KR20100052264A - Method of maximizing accuracy of measurement for saw type tpms sensor - Google Patents

Abstract

PURPOSE: A method of maximizing the measurement precision of a SAW type TPMS sensor is provided to maximize the measurement precision of a SAW type TPMS sensor by searching a precise peak frequency. CONSTITUTION: A method of maximizing the measurement precision of a SAW type TPMS is as follows. An answer signal is received from a sensor. The signal is sampled by the analog-digital conversion. The FFT(fast Fourier transform) is practiced on the sampled signal to develop the answer signal into the frequency spectrum. The frequency of the peak point with the maximum amplitude on the frequency spectrum is searched(310). The searched peak frequency is decided as the frequency of the answer signal from the sensor. The method comprises a calculating step, the peak frequencies is searched on the frequency spectrum. The frequency pattern of the frequency spectrum is analyzed. The frequency component of a second peak point existing at left and right sides of the peak point of the frequency is analyzed. The peak frequency of the answer signal is taken.

Description

Method of maximizing accuracy of measurement for SAW type TPMS sensor

The present invention relates to a method for maximizing the measurement accuracy of a SAW type TPMS sensor, and specifically, to maximize the accuracy in measuring the input frequency by comparing the magnitudes of the left and right peak points in addition to the center peak point of the FFT transformed frequency component. It is about a method.

In a tire pressure measuring system (TPMS), various types of sensors have been developed for measuring air pressure and temperature in a tire. In most cases, the sensor is attached to a car wheel or tire to sense the pressure or temperature inside the tire and output it to the outside.

Some known sensors use a power battery, but in this case, the size and weight of the sensor itself increases according to the weight and size of the power battery, which may affect the wheel balance when attached to the wheel. There is a problem such as the need to replace the battery when the end of life.

On the other hand, according to the prior art possessed by the applicant of the TPMS sensor of the SAW method that does not use a battery is known.

In FIG. 1A, a sensor 10 is attached to each wheel of a vehicle, and one reader 20 is installed in the vehicle. Each sensor 10 and reader 20 are each equipped with an antenna to communicate with each other RF. 1B conceptually illustrates one sensor 10 and a reader 20. When the reader 20 transmits an interrogation signal of a predetermined frequency to the sensor 10 as an RF signal, the sensor 10 transmits a response signal of a predetermined frequency resonant with the interrogation signal. The reader 20 receives the response signal and measures pressure and temperature information in the tire in which the sensor 10 is installed. For example, when the reader 20 transmits an RF wave of a predetermined frequency A MHz (for example, 433 MHz) as an inquiry signal, the sensor 10 responds to the pressure or temperature in the tire as a response signal resonated thereto. A + α) MHz (for example, 433.5 MHz), (A + β) MHz (for example, 434 MHz) and the like are output. The reader 20 receives this response signal and reads the frequency (peak frequency) to calculate the pressure and temperature of the tire on which the sensor 10 is installed.

1C briefly shows an internal configuration of the reader 20. It is controlled by the microcomputer 22 as a whole and consists of a transmitter 24 and a receiver 28. The transmission signal (that is, the inquiry signal) is RF-modulated by the RF unit 26 and transmitted from the transmitter 24 to the outside through an antenna. The response signal transmitted from the outside (i.e., the sensor 10) is received by the receiver 28, and the micom 22 performs reception value analysis, measurement value calculation, and the like.

In the microcomputer 22 of the conventional non-powered SAW resonant TPMS reader described with reference to FIGS. 1A to C, the peak frequency value of the response signal is measured in the same manner as in FIG. 2A. Upon receiving the response signal from the sensor 10, the signal is subjected to analog-digital conversion (ADC) sampling and fast Fourier transform (FFT) to expand the response signal into the frequency spectrum. The peak frequency is searched in this frequency spectrum, and the peak frequency is determined as the frequency of the response signal from the sensor and calculated.

In this case, the resolution f _res of the output frequency converted from the sampled response signal for FFT execution is inversely proportional to the duration T _spls of the sampled response signal. That is, f _res = 1 / T _spls . Therefore, as shown in FIG. 2B, the duration of the response signal received from the sensor is only about 20usec at most, so the frequency resolution after FFT of the response signal using the conventional sensor is f _res = 1 / T _spls = 1 / 20usec = 50kHz. Becomes

This frequency resolution of 50 kHz is a very large value. That is, since the peak frequency can be found only in units of 50 kHz or more in the search for the peak frequency in the spectrum converted from the response signal to the frequency domain, the precision is inferior.

As shown in the above equation, if the duration of the response signal from the sensor is long enough, the resolution at the FFT may be reduced. However, it is not possible to make the response signal duration of the existing SAW type TPMS sensor larger than 20usec.

An object of the present invention is to provide a method of maximizing the precision in peak frequency measurement by increasing the frequency resolution of the response signal transmitted from the conventional TPMS sensor using SAW.

A method of measuring pressure and temperature information in a tire in which a sensor is installed by sending an inquiry signal of a predetermined frequency to a sensor attached to a wheel of a vehicle and receiving a response signal sent from the sensor, the conventional SAW type TPMS sensor The present invention for the purpose of improving the peak frequency search method of the response signal,

Upon receiving a response signal from the sensor, sampling the signal to analog-digital conversion (ADC) and performing a fast fourier transform (FFT) on the signal to expand the response signal into the frequency spectrum, and the maximum amplitude on the frequency spectrum. Searching for a frequency (peak frequency) of a peak point having a peak, and determining and calculating the found peak frequency as a frequency of a response signal from the sensor, thereby increasing the frequency resolution in the FFT-converted frequency spectrum. A method of maximizing measurement accuracy of a SAW type TPMS sensor is disclosed.

Here, the step of searching for the peak frequency on the frequency spectrum, by analyzing the frequency pattern of the frequency spectrum, by analyzing the frequency components of the second peak point present on both sides of the peak point of the frequency peak frequency of the response signal Estimating may be further included. At this time, the peak frequency estimating step,

Extracting a frequency of a central peak point having a maximum amplitude in the frequency spectrum,

The secondary peak point located adjacent to the peak point (left and right sides) is examined, and the frequency f _{0 at the} center peak point, the magnitude of the output power (amplitude) Y ₀ at that point, the frequency f _{−1 at the} left peak point, and The output magnitude Y _{-1 of the} point, the frequency f ₊₁ of the right peak point, and the output magnitude Y ₁ value of the point are acquired as basic data, and the amplitude difference between the center peak point and the left peak point A _- at each obtained amplitude value- Acquiring an amplitude difference A ₁ (= Y ₀ -Y ₁ ) between ₁ (= Y ₀ -Y _-1 ) and the right peak point,

It is checked whether these basic data are A ₋₁ ≒ A ₁ (530), and if so, it is determined that f _in = f ₀ and outputs an f _in value; If it is not A ₋₁ 가 A _{1, it} is again checked whether A ₁ ≒ 0 (540). In this case, f _in = f ₀ + (f ₁ -f ₀ ) / 2 is calculated to output the value of f _in ; If A ₁ ≒ 0 is not checked again, check whether A _-1 <A _1, and if so, calculate f _in = f _0- (1-A _-1 / A ₁ ) * (f ₁ -f ₀ ) / 2 outputs the value of f _in ; If A _-1 <A ₁ is not checked again, check whether A _-1> A ₁ , _{in which} case f _in = f ₀ + (1-A ₁ / A _-1 ) * (f ₁ -f ₀ ) / 2 Calculating and outputting the value of f _in .

Here, the method may further include an analog-digital conversion (ADC) sampling of the response signal received from the sensor, and adding a silent signal having a duration longer than the duration of the response signal in the time axis direction to the rear of the sampled signal.

According to the present invention, it is possible to maximize the measurement accuracy of the SAW type TPMS sensor by searching for a precise peak frequency by increasing the frequency resolution of the response signal transmitted from the existing SAW type TPMS sensor.

First embodiment

The first embodiment of the present invention is an improvement on the conventional peak frequency search step shown in Fig. 2A. In FIG. 2A, the FFT is transformed into a frequency spectrum to find a peak frequency, and the peak frequency of the maximum amplitude is searched on the spectrum. However, in the present embodiment, this is further developed to find a more precise peak frequency.

As a method, the peak frequency of the response signal is estimated by analyzing the frequency pattern FFT-converted into the frequency domain instead of simply finding the maximum peak point. .

First, the basis of this estimation method will be described. According to experimental considerations, the case shown in FIG. 4A is A ₋₁ ≒ A ₁ , indicating that the input frequency (ie, frequency before FFT execution) f _in is the same as the output frequency (ie frequency after FFT execution) f _0. (I.e. f _in = f ₀ ), the relationship f ₀ <f _in <(f ₀ + Δf / 2) holds for A ₋₁ > A ₁ in FIG. 4B, and A for FIG. 4C. _{When 1} is almost equal to the peak point (that is, A ₁ ≒ 0), the relationship of f _in = (f ₀ + Δf / 2) is established, and in the case of FIG. 4D, when A ₋₁ <A ₁ ( It was found that the relationship of f ₀ -Δf / 2) <f _in <f ₀ was established.

Therefore, if the frequency pattern is analyzed as in the case of FIGS. 4A to 4D, it will be possible to estimate the peak point more accurately than simply finding the frequency peak point. The present invention proposes an algorithm that can estimate the peak frequency by analyzing such a frequency pattern as shown in FIG.

In FIG. 5, first, a surface detectable center peak point is extracted from a frequency spectrum obtained after performing FFT on a response signal (510). The secondary peak points (left and right peak points) located adjacent to the peak portion (left and right peaks) are examined (520) to determine the frequency f ₀ of the center peak point and the output power (amplitude) of the point. The magnitude Y ₀ , the frequency f ₋₁ of the left peak point and the output magnitude Y ₋₁ , the frequency f ₊₁ of the right peak point, and the output magnitude Y ₁ value of the point are obtained as basic data, Calculate the amplitude difference A- ₁ (= Y ₀ -Y _-1 ) and A ₁ (= Y ₀ -Y ₁ ) between the peak point and the center peak point (525). These basic data are checked whether A ₋₁ ≒ A ₁ (530). In this case, it is determined that f _in = f ₀ and the f _in value is output (535). If it is not A _-1 ≒ A _{1, it} is again checked whether it is A ₁ ≒ 0 (540). In that case, f _in = f ₀ + (f ₁ -f ₀ ) / 2 is calculated and the f _in value is output (545). ). If A ₁ ≒ 0 is not checked, check again whether A _-1 <A ₁ (550), _{in which} case f _in = f _0- (1-A _-1 / A ₁ ) * (f ₁ -f ₀ ) / 2 Calculate and output the f _in value (555). If A _-1 <A ₁ is not checked again, check whether A _-1> A ₁ (560), _{in which} case f _in = f ₀ + (1-A ₁ / A _-1 ) * (f ₁ -f ₀ ) Compute / 2 and output the f _in value (565). By the above algorithm, more accurate peak frequency can be estimated from the response signal from the sensor.

Second embodiment

3 is a process flowchart of a second embodiment of a method for maximizing measurement accuracy of a SAW type TPMS sensor according to the present invention. In a second embodiment, a step 300 of adding a null signal between the ADC sampling step and the FFT execution step is further included in the existing process shown in FIG. 2A. The virtual signal of a predetermined duration is added to the rear of the signal sampled from the received response signal.

The resulting resolution of the output frequency is then determined by the duration T _spls of the sampling frequency and the duration T _null of the silence signal. Here, the null signal is a virtual signal having a duration longer than the duration of the response signal, which is added to the time axis direction of the response signal in order to relatively increase the duration of the response signal from the existing SAW type TPMS sensor up to 20usec. It is defined as.

For example, if you add a silent signal with a duration of 236 msec, the resolution of the FFT frequency is f _res = 1 / (T _spls + T _null ) = 1 / (20usec + 236msec) = 3.9 kHz, much more than the existing 50 kHz. Frequency resolution is reduced. That is, since the frequency resolution in the frequency spectrum converted from the sensor response signal is reduced, it is possible to calculate the peak frequency more precisely on this spectrum.

However, in the second embodiment of the present invention, when the silence signal is too long in msec or more, the FFT analysis time is lengthened accordingly, and thus, the speed of the processor must be followed accordingly (one time FFT analysis time). Takes tens of thousands of msec). Therefore, in practical application, it is desirable to limit the silence signal to several tens of usec lengths (about 66usec length).

The second embodiment can of course be practiced alone, but since the effect is weaker than that of the first embodiment, it will be more preferable to use it in addition to the first embodiment.

1a to 1c illustrate a conventional SAW type TPMS sensor

2A and 2B illustrate an operation pattern of a reader in a conventional SAW type TPMS sensor.

3 illustrates a method according to a second embodiment of the present invention.

4a to 4d illustrate the basic principle of the first embodiment of the present invention.

5 is a method process flow diagram of a first embodiment of the present invention.

Claims (5)

In a method for measuring the pressure and temperature information in the tire in which the sensor is installed by sending a query signal of a predetermined frequency to the sensor attached to the wheel of the vehicle, receiving a response signal sent from the sensor, Receiving a response signal from the sensor, analog-digital conversion (ADC) sampling the signal, and performing a fast fourier transform (FFT) on the sampled signal to expand the response signal into a frequency spectrum, Searching for a frequency (peak frequency) of a peak point having the maximum amplitude on the frequency spectrum, and determining the found peak frequency as a frequency of a response signal from a sensor, and calculating the calculated peak frequency. Searching for the peak frequency on the frequency spectrum, by analyzing the frequency pattern of the frequency spectrum, by analyzing the frequency components of the second peak point present on both sides of the peak point of the frequency to estimate the peak frequency of the response signal And increasing the frequency resolution in the FFT-converted frequency spectrum. The method of claim 1, wherein the peak frequency estimating step included in the peak frequency searching step comprises: Extracting a frequency of a peak point having a maximum amplitude in the frequency spectrum, The secondary peak point located adjacent to the peak point (left and right sides) is examined, and the frequency f _{0 at the} center peak point, the magnitude of the output power (amplitude) Y ₀ at that point, the frequency f _{−1 at the} left peak point, and The output magnitude Y _{-1 of the} point, the frequency f ₊₁ of the right peak point, and the output magnitude A ₁ value of the point are obtained as basic data, and from these values the amplitude difference A _-1 ( Calculating = Y ₀ -Y _-1 ) and A ₁ (= Y ₀ -Y ₁ ), It is checked whether or not A _-1 ? A ₁ for these basic data, and if so, it is determined that f _in = f ₀ and outputs the value of f _in ; If A _-1 ≒ A ₁ is not checked again, and if it is A ₁ ≒ 0, then f _in = f ₀ + (f ₁ -f ₀ ) / 2 is calculated and the value f _in is output; If A ₁ ≒ 0 is not checked again, check whether A _-1 <A _1, and if so, calculate f _in = f _0- (1-A _-1 / A ₁ ) * (f ₁ -f ₀ ) / 2 outputs the value of f _in ; If A _-1 <A ₁ is not checked again, check whether A _-1> A ₁ , _{in which} case f _in = f ₀ + (1-A ₁ / A _-1 ) * (f ₁ -f ₀ ) / 2 Method for maximizing the measurement accuracy of the SAW-type TPMS sensor, including the step of calculating and outputting the f _in value. The method according to claim 1 or 2, wherein after the analog-digital conversion (ADC) sampling of the response signal received from the sensor, a silent signal having a duration longer than the duration of the response signal in the time axis direction is provided at the rear of the sampled signal. The method of maximizing the measurement accuracy of the SAW type TPMS sensor, characterized in that further comprising the step of adding. In a method for measuring the pressure and temperature information in the tire in which the sensor is installed by sending a query signal of a predetermined frequency to the sensor attached to the wheel of the vehicle, receiving a response signal sent from the sensor, Receiving a response signal from the sensor, analog-digital conversion (ADC) sampling the signal, and adding a silent signal having a duration longer than the duration of the response signal in the time axis direction at the rear of the sampled signal; Performing a fast Fourier transform (FFT) on the sampled signal to which the silent signal is added to expand the response signal into a frequency spectrum, Searching for a frequency (peak frequency) of a peak point having the maximum amplitude on the frequency spectrum, and determining the found peak frequency as a frequency of a response signal from a sensor, and calculating the calculated peak frequency. Searching for the peak frequency on the frequency spectrum, by analyzing the frequency pattern of the frequency spectrum, by analyzing the frequency components of the second peak point present on both sides of the peak point of the frequency to estimate the peak frequency of the response signal And reducing the frequency resolution in the FFT-converted frequency spectrum. The method of claim 4, wherein the peak frequency estimating step included in the peak frequency searching step comprises: Extracting a frequency of a peak point having a maximum amplitude in the frequency spectrum, The secondary peak point located adjacent to the peak point (left and right sides) is examined, and the frequency f _{0 at the} center peak point, the magnitude of the output power (amplitude) Y ₀ at that point, the frequency f _{−1 at the} left peak point, and The output magnitude Y _{-1 of the} point, the frequency f ₊₁ of the right peak point, and the output magnitude Y ₁ value of the point are acquired as basic data, and from these values, the amplitude difference A _-1 ( Calculating = Y ₀ -Y _-1 ) and A ₁ (= Y ₀ -Y ₁ ), It is checked whether or not A _-1 ? A ₁ for these basic data, and if so, it is determined that f _in = f ₀ and outputs the value of f _in ; If A _-1 ≒ A ₁ is not checked again, and if it is A ₁ ≒ 0, then f _in = f ₀ + (f ₁ -f ₀ ) / 2 is calculated and the value f _in is output; If A ₁ ≒ 0 is not checked again, check whether A _-1 <A _1, and if so, calculate f _in = f _0- (1-A _-1 / A ₁ ) * (f ₁ -f ₀ ) / 2 outputs the value of f _in ; If A _-1 <A ₁ is not checked again, check whether A _-1> A ₁ , _{in which} case f _in = f ₀ + (1-A ₁ / A _-1 ) * (f ₁ -f ₀ ) / 2 Method for maximizing the measurement accuracy of the SAW-type TPMS sensor, including the step of calculating and outputting the f _in value.

Images