Processing received radiation reflected from a target

Abstract

A method of and apparatus for processing received radiation (e.g. RADAR radiation) reflected from a target, the method comprising generating a set of predicted targets, the set of predicted targets comprising at least one member, each member representing a state of the target, generating a predicted waveform for the radiation for each member dependent upon the state of the target, and comparing each predicted waveform with a waveform of the received radiation to determine the accuracy with which the state of the target represented by the member for which the predicted waveform was generated matches an actual state of the target.

Claims

1. A method of processing received radiation reflected from a target, the method comprising: generating a set of predicted targets, the set of predicted targets comprising at least one member, the at least one member representing a state of the target, generating a predicted waveform for the radiation for the at least one member dependent upon a state of the target, and comparing each predicted waveform with a waveform of the received radiation to determine an accuracy with which the state of the target represented by the at least one member for which the predicted waveform was generated matches an actual state of the target, in which the state of the target includes a parameter set comprising at least one parameter of the target, in which the step of generating the set of predicted targets comprises generating a set of members scattered through a parameter space defined by the parameter set, comprising, after the comparison between predicted and received waveforms, repopulating the set of predicted targets with members in the parameter space so that the members are scattered around the members of the parameter set before repopulating preferentially with increasing degree of correlation.

2. The method of claim 1, in which, no spectral analysis is made of the received radiation.

3. The method of claim 2, in which no fast Fourier transform, is made of the received radiation.

4. The method of claim 1, in which the parameter set includes positional data for the target.

5. The method of claim 4, in which the parameter set includes the position of the target.

6. The method of claim 1, in which the parameter set includes at least one of a velocity, acceleration and jerk.

7. The method of claim 1, in which the parameter set includes a measurement of an extent of the target.

8. The method of any of claim 7, in which the extent of the target is a width or size of the target.

9. The method of claim 1, comprising transmitting radiation from at least one transmitter, and receiving the received radiation at at least one receiver.

10. The method of claim 9 in which there are a plurality of receivers, in which the step of generating the predicted waveform comprises generating a predicted waveform for the received radiation received at each receiver and the step of comparing each predicted waveform comprises comparing the predicted waveform for each receiver with radiation received at that receiver.

11. The method of claim 10, comprising mixing radiation received at each receiver with radiation as transmitted by the transmitter and performing the comparison on the received radiation as mixed with the transmitted radiation.

12. The method of claim 9, in which the step of generating a predicted waveform comprises estimating the waveform of the transmitted radiation and applying at least one transform to the waveform of the transmitted radiation in order to arrive at the predicted waveform.

13. The method of claim 12, in which the at least one transform will depend on the parameter set.

14. The method of claim 12, in which the at least one transform comprises modifying a frequency and phase of the waveform dependent upon a range of the target from the at least one receiver.

15. The method of claim 12, in which at least one transform comprises transforming an amplitude of the waveform dependent upon on a position of the target.

16. The method of claim 9, comprising the step of steering the transmitted radiation and/or the at least one receiver based upon a position of the target.

17. The method of claim 1, in which the step of comparing each predicted waveform with the waveform of the received radiation comprises determining a correlation between the predicted waveform and the waveform of the received radiation.

18. The method of claim 1, in which members having a low degree of correlation are removed from the set of predicted targets.

19. The method of claim 1 comprising, after repopulating the set of predicted targets, repeating the step of comparing each predicted waveform to the waveform of the received radiation received subsequently to that used for the previous step of comparing.

20. The method of claim 19, in which the steps of repopulating and comparing repeat indefinitely.

21. The method of claim 1, in which the step of repopulating the set of predicted targets comprises updating the parameter set of each target based on an elapsed time between the reception of the original received radiation and the reception of the subsequently received radiation reflected.

22. The method of claim 1, in which at least some of the members and an associated degree of correlation thereof are output by the method as potential targets.

23. A reflection processing apparatus, comprising an input for received reflected radiation, a processor arranged to process the received radiation by carrying out the method of claim 1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a RARAR reflection processing apparatus in accordance with an embodiment of the invention installed in a vehicle.

(2) FIG. 2 shows a flow chart showing the operation of the processor of the apparatus of FIG. 1.

(3) FIG. 3a shows a first predicted waveform.

(4) FIG. 3b shows a second predicted waveform.

(5) FIG. 3c shows a third predicted waveform.

DETAILED DESCRIPTION OF THE DRAWINGS

(6) FIG. 1 of the accompanying drawings shows a RADAR reflection processing apparatus 1 in accordance with an embodiment of the invention.

(7) The apparatus 1 comprises an oscillator 2 which generates a frequency modulated continuous wave waveform u(t) and transmits that through transmitter circuit 3 and transmission antenna 4. The radio frequency electromagnetic radiation (radio waves) transmitted is directed ahead of a vehicle 100 on which the apparatus 1 is installed. The radiation can be reflected off any targets 101 that may be present in the area ahead of the vehicle.

(8) Two reception antennas 5, 6 are provided on opposing lateral sides of the vehicle 100. These collect radiation reflected from any targets 101. The received signals are detected using reception circuits 7, 8 and mixed using mixers 9 with the transmitted signal u(t). The mixed signals are each passed through respective low pass filters 10 and analogue to digital converters (ADCs) 11, before being passed to a processor 12.

(9) The processor carries out the steps shown in FIG. 2 of the accompanying drawings. In step 200, a set of target candidates is generated. In the order of a thousand to ten thousand candidates can be generated. Each candidate will have a position in a parameter space. The parameter space can have as many dimensions as desired. Typically, the dimensions can include position (shown as x and y in FIG. 1 of the accompanying drawings), speed v (and its direction), acceleration a (and its direction), jerk j (and its direction, jerk being the rate of change of acceleration), higher derivatives of jerk, the width w of the target 101 and its length 1.

(10) FIG. 1 of the accompanying drawings shows a set of target candidates 50 scattered through the parameter space; Xs show the original set of target candidates. Because only two dimensions are available in the accompanying drawings, the points are shown scattered only through two-dimensional space, but each would also be scattered throughout the other parameter dimensions discussed. The scattering can be random or pseudo random, or can be regularly spaced throughout the parameter space, or can be spaced throughout parameter space with increasing density with increasing probability of a target being found in that area. For example, it may be more likely for there to be a target directly in front of the vehicle 100, and so the scattering of target candidates throughout the positional element of parameter space may be denser in that area than to either side.

(11) At step 201, for each target, a predicted waveform for the radiation as received at each antenna 5, 6, received, mixed 9 and filtered 10 is generated using processor 12. In order to generate the expected time series return signal, an equation is required which will allow obtaining the predicted waveform of this filtered signal. For the simple example of a transmitted cosine wave with a linear increase of its frequency f over time (t is time, f.sub.1 is a base frequency,

(12) $\frac{f}{}$
is the rate of change of frequency with time):

(13) $f\left(t\right)=f_1+{}_0^t\frac{f}{}\mathrm{dt}$

(14) A function of phase over time can be found as follows:

(15) $\frac{}{\mathrm{dt}}=f\left(t\right).fwdarw.=f\left(t\right)\mathrm{dt}=\left[f_1+{}_0^t\frac{f}{}\mathrm{dt}\right]$

(16) After integrating:

(17) $\left(t\right)=2f_{1}t+t^2\frac{f}{}$

(18) The transmitted waveform is u(t)=cos((t)). The received waveform is simply a delayed version of the same signal: r(t)=cos((tt)).

(19) Using this expression the low frequency component of signal s(t) output by the mixer can be obtained as a function of time by multiplying transmitted and received signals and discarding high frequency terms, which will give the predicted waveform at any given instant:

(20) $s\left(t\right)=u\left(t\right)r\left(t\right)=\cos \left(t\right)\cos \left(t-t\right)=\cos \left(2t\frac{f}{}t+2f_{1}t-t^2\frac{f}{}\right)$

(21) For clarity, note that for any particular transmitted waveform exactly the same process can be carried out to obtain the expected mixer output waveform. The output is simply the multiplication of the transmitted waveform with a delayed version of the same waveform:
s(t)=u(t)r(t)=u(t)u(tt)

(22) The transmitted waveform can be modified to allow for the operation of the oscillator 2, transmission circuit 3 and antenna 4. For instance, if it is known that the oscillator 2 produces a FMCW modulation with what is nominally a linear saw-tooth ramp, but is non-linear to some extent, then that can be addressed in the definition of u(t). This is not the case with frequency domain FMCW analysis, where the accuracy of the system will be degraded with a non-linear modulation.

(23) The predicted waveform can also be modified to correct for the performance of the reception antennas 5, 6, receiver circuits 7, 8, mixers 9, filters 10 and ADC 11. For example, if the reception antennas give higher gain directly ahead of the vehicle 100, but lower gain to either side, then the amplitude of the predicted waveform will depend upon the position of the targets.

(24) Each of the parameters will have an effect on the predicted waveform. The position of the target will affect the delay between the transmitted and received signals, and so increasing range from the respective antenna 5, 6 will increase a phase shift between the transmitted and predicted signals. Increasing range may also decrease the amplitude of the predicted signals, in line with the inverse square law. Each of the predicted signals for the two antennas 5, 6 will have a different range, which can be used to triangulate the position of the target candidate relative to the vehicle.

(25) The speed, acceleration and jerk of the vehicle will affect the frequency of the output signal, in accordance with the continuously varying time delay, as shown in FIGS. 3a to 3c of the accompanying drawings. FIG. 3a shows a target candidate with zero relative speed (constant frequency), constant relative speed (higher frequency) and constant relative acceleration (increasing frequency).

(26) The width of the target will affect the amplitude of the received radiation; a wider target (having higher radar cross section) will have a higher amplitude response.

(27) Once each predicted waveform has been generated, at step 202, a comparison is made between each predicted waveform and the output of the ADC. The correlation between each predicted waveform and the output of the ADC is calculated. This indicates how accurately the state of each target candidate reflects the actual target 101. Thus, in FIG. 1, those targets shown with a circled X will correlate particularly highly. Target candidates with a high correlation can be output by the processor; the correlation can be output as well to indicate a confidence which the processor 12 has with each target candidate.

(28) In one embodiment where the transmitter antenna 4 and/or the reception antennas 5, 6 are steerable (for example, mechanically steerable, or phased array antennas, the antennas 4, 5, 6 may be directed to more particularly focus on any area where the correlation of targets in that area is particularly high. Where the antennas sweep across a field of view (FOV), having a dwell time in each area of the FOV, the dwell time (for subsequent iterations) may be higher for those areas with targets with higher correlation, and lower for those areas with targets with lower correlation (or which are lacking in targets).

(29) At step 203, the set of target candidates is repopulated. Typically, those target candidates with a low correlation will be removed. Those with a high correlation will have their parameters updated based upon the time elapsed since the last received radiation (because, due to the speed, acceleration and jerk of the candidates, they will have moved within parameter space). Further new target candidates will be added, concentrated around the successful candidates. In the example of FIG. 1 (and only depicting two-dimensional space), the new target candidates could be those indicated with a + sign; some will still be positioned away from the successful candidates, but less densely than closer to the successful targets.

(30) The method then repeats from step 201, with new predicted waveforms being generated and a comparison made to those predicted waveforms with newly-received radiation. Thus, each section of received radiation can be analysed as it is received; typically, prior art spectral analysis methods required 2.sup.n samples, where n was between 10 and 14, whereas the current method can process received data down to individual samples.

(31) As such, this method can have the following potential advantages over the prior art spectral analysis methods:

(32) No reliance on frequency domain processing so easier to understand based on simple time-series principles.

(33) Can process each return sample as it is captured. No need to capture blocks of data before processing. Reduces latency.

(34) Easier treatment of arbitrary waveform modulation.

(35) Ability to include higher order target motion models (that directly measure acceleration, jerk, higher order derivatives).

(36) Ability to include other target parameters (e.g. width).

(37) Ability to use information about antenna characteristics (e.g. sidelobes with differential gain) directly.

(38) Easy extension to multiple transmit and receive antennas (including arbitrary array patterns).

(39) Easy extension to 3-Dimensional target detection/tracking.

(40) Processing technique is very highly parallelisable.

(41) Easier to embed in low-cost hardware (e.g. FPGA).

(42) Scales easily for more complex systems.

(43) Ability to handle weak target returns due to removal of thresholding (where in spectral systems, the signal would be lost in noise; typically any frequency domain signal that is less strong than a threshold is discarded as noise).

(44) Whilst this embodiment has been described with reference to RADAR, it is equally applicable to LIDAR or SONAR or other such systems.

(45) In accordance with the provisions of the patent statutes, the principle and mode of operation of this invention have been explained and illustrated in its preferred embodiments. However, it must be understood that this invention may be practiced otherwise than as specifically explained and illustrated without departing from its spirit or scope.