Method and device for determining distance and radial velocity of an object by means of radar signal

Abstract

The present invention relates to a method for determining distance (R) and radial velocity (v) of an object in relation to a measurement location, in which method radar signals are emitted and after reflection on the object are received again at the measurement location, wherein the emitted radar signals are subdivided within a measuring cycle into numerous segments (10) in which the frequency of the radar signals is gradually changed from an initial value (f.sub.A, f.sub.B) to the end value and each received reflected signal is subjected across one segment (10) to a first evaluation to detect frequency peaks and additionally a subsequent second evaluation of the signals for the frequency peaks of all segments (10) of the measuring cycle is carried out to determine a Doppler frequency component as a measure of the radial velocity (v). According to said method, an ambiguity in the determination of the relative velocity (v) is eliminated by subdividing the segments (10) into at least two groups (A, B), the initial value (f.sub.A, f.sub.B) of which and/or end value of the changing frequency are different, by subjecting the segments (11, 12) of each group (A, B) separately to the second evaluation and by determining a phase difference of the signals occurring during the second evaluation of the segments (11, 12) of each group (A, B) and corresponding to each other, thereby removing ambiguities in the determined velocity.

Claims

1. A method for determining the distance (R) and radial velocity (v) of an object in relation to a measurement location, with which radar signals are transmitted and following reflection at the object are received again at the measurement location, wherein the transmitted radar signals are subdivided within a measurement cycle into numerous segments, in which they are varied in their frequency from an initial value (f.sub.A, f.sub.B) to a final value, and the received reflected signals are subjected over a segment in each case to a first evaluation for detecting frequency peaks and additionally a subsequent second evaluation of the signals is carried out for the frequency peaks of all segments of the measurement cycle to determine a Doppler frequency component as a measure of the radial velocity (v), wherein the segments are subdivided into at least two groups (A, B) whose initial value (f.sub.A, f.sub.B) and/or final value of the varying frequency are different, where the segments of each group have the same form and the same initial values and final values, the segments of different groups are different with respect to initial values and final values, and the segments of each group (A, B) are separately subjected to the second evaluation and that elimination of ambiguities of the determined velocity is carried out by determining a phase difference of the mutually corresponding signals arising during the second evaluation of the segments of each group (A, B).

2. The method as claimed in claim 1, wherein the first evaluation is carried out as a first FFT using the sampling signals within a segment for determining the frequency peaks.

3. The method as claimed in claim 2, wherein the second evaluation is carried out as a second FFT using the mutually corresponding frequency peaks of the segments of the measurement cycle.

4. The method as claimed in claim 3, wherein determining the phase difference for the frequency peaks arising during the second FFT for the Doppler frequency takes place after at least two groups (A, B).

5. The method as claimed in any claim 1, wherein the segments of the two groups (A, B) are generated with the same real frequency profiles, but are used for a first group (A) from a first initial value (f.sub.A) to a first final value and for the second group (B) from a second initial value (f.sub.B) to a second final value for the measurement, wherein the initial values f.sub.A, f.sub.B and final values are different from each other.

6. The method as claimed in claim 1, wherein the segments of the at least two groups (A, B) all have the same frequency shift.

7. The method as claimed in claim 1, wherein the frequency change of the segments in the groups (A, B) is constant and of the same size.

8. The method as claimed in claim 1, wherein the determined phase difference is also evaluated for accurate determination of the distance (R) of an object.

9. A device for determining the distance and radial velocity of an object in relation to a measurement location, with a radar transmitter, a receiver disposed at the measurement location for radar signals of the radar transmitter reflected from the object, wherein the radar signals are subdivided within a measurement cycle into numerous segments, in which they are varied in their frequency from an initial value (f.sub.A, f.sub.B) to a final value, with a first evaluation device connected to the receiver for detecting frequency peaks within each of the segments of the received signal, with a second evaluation device connected to the first evaluation device for evaluation of a phase difference of the determined frequency peaks for determining a Doppler frequency component as a measure of the radial velocity (v), wherein segments of at least two groups (A, B) are used for the evaluation in the evaluation devices, the initial value (f.sub.A, f.sub.B) and/or final value of the varying frequency of said segments being different, where the segments of each group have the same form and the same initial values and final values, the segments of different groups are different with respect to initial values and final values, and the second evaluation device comprises at least two evaluation stages for the separate evaluation of the signals of the at least two groups (A, B) and that at least one phase difference detector, whose output signals can be used for unique determination of radial velocities, is connected to the at least two evaluation stages.

10. The device as claimed in claim 9, wherein the output signal of the phase difference detector is also evaluated for determination of the distance (R).

Description

(1) In a similar manner, initial or final segments of the received signals of the two groups can be “truncated” by discarding corresponding sampling values at the start or at the end of the sampling, i.e. leaving the same unevaluated. Graphical representations of exemplary embodiments are used in order to explain the invention. In the figures:

(2) FIG. 1 shows a curve profile and schematic evaluation signals for forming a range Doppler matrix by two dimensional FFT according to the prior art;

(3) FIG. 2 shows a curve profile according to the invention in accordance with an exemplary embodiment of the invention with two evaluation range Doppler matrices;

(4) FIG. 3 shows a first version for forming the modulation of the transmission signals according to the invention;

(5) FIG. 4 shows a second version for forming the modulated signals according to the invention.

(6) FIG. 2 contains a graphical representation of the profile of a transmission signal, wherein the frequency of the transmission signal f(t) is plotted against time t. The transmission signal consists of 2 L segments 10, which form two groups A, B of frequency ramps. The segments 11 of the first group A extend from an initial value f.sub.A over a modulation shift f.sub.SW, whereas the segments 12 of the second group B extend from an initial value f.sub.B with the same modulation shift (bandwidth) f.sub.SW. The segments 11, 12 of groups A, B adjoin each other alternately, so that all even-numbered segments belong to group A and all odd-numbered segments belong to group B.

(7) As in the prior art, a respective evaluation is carried out for each segment 10, preferably in the form of an FFT. Using a second evaluation, especially a second FFT, a range Doppler matrix is formed for the segments 11 of the first group on the one hand and for the segments 12 of the second group B on the other hand. there are thus different measured beat frequencies f.sub.Beat A and f.sub.Beat B for the two matrices.

(8) The transmission signal according to the invention consists initially of a classic transmission signal, i.e. of short rapid ramps, with a fixed specified ramp duration T.sub.chirp. However, the two groups of ramps A and B are transmitted in a nested “intertwined” mode. Only a very little changed lower carrier frequency is set between the first segments (ramps) 11 and the second segments (ramps) 12, differing e.g. by 10 kHz. Thus in the first group A in the exemplary embodiment the transmission signal is modulated from f.sub.0 to f.sub.0+100,000 MHz and in the other group of ramps B from f.sub.0+10 kHz to f.sub.0+100,010 MHz.

(9) The echo signals are mixed with the current transmission frequency in the baseband. The range Doppler matrices are generated for the two groups of ramps A and B. A target or object is accordingly observed and detected in both groups of ramps A and B in exactly the same cell of the two range Doppler matrices (RDM).

(10) Because the Doppler frequency analysis (second FFT) is carried out for each group of ramps A, i.e. over two ramp intervals in each case, the already small uniqueness range of the Doppler frequency in the prior art is halved again.

(11) However, owing to the measures according to the invention, this does not result in disadvantages. With the transmission signal according to the invention and the two lower carrier frequencies f.sub.A=f.sub.0 and f.sub.B=f.sub.0+10 kHz, the two range Doppler matrices for the two nested signals exist with the following spectra following the two-dimensional FFT:

(12) $\begin{array}{cc}{S}_A\left(m,n\right)=\underset{l=0}{\overset{L-1}{.Math.}}\underset{k=0}{\overset{K-1}{.Math.}}{e}^{\left(\mathrm{j2\pi }\left(f_{\mathrm{Beat}}\frac{k}{f_{\mathrm{sa}}}-f_{D,\mathrm{md}}{T}_{\mathrm{chirp}}2l+f_A\frac{2R}{c}\right)\right)}.Math.e^{-j2\pi \frac{k.Math.m}{K}}{e}^{-j2\pi \frac{l.Math.n}{L}}& \left(7\right)\\ S_B\left(m,n\right)=\underset{l=0}{\overset{L-1}{.Math.}}\underset{k=0}{\overset{K-1}{.Math.}}{e}^{\left(j2\pi \left(f_{\mathrm{Beat}}\frac{k}{f_{\mathrm{sa}}}-f_{D,\mathrm{md}}{T}_{\mathrm{chirp}}\left(2l+1\right)+f_B\frac{2R}{c}\right)\right)}.Math.e^{-j2\pi \frac{k.Math.m}{K}}{e}^{-j2\pi \frac{l.Math.n}{L}}& \left(8\right)\end{array}$

(13) In total 2L ramp signals 11, 12 are transmitted during this. All even-numbered ramps (group A) are associated with the signal S.sub.A, whereas the signal S.sub.B is composed of the odd-numbered ramps (group B) (2L+1). Compared to the known arrangement, the initial values f.sub.A and f.sub.B of the carrier frequencies in the two groups A, B are slightly shifted relative to each other. The segments (ramps) of a group A, B to be processed are separated from each other by a ramp length T.sub.chirp owing to the nested arrangement.

(14) In this situation there are two range Doppler matrices, which are evaluated for specific cells. For detection purposes the signals are simply added incoherently by magnitude for each cell. For each detected target, the frequency f.sub.Beat and the ambiguous Doppler frequency f.sub.D,md can be read directly from the range Doppler matrix or calculated by an interpolation technique for increased accuracy. In this respect there are two range Doppler matrices with identical magnitude information (but different phase infatuation).

(15) According to the invention, the phase difference per cell in the range Doppler matrix is now evaluated, advantageously only for those cells in which a target has been detected.

(16) Mathematically, this is given by

(17) $\begin{array}{cc}\begin{array}{c}\mathrm{\Delta \Phi }=\left[\left(\arg \left(\frac{S_A\left(m,n\right)}{S_B\left(m,n\right)}\right)-2\pi .Math.f_{D,\mathrm{md}}{T}_{\mathrm{chirp}}\right)\mathrm{mod}\pi \right]\\ =2\pi \left(\left(f_A-f_B\right)\frac{2R}{c}\right)\end{array}& \left(9\right)\end{array}$

(18) f.sub.D,md T.sub.chirp is a phase correction factor that arises owing to the (possibly ambiguous) measured Doppler frequency f.sub.D,md from ramp to ramp. The phase rotates further from ramp to ramp by said value. This must be taken into account for the evaluation of the received nested signal arrangement. The target distance R and hence f.sub.R can now be calculated from the above equation and the phase difference measurement as follows:

(19) $\begin{array}{cc}R=\frac{\mathrm{\Delta \Phi }}{2\pi }.Math.\frac{c}{2}\frac{1}{\left(f_A-f_B\right)}& \left(10\right)\\ f_R=-\frac{T_R}{T_{\mathrm{chirp}}}.Math.f_{\mathrm{sw}}=-\frac{2R}{c}.Math.\frac{f_{\mathrm{sw}}}{T_{\mathrm{chirp}}}=-\frac{\mathrm{\Delta \Phi }}{2\pi }.Math.\frac{1}{\left(f_A-f_B\right)}\frac{f_{\mathrm{sw}}}{T_{\mathrm{chirp}}}& \left(11\right)\end{array}$

(20) Finally, the unique Doppler frequency f.sub.D is given by the above equation taking into account the measured beat frequency f.sub.Beat and the measured phase difference:

(21) 0$\begin{array}{cc}{f}_D=f_R-f_{\mathrm{Beat}}& \left(12\right)\\ =-\frac{\mathrm{\Delta \Phi }}{2\pi }.Math.\frac{1}{\left(f_A-f_B\right)}\frac{f_{\mathrm{sw}}}{f_{\mathrm{chirp}}}-f_{\mathrm{Beat}}& \left(13\right)\end{array}$

(22) The evaluation of the measured phase difference results in a maximum unique measurable distance of

(23) $\begin{array}{cc}{R}_{\max }=\frac{1}{2}.Math.\frac{c}{2}\frac{1}{\left(f_A-f_B\right)}& \left(14\right)\end{array}$

(24) For a frequency difference (f.sub.A−f.sub.B) of 10 kHz, there is a maximum unique measurable distance of R.sub.max=7.5 km. For a frequency difference (f.sub.A−f.sub.B) of 4 kHz there is a maximum unique measurable distance of R.sub.max=18.75 km.

(25) Equation 11 thus results in not only an approximate but an accurate determination of the frequency relating to the distance R, which according to equations 12 and 13 enables accurate determination of the Doppler frequency f.sub.D in a unique manner.

(26) The use according to the invention of two groups A, B of segments 11, 12 with nested frequency shifts thus enables unique and accurate determination of the distance and the radial velocity by means of the determination of the Doppler frequency. The described transmission signal can be generated in the required manner by a suitably controlled frequency generator. However, it is also possible that the real segments 10, 11 can be generated in the same way, but using a different virtual modulation. For this purpose, according to FIG. 3 the so-called “zero filling” is used. The real modulated frequency shift is thereby f.sub.sw, but is not fully utilized. The frequency shift used in each case for the segments 11, 12 is f.sub.SW−(f.sub.B−f.sub.A).

(27) FIG. 3 shows that for the segments 10 a real modulation is always used, which starts from the initial value f.sub.A and extends over the entire bandwidth f.sub.SW. For the segments 11 of the first group A, the segment 11 starting with f.sub.A is used, whereas at the upper end a segment of width f.sub.B−f.sub.A is not used.

(28) For the segment 12 of group B, by contrast, the lower segment f.sub.B−f.sub.A is not used, so that the same frequency shift f.sub.SW−(f.sub.B−f.sub.A) occurs for both segments 11, 12.

(29) According to the version illustrated in FIG. 4, the segments 10 for both groups A, B are generated equally in real form. The length of the segments here is f.sub.SW+(f.sub.B−f.sub.A). The real modulated region is thus increased by f.sub.B−f.sub.A. Unused sampling values at the upper end of the segments 11 of the group A and at the lower end of the segment 12 of group B are discarded.

(30) In all the described cases, the segments 10, 11 have the same frequency shift and the same gradient. This is not absolutely necessary. Different frequency shifts and different gradients can also be used in the method described here. However, the mathematical evaluation for this is somewhat complex.