which consumes more power than other elements, into sleep, intermittent operation can be realized, and by continuously operating only the analog front end (AFE), the cost and power consumption can be reduced (Non-Patent Document 2).

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, when a time stamp is executed at the data collection terminal, the delay time associated with the communication between the sensor and the data collection terminal cannot be canceled, and an error occurs in the time stamp.

[0006] Also, even if the intervals at which the MPU starts are accurate, there are constantly errors in the clock mounted on the AFE, fluctuations occur in the intervals of the packets to be transmitted, and errors occur in the time stamp.

[0007] In addition, due to external noise to the wireless section, the reception timing is shifted, and an error occurs in the time stamp.

Means for Solving the Problem

[0008] In order to solve the problems as described above, a time stamp correction system according to the present invention corrects the time stamps assigned to a plurality of packets when the plurality of packets transmitted at a predetermined transmission interval time are received. A time stamp correction system, comprising: a delay circuit that delays the time stamp of one of the plurality of packets; the time stamp of another packet that is subsequently received by the one packet; and the time stamp of the one packet that has been delayed. A subtraction unit that calculates the difference as the arrival interval time; quantizes the arrival interval time; a packet number estimation unit that calculates the average value of the quantization numbers as an estimated value of the number of packets; and the estimated value of the number of packets. A calculation unit that multiplies by the transmission interval time to calculate an inverse-quantized arrival interval time; a moving average filter that performs a moving average on the inverse-quantized arrival interval time; and an addition unit that adds to the value obtained by the moving average. By comparing with the transmission interval time Quantize, calculate the sum of the quantization numbers of the quantized arrival interval times in a predetermined period, and by dividing the calculated sum of the quantization numbers by the number of packets arrived in the predetermined period The average value of the quantization numbers calculate , the average value of the quantization numbers as the Estimated value of the number of packets same as A packet number estimation unit; a calculation unit that multiplies the estimated value of the number of packets by the transmission interval time to calculate an inverse-quantized arrival interval time; a moving average filter that performs a moving average on the inverse-quantized arrival interval time; and a value obtained by the moving average. the time stamp of the one packet And an addition unit that adds.

[0009] Also, a time stamp correction method according to the present invention is a method for correcting a time stamp assigned to a packet when a plurality of packets transmitted at a predetermined transmission interval time are received. The delay circuit delays the time stamp of one packet among the plurality of packets, and The subtraction unit the time stamp of another packet continuously received after the one packet and the delayed Thetime stamp of the one packet Time stamp and of calculate the difference calculating as the arrival interval time, and the packet number estimation unit quantizes the arrival interval time into the By comparing with the transmission interval time step of quantization, and the packet number estimation unit calculate the sum of the quantization numbers of the quantized arrival interval times in a predetermined period, and by dividing the calculated sum of the quantization numbers by the number of packets arrived in the predetermined period average value of the quantization number is calculate , the average value of the quantization numbers as the estimated value of the number of packets same as step of, and the calculation unit multiplies the estimated value of the number of packets by the transmission interval time to obtain the inverse-quantized arrival interval time calculate step, the moving average filter executes a moving average on the inverse-quantized arrival interval time, and the addition unit adds the time stamp of the one packet to the value obtained by the moving average.

Advantages of the Invention

[0010] According to the present invention, a time stamp correction system, a sensor system, and a time stamp correction method for reducing errors in time stamps attached to packets can be provided.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2A

Figure 2B

Figure 3

Figure 4

Figure 5

Figure 6A

Figure 6B

Figure 6C

Figure 6D

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

[0012] <First Embodiment> The timestamp correction system, sensor system, and timestamp correction method according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 8.

[0013] <Configuration of the Sensor System> As shown in FIG. 1, the sensor system 10 according to this embodiment includes a sensor 11 and a receiver 12. A signal is transmitted and received wirelessly or wiredly between the sensor 11 and the receiver 12.

[0014] The sensor 11 operates intermittently and includes an AFE 111, a memory 112, an MPU 113, and a transmission unit 114. It also includes an AFE clock 115 connected to the AFE 111 and a packet clock 116 connected to the MPU 113.

[0015] The AFE 111 samples and quantizes the measurement signal 1 at the time T counted by the AFE clock 115. AFE

[0016] The memory 112 stores the quantized measurement signal 1 as sensor data 2.

[0017] Here, one packet 3 is composed of a predetermined number of sensor data 2.

[0018] The MPU 113 is activated at a predetermined interval T counted by the packet clock 116 and checks the memory 112. When the sensor data 2 to be stored in one packet 3 is accumulated in the memory 112, a wireless circuit (for example, BLE, Bluetooth Low Energy) is activated, and the packet 3 is transmitted from the transmission unit 114. Thus, the packet transmission interval time is T. packet packet

[0019] The receiver 12 includes a timestamp correction system 121 according to this embodiment, a timestamp adding unit 122, and a receiving unit 123. It also includes a timestamp clock 125.

[0020] The timestamp adding unit 122 according to this embodiment, after receiving a packet by the receiving unit 123, adds a timestamp T counted by the timestamp clock 125 and provided via the receiver's OS 124. stamp ​​​Attach it to the packet. Here, the timestamp is synchronized by GPS (Global Positioning System), NTP (Network Time Protocol), NITZ (Network Identity and Time Zone), etc.

[0021] As will be described later, the timestamp correction system 121 corrects the timestamp attached to the packet input from the timestamp attachment unit 122.

[0022] <Configuration and Operation of Timestamp Correction System> The configuration and operation of the timestamp attachment unit 122 according to this embodiment will be described below.

[0023] First, an explanation will be given of the case without a timestamp correction system. In this configuration, the T at the AFE clock 115 AFE has a large error for power consumption reduction and cost reduction, and may deviate by up to about several minutes per day at most. This error becomes a problem for biosensors that measure for more than one day. Details will be described below.

[0024] Figs. 2A and B show examples of the transmission modes 131_1 and 132_1 and the reception modes 131_2 and 132_2 of packets when timestamp correction is not performed. Here, the length Lp of one packet corresponds to 8 sensor data 2. The white circles 3_1 to 3_4 indicate packets, and T1 to T4 indicate the packet arrival interval times.

[0025] When the AFE clock 115 is faster than the startup interval of the MPU 113, as shown in Fig. 2A, there may be a case where data for more than two packets (for example, corresponding to packets 3_2 and 3_3) is held at startup (131_1). In this case, since packets 3_2 and 3_3 are transmitted continuously, the packet arrival interval time T2 becomes very short on the receiver side (132_1).

[0026] Also, when the AFE clock is slower than the MPU startup interval, as shown in FIG. 2B, the data volume of the packet to be transmitted at startup is insufficient (dotted white circle 3'). At this time, the packet is not transmitted (131_2). In this case, the arrival interval time T4 of the packet becomes longer on the receiver side (132_2).

[0027] As described above, in a configuration without a timestamp correction system, the arrival interval time of the packet varies greatly depending on the AFE clock, and errors due to transmission errors during wireless transmission also occur. Therefore, if the timestamp at reception is used as it is, the error increases.

[0028] Next, the timestamp correction system 121 and the timestamp correction system according to the present embodiment will be described. FIG. 3 shows the configuration of the timestamp correction system 121 according to the present embodiment. FIG. 4 shows a flowchart of the timestamp correction method according to the present embodiment.

[0029] As shown in FIG. 3, the timestamp correction system 121 includes a delay circuit 1211, a subtraction unit 1212, a packet number estimation unit 1213, an arithmetic unit 1214, a moving average filter 1215, and an addition unit 1216.

[0030] Here, the timestamp correction system 121 acquires and holds the transmission interval time T packet in advance. The transmission interval time T packet may be stored in the timestamp correction system 121 in advance, or may be transmitted from the sensor 11.

[0031] The arrival time T arrival assigned to the packet is input to the timestamp correction system 121. Here, the arrival time T arrival is the same as the timestamp T stamp .

[0032] The delay circuit 1211 is composed of a single-stage delay circuit, and the arrival time T arrival (timestamp Tstamp ) For example, T arrival Delay [i - 1].

[0033] At the subtraction unit 1212, when a continuously received packet (another packet) arrives at time T arrival (For example, T arrival [i]) and the T delayed by the delay circuit arrival (For example, T arrival [i - 1]), calculate the difference (for example, T arrival [i] - T arrival [i - 1]) (step S11). This difference in arrival time T arrival (timestamp T stamp ) is the arrival interval time T interval [i].

[0034] The packet number estimation unit 1213 estimates the packet number from the arrival interval time T interval [i].

[0035] Here, as shown in FIG. 5, the arrival interval time is distributed with noise with respect to an integer multiple (n) of the transmission interval time T packet from the sensor. Therefore, in order to cancel this noise, the estimated value of T packet is quantized with n = 0, 1, 2,... as the number of packets estimated to be sent within the transmission interval with n = 1 (step S12).

[0036] Regarding quantization by the number of packets, taking the packets received when the clock of the AFE is faster than the startup interval of the MPU as an example, FIGS. 6A to 6D will be referred to for explanation.

[0037] Packets P(i) to P(i - 5) are received, and the arrival interval times thereof are T interval (i) to T interval (i - 4) (FIG. 6A).

[0038] First, set the time between received packets as the transmission interval time T packet .

[0039] The arrival interval time T of the actually received packet interval (i)~T interval (i - 4) each is T packet By comparing with the arrival interval time T of the packet interval (i)~T interval (i - 4) is T packet When it corresponds to n times of T, the quantization number is set to n (Fig. 6B).

[0040] Specifically, since T interval (i) is equal to T packet the quantization number is "1". Similarly, since T interval (i - 1) is equal to T packet the quantization number is "1". Next, since T interval (i - 2) is shorter than T packet the quantization number is "0". Next, since T interval (i - 3) is equal to T packet the quantization number is "1". Similarly, since T interval (i - 4) is equal to T packet the quantization number is "1".

[0041] In this way, the quantized arrival interval time fluctuates as 1→1→0→1→1 (Fig. 6C).

[0042] Next, calculate the total quantization number of the quantized arrival interval time, that is, the total number of packets (step S13). In this case, 1 + 1 + 0 + 1 + 1 = 4. Next, calculate the average of the quantized arrival interval time as the estimated value of the number of packets (step S14). In this case, since the number of arrived packets is 5, the average value is 4 / 5. In this way, the quantized arrival interval time is flattened and the fluctuation of the arrival interval time is suppressed (Fig. 6D).

[0043] As a result, the real value of the arrival interval time is converted to an integer value, and noise can be reduced.

[0044] Next, the operation unit 1214 converts (returns) the quantized arrival interval time (integer value) back to a real value. That is, inverse quantization is performed. Specifically, the quantized arrival interval time (integer value) is used as the transmission interval time Tpacket Multiply it (step S15). In the above example, (4 / 5)×T packet is calculated.

[0045] Next, input the inverse-quantized arrival interval time into the moving average filter 1215 and perform a moving average (step S16). Here, the moving average value is calculated using the average value of T interval and the average value of T interval continuously obtained.

[0046] For example, as shown in FIG. 7, for the arrival interval times from an arbitrary (the i-th) arrival interval time T interval [i] to the arrival interval time N-1 arrivals ago T interval [i-N+1], a moving average is calculated.

[0047] Here, for the moving average, a simple moving average, a weighted moving average, an exponential moving average, etc. can be used.

[0048] The effect of the processing (moving average) of the moving average filter 1215 will be described below.

[0049] The cut-off frequency Fc of the moving average filter 1215 is designed to be lower than the Nyquist frequency. Here, when the Nyquist frequency is 1 / T packet =f packet , it is the frequency represented by f packet / 2.

[0050] As shown in FIG. 8, in the process of quantization and inverse quantization, quantization noise 151 is distributed over the entire frequency range. The quantization noise 151 is noise caused by the information (analog value) missing during quantization.

[0051] Also, when actually transmitting sensor data in packets, the packet has a high-frequency signal 152 such as a measurement signal and a low-frequency signal 153 including information regarding a time stamp. Here, the high-frequency signal 152 varies in seconds, and the low-frequency signal 153 varies in minutes.

[0052] By setting the cut-off frequency Fc of the moving average filter 1215 lower than the Nyquist frequency, the moving average filter 1215 functions as a low-pass filter, blocking the high-frequency signal 152, transmitting the low-frequency signal 153, and correcting the time stamp included in the low-frequency signal 153.

[0053] Here, for the high-frequency signal 152 that varies in seconds, since the low-frequency signal 153 varies in minutes, it is effective to set the cut-off frequency Fc to 1 / 60 or less of the frequency of the high-frequency signal 152.

[0054] Also, since the quantization noise 151 is uniformly distributed with respect to frequency, the noise can be reduced by f packet / 2Fc.

[0055] Finally, in the adder 1216, the initial time T 0 , which is an offset, is added to the moving average value of the calculated arrival interval time, thereby obtaining the time stamp correction value T correct (step S17).

[0056] Here, regarding the determination of the initial time T 0 , whether using the first arrival time or estimating using the least squares method or the like, only an absolute time error of about several tens of milliseconds occurs, so the influence is small compared to before application where a deviation occurs in units of seconds or more.

[0057] Also, in the time stamp correction system 121, by configuring the moving average filter 1215 in multiple stages, the slope 154 of the low-pass filter characteristics of the moving average filter 1215 becomes steeper, the jitter component can be reduced, and the high-frequency signal can be blocked more accurately and stably. On the other hand, there is a delay in tracking the variation of T packet , but this delay is not a problem in a biological data measurement system that cuts off the DC component.

[0058] Also, since the variation speed of T packet is very slow depending on temperature changes and aging deterioration, the influence of narrowing the filter bandwidth is small.

[0059] Also, in the timestamp correction method, the output of the moving average filter 1215 may be fed back to the step of quantizing the arrival interval time (step S12) to calculate the moving average value again (in FIG. 4, the dotted arrow). As a result, since T packet changes over time, by performing the calculation again using the updated T packet , the timestamp can be corrected with higher accuracy.

[0060] According to the timestamp correction system according to this embodiment, noise can be removed, delay can be alleviated, and the error of the timestamp can be reduced by the moving average filter.

[0061] Furthermore, when a packet transmission error occurs, two or more packets are continuously transmitted immediately, so T interval includes temporally correlated noise. Therefore, this noise can be easily removed by the moving average filter 1215. That is, it is possible to reduce jitter and obtain a timestamp with high accuracy by using the transmission interval time T packet as a reference rather than using the arrival time including noise and delay as it is.

[0062] Also, it is useful when real-time correction is required in a system where packets arrive on a stream. Also, it is useful in that the applicable application range is wider than that of a system that performs batch processing on a server.

[0063] Also, since signal processing is basically composed of delay and sum-of-products operations, acceleration processing by a DSP (Digital Signal Processor) can be applied, so it is suitable for real-time data processing.

[0064] Also, since correction is executed with a clock synchronized by GPS, NIP, NITZ, etc., high-precision correction can be achieved.

[0065] <Second Embodiment> A time stamp correction system, a sensor system, and a time stamp correction method according to a second embodiment of the present invention will be described with reference to FIGS. 9 and 10.

[0066] <Configuration and Operation of Time Stamp Correction System> The sensor system 20 according to the present embodiment has substantially the same configuration as that of the first embodiment, and the configuration of the time stamp correction system 221 is different.

[0067] FIG. 9 shows the configuration of the time stamp correction system 221 according to the present embodiment. FIG. 10 shows a flowchart of the time stamp correction method according to the present embodiment.

[0068] As shown in FIG. 9, the time stamp correction system 221 includes a multi-stage delay circuit 2211, a subtraction unit 2212, a packet number estimation unit 2213, an arithmetic unit 2214, a moving average filter 2215, and an addition unit 2216.

[0069] The time stamp correction system 221 receives the arrival time T arrival (time stamp T stamp ) attached to the packet.

[0070] The delay circuit 2211 is composed of M-stage delay circuits and delays the input T arrival (for example, T arrival [i - 1]). Here, T arrival is the same as T stamp .

[0071] The subtraction unit 2212 calculates the difference in arrival times, that is, the arrival interval time, using the outputs of the M-stage delay circuits (step S21). As a result, M arrival interval times are obtained.

[0072] The packet number estimation unit 2213 quantizes the arrival interval time and calculates the average value of this quantization number as an estimated value of the packet number (steps S22 to S24).

[0073] Next, in the arithmetic unit 2214, the quantized arrival interval time is multiplied by the transmission interval time T packet and inverse-quantized to convert it into a real value (step S25).

[0074] Here, as described above, since the arrival interval time is multiplied by M by the M-stage delay circuit, the quantized arrival interval time is multiplied by the T packet value and divided by M (step S26).

[0075] Next, the inverse-quantized and M-divided arrival interval time is input to the moving average filter 2215 to perform a moving average (step S27).

[0076] Finally, in the addition unit 2216, the initial time T that is the offset is added to the moving average value of the calculated arrival interval time 0 to obtain the time stamp correction value T correct (step S28).

[0077] Here, similar to the first embodiment, the output of the moving average filter 2215 may be fed back to the step of quantizing the arrival interval time (step S22) to calculate the moving average value again (in FIG. 10, the dotted arrow).

[0078] According to the time stamp correction system according to the present embodiment, noise can be removed, delay can be alleviated, and error of the time stamp can be reduced by the moving average filter.

[0079] Furthermore, by the multi-stage delay circuit, correlation noise can be removed before the number of packets is quantized. Also, since the M-multiplied data is made 1 / M times in the inverse quantization, noise itself due to the inverse quantization can be reduced.

[0080] Also, in the time stamp correction system, since M-stage delay is required in the process of estimating the number of packets, noise is reduced as the number of stages (M stages) of the delay circuit increases, but real-time performance decreases. Therefore, it is necessary to adjust the number of stages (M stages) of the delay circuit according to the application. However, for the T to be trackedpacket Since the fluctuations are in units of minutes, it is considered that a delay of about 10 seconds will not affect the real-time performance.

[0081] <Third Embodiment> The timestamp correction system and the sensor system according to the third embodiment of the present invention will be described with reference to FIG. 11.

[0082] <Configuration of the Sensor System> As shown in FIG. 11, the sensor system 30 according to the present embodiment includes sensors 31_1 to 31_M and portable information terminals 32_1 to 32_M such as smartphones. Each of the portable information terminals 32_1 to 32_M includes a timestamp correction system and a timestamp adding unit according to the first embodiment. Here, a timestamp correction system according to the second embodiment may be provided.

[0083] In the sensor system 30, the data acquired by the sensors 31_1 to 31_M is collected by the portable information terminals 32_1 to 32_M. The collected data is given a timestamp by the portable information terminals 32_1 to 32_M, this timestamp is corrected, and then transmitted to the network system 4 and processed in a form such as the cloud.

[0084] According to the sensor system of the present embodiment, noise can be removed, delay can be mitigated, and the error of the timestamp can be reduced by the timestamp correction system. In addition, since the correction is executed with a clock synchronized by GPS, NIP, NITZ, etc., high-precision correction can be achieved.

[0085] Furthermore, for example, in Android and iOS, which are representative operating systems of smartphones, since the resolution of timestamps is limited to 1 ms, the achievable jitter value may be limited. According to the sensor system according to the present embodiment, the accuracy is reduced to the number of arriving packets at an arbitrary time interval, and the accuracy of the timestamp itself does not become a problem. Therefore, when performing software processing on the operating system of a smartphone, the sensor system according to the present embodiment is effective.

[0086] In the sensor system according to the present embodiment, one smartphone may be connected to a plurality of sensors to assign timestamps. In this configuration, the influence of delay is large, and the influence on other applications is also large. Therefore, a configuration in which one smartphone is connected to one sensor is more desirable than a configuration in which one smartphone is connected to a plurality of sensors.

[0087] <Fourth Embodiment> The timestamp correction system and the sensor system according to the fourth embodiment of the present invention will be described with reference to FIG. 12.

[0088] <Configuration of Sensor System> As shown in FIG. 12, the sensor system 40 according to the present embodiment includes sensors 41_1 to 41_M, data collection terminals 42_1 to 42_N, and a server 43. Here, the data collection terminals 42_1 to 42_N include a timestamp adding unit, and the server 43 includes the timestamp correction system according to the first embodiment. Here, the timestamp correction system according to the second embodiment may be provided.

[0089] In the sensor system 40, the data acquired by the sensors 41_1 to 42_M is collected by the data collection terminals 42_1 to 42_N, and timestamps are added. The data with timestamps added is transmitted to the server 43, the timestamps are corrected by the server 43, and then transmitted to the network system 4 and processed in the form of a cloud or the like.

[0090] In this way, the sensor system 40 is configured such that the data collection terminals 42_1 to 42_N only perform timestamping, and the server 43 corrects the timestamp.

[0091] According to the sensor system according to the present embodiment, the timestamp correction system can remove noise, mitigate delay, and reduce the error of the timestamp. Further, since the data collection terminal can allocate resources only for timestamping, it is possible to prevent the degradation of the timestamp accuracy due to calculation in an environment where a plurality of sensors are connected to the data collection terminal.

[0092] Also, in a normal sensor network, the higher-level (closer to the network system) server has higher computing power, and the lower-level (farther from the network system) sensor has lower computing power. Therefore, it is useful to correct the timestamp with a server having high computing power in that resources can be concentrated.

[0093] In particular, in the sensor system according to the second embodiment, packet transmission failures increase when a plurality of sensors are connected. However, since the timestamp can be corrected and transmitted again after a packet transmission failure, noise becomes correlated and accuracy can be maintained.

[0094] In the embodiments of the present invention, in the configuration of the timestamp correction system and the sensor system, the timestamp correction method, etc., examples of the structure, dimensions, etc. of each component are shown, but the present invention is not limited thereto. Any configuration, method, etc. that can exhibit the functions of the timestamp correction system, the sensor system, and the timestamp correction method and achieve the effects may be used.

Industrial Applicability

[0095] The present invention relates to a timestamp correction system, a sensor system, and a timestamp correction method, and can be applied to a system for transmitting and receiving data acquired by a sensor and a communication system.

Explanation of Signs

[0096] 121 Timestamp Correction System 1211 Delay Circuit 1212 Subtraction Unit 1213 Packet Number Estimation Unit 1214 Arithmetic Unit 1215 Moving Average Filter 1216 Addition Unit

Claims

1. A timestamp correction system that corrects timestamps assigned to each of a plurality of packets when the plurality of packets transmitted at a predetermined transmission interval time are received, comprising: a delay circuit that delays the timestamp of one packet among the plurality of packets; a subtraction unit that calculates, as an arrival interval time, a difference between the timestamp of another packet received subsequently to the one packet and the timestamp of the delayed one packet; a packet number estimation unit that quantizes by comparing the arrival interval time with the transmission interval time, calculates a total number of quantization numbers of the quantized arrival interval time in a predetermined period, and calculates an average value of the quantization numbers by dividing the calculated total number of quantization numbers by the number of packets that have arrived in the predetermined period, and uses the average value of the quantization numbers as an estimated value of the number of packets; an arithmetic unit that multiplies the estimated value of the number of packets by the transmission interval time to calculate an inverse-quantized arrival interval time; a moving average filter that performs a moving average on the inverse-quantized arrival interval time; an addition unit that adds the timestamp of the one packet to the value obtained by the moving average A timestamp correction system comprising:

2. The delay circuit is multi-stage The timestamp correction system according to claim 1, characterized in that:

3. A sensor that transmits packets generated from a measurement signal at a transmission interval time; a timestamp assigning unit that receives the packet and assigns a timestamp; The sensor system comprising the timestamp correction system according to claim 1 or claim 2

4. The sensor; A mobile information terminal comprising the timestamp assigning unit and the timestamp correction system, and transmitting the corrected timestamp to a network The sensor system according to claim 3, comprising:

5. The sensor; A data collection terminal comprising the timestamp assigning unit; A server comprising the timestamp correction system and transmitting the corrected timestamp to a network The sensor system according to claim 3, comprising:

6. A method for correcting a timestamp assigned to a packet when a plurality of packets transmitted at a predetermined transmission interval time are received, comprising: a step in which a delay circuit delays the timestamp of one packet among the plurality of packets; ​ The subtraction unit calculates the difference between the timestamp of another packet continuously received following the one packet and the timestamp of the delayed one packet as the arrival interval time; The packet number estimation unit quantizes by comparing the arrival interval time with the transmission interval time; The packet number estimation unit calculates the sum of the quantization numbers of the quantized arrival interval times in a predetermined period, and calculates the average value of the quantization numbers by dividing the calculated sum of the quantization numbers by the number of packets that arrived in the predetermined period, and sets the average value of the quantization numbers as the estimated value of the number of packets; The calculation unit multiplies the estimated value of the number of packets by the transmission interval time to calculate the dequantized arrival interval time; The moving average filter executes a moving average on the dequantized arrival interval time; The addition unit adds the timestamp of the one packet to the value obtained by the moving average A timestamp correction method comprising the above.

Images