Echo curve determination at a resolution that differs on area-by-area basis

Abstract

To determine an echo curve using a fill level measurement device operating according to the FMCW principle, the echo curve is calculated from corresponding sampling values at a first resolution. After this, a particular portion of the echo curve is calculated at a second, higher resolution using the DTFT algorithm. This can reduce the complexity required for calculating the echo curve.

Claims

1. A method for determining an echo curve using a fill level measurement device which operates according to a distance measurement method and carries out a spectral analysis of a measurement signal, which has been received by the fill level measurement device during the fill level determination, comprising the steps of: detecting the measurement signal, which is a transmission signal transmitted by the fill level measurement device and reflected at least at the surface of a medium; converting the detected transmission signal into an intermediate frequency signal; sampling the intermediate frequency signal at discrete times, resulting in sampling values; transforming the sampling values, obtained by the sampling, from a time range into a frequency range and calculating first points of an echo curve from the sampling values at a first spacing, and using the calculated first points to form the echo curve; calculating second, additional points on a portion of the echo curve within a defined region of the echo curve at a second spacing which is smaller than the first spacing, and using the calculated second, additional points to form a portion of the echo curve; determining a distance value from the surface of the medium using the second, additional points; and providing the distance value to an external interface of the fill level measurement device.

2. The method according to claim 1, wherein the sampling values obtained by the sampling are transformed from the time range into the frequency range and the second, additional sampling points on the portion of the echo curve are calculated using a discrete-time Fourier transform (DTFT).

3. The method according to claim 2, wherein the second, additional sampling points are each at a predetermined frequency spacing f from the sampling points adjacent thereto.

4. The method according to claim 1, wherein the transmission signal is a frequency-modulated signal, an electromagnetic signal or an optical signal.

5. The method according to claim 1, wherein the fill level measurement device operates according to a Frequency-Modulated Continuous Wave (FMCW) principle.

6. The method according to claim 1, wherein the portion of the echo curve corresponds to the location of an echo corresponding to the surface of the medium.

7. The method according to claim 1, wherein the first sampling points on the echo curve are calculated from the sampling values at the first spacing using a fast Fourier transform (FFT).

8. A program element which, when executed on a processor of a fill level measurement device, causes the processor to carry out steps of a method according to claim 1.

9. A non-transitory computer-readable medium, on which a program element is stored which, when executed on a processor of a fill-level measurement device, causes the processor to carry out steps of a method according to claim 1.

10. A high-resolution fill level measurement device which operates according to a distance measurement method and carries out a spectral analysis of a measurement signal, which has been received by the device during the fill level determination, comprising: an antenna configured to detect a measurement signal, which is a transmission signal transmitted by the device and reflected at least at the surface of a medium; a high-frequency unit configured to convert the detected transmission signal into an intermediate frequency signal; an external interface; a sampling unit configured to sample the intermediate frequency signal at discrete times resulting in sampling values; a processor; and a non-transitory computer-readable storage medium on which a program element is stored which, when executed by a processor, causes the fill level measurement device to: calculate first points from the sampling values at a first spacing, and using the calculated points to form an echo curve; calculate second, additional points from the sampling values at a second spacing which is smaller than the first spacing, and using the calculated second, additional points to form a portion of the echo curve; determine a distance value from the surface of the medium using the second, additional points; and providing the distance value to the external interface of the fill level measurement device.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a fill level measurement device installed in a container.

(2) FIG. 2A shows an intermediate frequency signal (beat signal) in the time range, which signal can be attributed to a reflected transmission signal received by a fill level measurement device.

(3) FIG. 2B shows the intermediate frequency signal of FIG. 2A following the conversion thereof into the frequency range.

(4) FIG. 3A shows the sampling of an intermediate frequency signal in the time range.

(5) FIG. 3B shows the signal of FIG. 3A following the conversion thereof into the frequency range.

(6) FIG. 4 shows a fill level measurement device according to an embodiment of the invention.

(7) FIG. 5 shows a flow diagram of a method according to an embodiment of the invention.

(8) FIG. 6 shows sampling values of an echo curve in the frequency range.

(9) FIG. 7 shows the calculation of additional sampling points on the echo curve of FIG. 6 within a defined region of the echo curve according to an embodiment of the invention.

DETAILED DESCRIPTION OF EMBODIMENTS

(10) The drawings are schematic and not to scale. In the following description of the drawings, where like reference numerals are used in different figures, they denote like or similar elements. However, like or similar elements may also be denoted by different reference numerals.

(11) FIG. 1 shows a fill level measurement device 101 in the form of a fill level radar, which is installed in or on a container.

(12) It should be mentioned at this point that the method described below can also be carried out by other fill level measurement devices which run a spectral analysis during the signal evaluation, for example fill level measurement devices which operate according to the guided microwave principle, or ultrasonic measurement devices or laser measurement devices.

(13) The fill level measurement device in FIG. 1 is, for example, an FMCW radar device. Particular importance is given to the FMCW fill level measurement method owing to the integrated high-frequency components which are now available for fill level measurements in the W band (from 75 GHz to 110 GHz). However, said method can also be used in all other frequency bands.

(14) Via the antenna 102, the fill level measurement device 101 emits, towards a filling material surface 105, a frequency-modulated transmission signal 104 which has been generated by the high-frequency unit 103. The transmission signal is reflected at the surface 105 of the medium 106 and then spreads out towards the fill level measurement device 101, where it is received again by the antenna 102 and forwarded to the high-frequency unit 103.

(15) Using the signal currently being transmitted, the received signal is converted in the high-frequency unit 103 into an intermediate frequency signal, which substantially still only comprises low-frequency signal portions.

(16) By way of example, FIG. 2A shows an intermediate frequency signal or beat signal 201 which has been generated by the fill level measurement device 101 from the received, reflected transmission signal (measurement signal) and the emitted transmission signal, the amplitude 205 of which beat signal is mapped as a function of the signal transit time t 207.

(17) The intermediate frequency signal 201 is provided continuously by the high-frequency unit 103 (cf. FIG. 1) during a measurement cycle of length T.sub.M.

(18) An A/D converter 107 connected to the high-frequency unit 103 samples the provided intermediate frequency signal 201 during a measurement cycle and stores the resultant digitised amplitude values of the intermediate frequency signal 201 in a memory, which is not shown here but is contained in the fill level measurement device 101.

(19) Once the actual measurement cycle has finished, an echo curve generation unit 108 calculates an echo curve from the stored amplitude values for the intermediate frequency signal. Depending on the method, it may be expedient to, in this step, transform the intermediate frequency signal from the time range into the frequency range.

(20) The curve 202 in FIG. 2B shows the result of converting the intermediate frequency signal 201 into the frequency range. Here, the amplitude 206 of the signal 202 is plotted as a function of the frequency f 208.

(21) In a further step, the echo curve 202 thus obtained is forwarded to an evaluation unit 109 (cf. FIG. 1) which, in accordance with known methods, identifies the echo 203 that is highly likely to correspond to the reflection at the filling material surface 105, and passes said echo on to a measurement unit 110.

(22) Using the rough frequency location of both the useful echo 203 and the echo curve 202, the measuring unit 110 determines the exact frequency that can be assigned to the echo pulse 203. Known methods are again used in this step. On one hand, the position of the maximum of the echo pulse 203 can be determined (f.sub.Target 204). Furthermore, it may also be possible to select another measurement point, for example the point on the rising echo flank that has an amplitude difference of exactly 3 dB compared with the maximum of the echo pulse 203.

(23) The frequency value 204 determined thus can be transformed into a distance value 112 using known equations.

(24) This distance value d is optionally then linearised and scaled in an output unit 111 of the fill level measurement device 101, and then provided to the outside on a suitable interface 113 digitally via a HART line, a profibus, a foundation fieldbus (FF), the Ethernet, a USB interface and/or even in analogue form via a 4 . . . mA loop.

(25) The relationship shown in FIG. 2B shows an ideal situation that can only be achieved approximately when the signals are processed digitally. FIG. 3A illustrates the problems with digital signal processing.

(26) The time-continuous beat signal 201 which is provided by the high-frequency unit 103 and has the length T.sub.M is only detected at certain times 302, 303, 304, etc. owing to the samples in the A/D converter 107.

(27) In FIG. 3A, the values determined in the process are marked by corresponding circles 305, 306, 307, etc.

(28) These values are converted into the frequency range by using, for example, a fast Fourier transform (FFT). The values determined therefrom are shown in FIG. 3B.

(29) The computed values have a frequency spacing of 1/T.sub.M. The frequency-continuous spectrum 202 of the beat signal 201 is calculated by the fast Fourier transform only for individual frequency values (see the circle marks in FIG. 3B).

(30) During processing of the signal propagation cycle in the measurement device 101, a false value 301 is now determined for the maximum of the useful echo 203, the result of which is a relatively large measurement error owing to the difference between the associated frequency value and the actual target frequency 204.

(31) To prevent this, zero padding can be used. In the process, prior to the calculation of the fast Fourier transform, additional sampling values having an amplitude value of 0 are computationally appended to the actual beat signal. Following the FFT, additional calculated sampling points on the echo curve 202 are thus produced.

(32) For example, up to 4096 real sampling values (or even more) of the beat signal can thus be detected. Therefore, to calculate three additional intermediate values, 3 times 4096 zeros would have to be appended to the signal. The transformation of 4096 values thus results in a transformation of 16384 values, which firstly places great demands on the size of the memory in the fill level measurement device and secondly can also lead to very long calculation times and, as a result, high levels of energy consumption.

(33) When even greater demands are placed on the accuracy of the echo curve, the complexity increases just as sharply.

(34) Alternatively to zero padding, a method can be carried out which allows for any number of sampling points on the spectrum 202 of the intermediate frequency signal 301 to be determined, without excessively increasing the demands on the memory space required therefor in the process.

(35) Applying this method in the field of energy-optimised two-wire measurement devices is particularly advantageous.

(36) FIG. 4 shows a fill level measurement device 401 which has been modified with respect to FIG. 1. This fill level measurement device differs from the fill level measurement device in FIG. 1 on account of a modified echo curve generation unit 402 and a modified measurement unit 403, which are interconnected via the data line 404.

(37) It should be noted at this point that the different signal processing units that the fill level measurement device 401 comprises can also be integrated in one single unit.

(38) By way of example, FIG. 5 shows a sequence as can be carried out in a fill level measurement device 401 according to an embodiment of the invention.

(39) FIGS. 6 and 7 also illustrate important intermediate results from the method described below.

(40) The method starts in the start state 501 (cf. FIG. 5). As with a conventional fill level measurement device, in step 502 a beat curve is firstly generated, digitised and stored in the memory of the fill level measurement device 401.

(41) In the optional step 503, the beat signal, which is in digital form, is weighted with a known window function, e.g. a Hamming window, a Bartlett window or another window. The use of window functions can improve the display in the spectral range.

(42) In step 504, the beat signal of the modified echo curve generation unit 402 (cf. FIG. 4) is converted into the spectral range using the fast Fourier transform. In this respect, the hardware configuration of the fill level measurement device 401 may be capable of doing this in a specific way, for example by using a digital signal processor having a specific hardware unit for calculating an FFT.

(43) In step 505, the location of the echo 203 corresponding to the filling material surface is determined according to known methods. The location of the echo can, for example, be defined by the frequency of the highest value 601 in the spectral range. In step 506, the modified measurement unit 403 determines the frequency values of the adjacent two, three, four or more sampling values 602, 603 of the previously identified sampling value 601 having the greatest amplitude. It is assumed that these sampling values are the sampling values for the filling material echo 203.

(44) These frequency values are communicated to the modified echo curve generation unit 402 of the fill level measurement device in FIG. 4, which unit then calculates additional sampling points 702 on the echo curve 202 in the region of the filling material echo 203 in a predeterminable frequency grid 701 (cf. FIG. 7) and communicates these to the modified measurement unit 403.

(45) The portion 704 of the echo curve within the defined region 703 can then be calculated at a higher resolution from these additional sampling points 702.

(46) The method used for this purpose is known as a discrete-time Fourier transform (DTFT). The results of this calculation agree with those of the fast Fourier transform in conjunction with zero padding. Since it is necessary to append zeros for said fast Fourier transform, the storage space required therefor is much greater but can be very limited, especially in commercially available, energy-optimised digital signal processor solutions.

(47) In step 507, in accordance with known methods, the measurement unit 403 determines the distance 112 from the filling material surface 105 (cf. FIG. 4) using the sampling points 702 which have been additionally computed. The method finishes once the measurement value 508 has been output.

(48) For the sake of completeness, it should be noted that comprising and having do not exclude the possibility of other elements or steps, and one or a does not exclude the possibility of a plurality. It should further be noted that features or steps which have been described with reference to one of the above embodiments may also be used in combination with other features or steps of other above-described embodiments. Reference numerals in the claims should not be treated as limiting.