Radar sensor and method of operating a radar sensor

Abstract

A radar motion sensor detects motion using a currently set sampling frequency, the detected motion being associated with a Doppler frequency. It is determined where the Doppler frequency of the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency. The sampling frequency is adapted over time based on the Doppler frequency of previously detected motions. In this way, the sampling frequency used by the radar transmitter is adapted by analyzing the relationship between the detected Doppler frequency and the sampling frequency. The system can thus adapt to the conditions in which it is used, such as the typical speed of the motions that it detects.

Description

[0001] RADAR SENSOR AND METHOD OF OPERATING A RADAR SENSOR

[0002] FIELD OF THE INVENTION

[0003] This invention relates to radar sensors, for example for motion sensing in order to detect the presence (and optionally also movement characteristics) of an object or person. The detection is used for automating control of an apparatus.

[0004] BACKGROUND OF THE INVENTION

[0005] The use of motion sensing is well-known for controlling an apparatus. For example, motion sensing, in order to provide presence detection, is used for controlling lighting systems, alarm systems, camera and security systems, and smart home devices.

[0006] The most common low-cost motion detectors use passive infrared (PIR) sensors. Radar sensors are more sensitive than PIR sensors and thus can give improved performance. However, radar sensors, as active RF devices, consume significantly more power than PIR sensors. This is a major limitation in the use of radar sensors to replace PIR sensors, especially for applications with a tight power budget (e.g., solar or battery solutions).

[0007] It is also more and more important to minimize the power consumption of Internet of Things (loT) devices, such as sensors in lighting products, not only during a standby mode but also during normal operation, in order to achieve better efficiency. In the case of lighting systems, a particular lighting efficiency (Im / W) is needed to meet certain energy labels (e.g., Class A).

[0008] It is known that a pulsed operation can reduce the power consumption of radar sensors. A simple illustration of a pulsed operation for a radar sensor is shown in Figure 1, which plots the operating voltage of the radar transmitter over time. The radar sensor, especially its radio transmitter, works in an intermittent, periodic, mode. In each period, having a pulse repetition time 10, the radar sensor works only for a fraction of the whole period, i.e., the duration of the pulse width 12.

[0009] The duty cycle of the pulsed operation is defined as the ratio (e.g., as a percentage) of the pulse width 12 to the pulse repetition time 10: Pulse width

[0010] Duty cycle = — — -

[0011] Pulse repetition time

[0012] The actual power consumption of the radar sensor depends on the duty cycle of the pulsed operation:

[0013] Pactual = Pew x Duty cycle

[0014] Pew is the power consumption when the radar sensor works in continuous mode.

[0015] In order to minimize the power consumption, the duty rate needs to be reduced either by decreasing the pulse width or increasing the pulse repetition time. The pulse repetition time determines not only the power consumption, but also the achievable maximal sampling frequency which in turn determines the maximum detectable speed of moving objects by using the Doppler effect.

[0016] The sampling frequency is given by:

[0017] 1

[0018] Sampling frequency = — — -

[0019] Pulse repetition time

[0020] Increasing the pulse repetition time enables a power reduction but causes a reduction of sampling frequency. Eventually the maximum detectable Doppler frequency of a moving object, and thus the maximum detectable speed, will reduce as well.

[0021] For this reason, systems are typically designed to operate at a sampling frequency which is able to detect the maximum object speed that is intended to be detected by the system. This has a power consumption penalty.

[0022] US 2022 / 0295621 discloses a radar-based motion detection system in which a sampling frequency the radar sensor is configured according to the user scenario. The sampling frequency is set in a standby mode based on a maximum frequency of motion to be detected.

[0023] CHEN YIJUN ET AL. discloses a micromotion feature extraction method using tracking pulses with adaptive repetition frequencies to relieve the conflict of the radar resource distribution in their paper titled “Micromotion feature extraction of radar target using tracking pulses with adaptive pulse repetition frequency adjustment”. It aims to using a varying PRF to solve the frequency-domain aliasing problem of the micro-doppler signal.

[0024] KR20190014917A relates to a tracking pulse repetition frequency (PRF) selection method in a medium pulse repetition frequency (MPRF) mode in aircraft radar, and aircraft radar using the same to prevent a target from being positioned in a clutter zone (blind zone) in the aircraft radar.

[0025] SUMMARY OF THE INVENTION

[0026] The invention is defined by the claims.

[0027] According to examples in accordance with an aspect of the invention, there is provided a radar motion sensor comprising: a radar transmitter for transmitting radar pulses with a give pulse repetition time; a radar receiver for receiving the radar pulses after reflection from a monitoring area of the radar sensor; and a controller; wherein the controller is configured to: detect motion using a currently set sampling frequency, the detected motion being associated with a Doppler frequency; determine where the Doppler frequency of the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency; and adapt the sampling frequency based on the Doppler frequency of previously detected motions.

[0028] This motion sensor adapts the sampling frequency used by the radar by analyzing the relationship between the detected Doppler frequency and the sampling frequency. The detected Doppler frequency with which the motion is associated depends on the speed of motion of the detected object. In this way, the sampling frequency adapts to the type of motion being detected. The system can thus adapt to the conditions in which it is used, such as the typical speed of the motions that it detects. Eventually the radar sensor can in this way self-adapt its sampling frequency to an optimal setting that can best fit the application while keeping the sampling frequency as low as possible to minimize power consumption. The "Doppler frequency" is a frequency shift between a signal transmitted by the radar transmitter and the and echo signal received by the radar receiver, caused by the Doppler effect. This frequency shift is associated with the velocity of the objects that generate the echo. Thus, a detection object having particular motion i.e., velocity, is associated with a particular Doppler frequency. Via a time-frequency transformation (such as an FFT), the Doppler frequency of a moving object can be obtained in the frequency domain signal of a Doppler radar system, in known manner.

[0029] The dominant frequency component of the result of the time-frequency processing can be used to determine the Doppler frequency generated by the motion (i.e. the Doppler frequency associated with that motion), if the sampling frequency is sufficiently high without aliasing and folding. Once a motion is detected, it can be associated with the dominant frequency that results from the time-frequency processing.

[0030] The controller is for example configured to label each detected motion to indicate the suitability of the sampling frequency, based on the detected Doppler frequency, and adapt the sampling frequency based on the labels.

[0031] This labeling provides a simple way to analyze past detection performance and modify the sampling frequency if needed.

[0032] The controller is for example configured to label each detected motion with: a first label to indicate a correct sampling frequency; a second label to indicate the sampling frequency is too low; or a third label to indicate the sampling frequency is too high.

[0033] Thus, simple processing with only three possible label values is possible.

[0034] The controller is for example configured to adapt the sampling frequency based on the labels of a set of previously detected motions. Thus, the sampling frequency can adapt over time based on previous historical detections.

[0035] The controller is for example configured to adapt the sampling frequency in a step by step manner.

[0036] The motion sensor for example comprises a time domain processor and associated first, time domain, detector for detecting movements from the received radar pulses.

[0037] The time domain processing enables motion events to be detected without the issue of the sampling frequency needing to be at a suitable level. A simple energy calculation may instead be performed.

[0038] The motion sensor may then further comprise: a frequency domain processor for performing time to frequency conversion and for splitting the frequency domains signal into a set of frequency ranges; and a set of frequency domain detectors for detecting movements, each frequency domain detector being configured to process the frequency domains signal in a respective frequency range.

[0039] Via frequency domain processing, more information about the motion types and moving speed can be obtained compared to the time domain processing, and thus more advanced filtering can be performed to minimize unwanted false detections. Furthermore, via frequency domain processing, clear motion patterns from interested objects but with smaller energy can be detected (e.g. moving objects at a far distance from the sensor)., Thus, the frequency domain processing is more sensitive than the time domain processing.

[0040] The set of frequency detectors for example comprises a second detector for the frequency range from one quarter of the sampling frequency to half of the sampling frequency and a third detector for the frequency range from zero to one quarter of the sampling frequency.

[0041] The controller may be configured to label a detection motion with: a first label to indicate a correct sampling frequency if motion is detected by the second detector then the third detector; a second label to indicate the sampling frequency is too low if motion is detected by the first detector before the second and / or third detectors, or if the motion is detected by the second and third detectors simultaneously; or a third label to indicate the sampling frequency is too high if motion is detected by the third detector but not the second detector.

[0042] In this way, logic is applied to the order in which detections take place in order to determine if the currently set sampling frequency is too low, too high or suitable.

[0043] In this way, the sensor of the invention makes use of a detector and controller arrangement to determine aliasing and folding exist and if so the detected motion is associated with a Doppler frequency that higher than the current detection limits. If there is not aliasing and folding, then the motion is associated with the frequency range currently used for motion detection.

[0044] The invention also provides a lighting system, comprising: a lighting arrangement; a lighting controller; and the motion sensor as defined above for providing a motion detection signal to the lighting controller.

[0045] The invention also provides a method of operating a radar motion sensor, comprising: controlling a radar transmitter of the radar motion sensor to transmit radar pulses with a given pulse repetition time and receiving the radar pulses after reflection from a monitoring area of the radar motion sensor using a radar receiver; detecting motion using a currently set sampling frequency, the detected motion being associated with a Doppler frequency; determining where the Doppler frequency associated with the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency; and adapting the sampling frequency based on the Doppler frequency of previously detected motions.

[0046] The method may comprise labeling each detected motion to indicate the suitability of the sampling frequency, based on the Doppler frequency, and adapting the sampling frequency based on the labels.

[0047] The method may comprise labeling each detected motion with: a first label to indicate a correct sampling frequency; a second label to indicate the sampling frequency is too low; or a third label to indicate the sampling frequency is too high.

[0048] The method may comprise detecting motion using a time domain processor and associated time domain detector and also using a frequency domain processor and a set of frequency domain detectors, each frequency domain detector being configured to process the frequency domain signal in a respective frequency range.

[0049] The invention also provides a computer program comprising computer program code which is adapted, when said program is run on a computer, to implement the method defined above.

[0050] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment s) described hereinafter.

[0051] BRIEF DESCRIPTION OF THE DRAWINGS

[0052] For a better understanding of the invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0053] Fig. 1 shows a pulsed operation of a radar sensor;

[0054] Fig. 2 shows how a motion event is presented under different sampling frequencies;

[0055] Fig. 3 shows the basic units of a motion sensor;

[0056] Fig. 4 shows that the ADC sampling frequency is synchronized with the pulse repetition rate of the radar sensor;

[0057] Fig. 5 shows one possible implementation of a motion detection algorithm implemented;

[0058] Fig. 6 shows an example of a histogram of sampling rate labels; and Fig. 7 shows a method of operating a radar motion sensor.

[0059] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The invention will be described with reference to the Figures.

[0061] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of the apparatus, systems and methods, are intended for purposes of illustration only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, systems and methods of the present invention will become better understood from the following description, appended claims, and accompanying drawings. It should be understood that the Figures are merely schematic and are not drawn to scale. It should also be understood that the same reference numerals are used throughout the Figures to indicate the same or similar parts.

[0062] The invention provides a radar motion sensor in which motion is detected using a currently set sampling frequency, the detected motion being associated with a Doppler frequency. It is determined where the Doppler frequency associated with the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency. The sampling frequency is adapted over time based on the detected Doppler frequency of previously detected motions. In this way, the sampling frequency used by the radar is adapted by analyzing the relationship between the detected Doppler frequency and the sampling frequency. The system can thus adapt to the conditions in which it is used, such as the typical speed of the motions that it detects. Table 1 below gives a simple illustration of how the sampling frequency plays an important role based on a 24GHz pulsed doppler radar sensor with Pcw=60mA and Pulse width=10ps.

[0063] Table 1

[0064] Table 1 shows the current consumption rate and velocity detectability as a function of the sampling frequency.

[0065] A conventional approach uses a fixed sampling frequency which is predefined by the maximal speed of the objects that should be detected. However, a too high sampling frequency will give wasted measurement range and result in a sacrifice of frequency resolution and power consumption. A too low sampling frequency may result in dominant motion signals being missed and can cause a delay or even a false negative detection signal.

[0066] Figure 2 shows an example of how a mismatched sampling frequency setting will impact on the radar sensor signal and power consumption.

[0067] The top image shows a raw time serial sensor signal. The second image shows the frequency domain signal (the FFT result in Hz) after time-frequency conversion with a 2000Hz sampling frequency. The third image shows the frequency domain FFT signal (Hz) with 1000Hz sampling frequency and the bottom image shows the frequency domain FFT signal (Hz) with 200Hz sampling frequency.

[0068] When the sensor operates at 2000Hz sampling frequency, the motion signal can be presented after a Fast-Fourier Transform (FFT) but dominantly within the 0Hz to 500Hz range only (because the detected motion has a speed which does not result in a signal above 500Hz). This means there is waste in measurement range between 500Hz and 1000Hz and a sacrifice of frequency resolution between 0Hz and 500Hz (because the whole range 0 to 1000Hz is measured). There is also excess power consumption.

[0069] If the sensor operates at 200Hz sampling frequency only, a dominant walking movement signal that appears in the regions 20 of the second and third images is no longer present anymore after FFT. Although the power consumption can be significantly reduced, the motion detection performance is degraded significantly as well.

[0070] For the specific motion event represented in Figure 2, 1000Hz would be considered as the most suitable sampling frequency, to reliably detect the motion while maintaining the power consumption at a low level. The frequency domain signals are obtained using a sliding window applied to the raw sensor signal.

[0071] The invention is based on operating a radar sensor in a pulsed operation mode which has a given or fixed pulse repetition time, but with a varied sampling frequency to selfadapt to the optimal sampling frequency setting according to the application. This selfadaptation takes place after the motion sensor is installed.

[0072] Figure 3 shows the basic motion sensor units. The radar sensor unit 30 implements fast on / off pulse control via a pulse width modulation, PWM, control signal PWM signal which is provided by an external main controller 32. The radar sensor unit generates an analog intermediate-frequency (IF) signal which represents the frequency components from the Doppler shifts of moving objects within the sensor field-of-view.

[0073] The controller 32 carries out analog-to-digital conversion (ADC) with a clock signal that is synchronized with the PWM pulse control signal. In this way, the sensor will carry out one sampling operation at every pulse-on timing. The ADC clock signal determines the sampling frequency F sampling, i.e. the frequency of the PWM pulse control signal.

[0074] All of the digital signal, after the analog to digital conversion, is temporally stored and further processed by the controller 32.

[0075] Figure 4 shows that the ADC sampling frequency ("ADC clock") is synchronized with the pulse repetition rate of the radar sensor control signal ("PWM").

[0076] The radar sensor has an initial default sampling frequency setting (F default), and the controller 32 carries out the motion detection based on the ADC sampled signal.

[0077] Figure 5 shows one possible implementation of a motion detection algorithm implemented by the controller 32. It shows various processing elements which process the output signal 50 (the digitally converted IF signal) from the radar sensor unit 30. The timedomain analysis and the FFT processing are applied to the IF signal generated by the ADC sampling. In practice, these software modules are part of an overall detection algorithm. The motion detection algorithm comprises a time domain processor 51 and associated first detector 52 that can process the time serial signal from the digital sensor signal and detect moving objects accordingly.

[0078] The time-domain signal processor carries out various processing function to the input signal, such as DC offset removal, AC amplitude calculation, zero-crossing detection, etc. All the processed results then feed into the first detector 52. The first detector 52 is designed to be capable for motion detection based on a time-domain signal only.

[0079] The motion detection algorithm also comprises a frequency domain processor which first performs time to frequency conversion using FFT 54, and then splits the converted signal into at least two frequency ranges. This processing takes place in parallel.

[0080] The converted signal is in this way divided into frequency bins. The bins cover frequencies in the range from zero to half the sampling frequency.

[0081] An example is shown for two frequency bins, but there may be more. In this example, a lower frequency range is defined as a lower range [0,1 / 4 F default] and a second, higher, range is defined as [1 / 4 F default, 1 / 2 F default], However, the split into bins is not necessarily an equal split.

[0082] There is a signal processor and detector for each frequency range, hence in this example a second signal processor 56 and a second detector 58 for the high frequency range and a third signal processor 60 and third detector 62 for the low frequency range.. The second and third detectors are designed to be capable of independent motion detection based on the frequency-domain signal only, but focusing on different frequency ranges. Thus, the FFT output signal is processed within each split frequency domain and the processing detects moving objects using those frequency ranges.

[0083] The radar sensor applies certain criteria according to the outputs from the three detectors 52, 58, 62 in order to judge whether the current sampling frequency setting is suitable or not (e.g. too low or too high) for the latest detected motion event. The detection can then be classified by applying certain predefined labels (e.g.. ‘GOOD’, ‘LOW’ or ‘HIGH’).

[0084] Each motion event has an associated Doppler frequency, relating to the speed of motion that has been detected. The Doppler frequency needs to be below half of the sampling frequency in order for the motion to be detectable. Thus, there is a suitable range of possible sampling frequencies for each motion event, depending on the speed of motion. The allocation of a detection label involves determining where the Doppler frequency of the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency. For example, the Doppler frequency of a very slowly moving object will be close to zero, whereas the Doppler frequency of an object moving close to the maximum detectable speed will be close to half the sampling frequency. If the Doppler frequency is low, the sampling frequency can be reduced to save power. If the Doppler frequency is high, then there is a risk that some motions are not being detected. Thus, the sampling frequency may be too low or too high, depending on the speed of the objects that are being (or should be) detected. Thus, the invention is based on adapting the sampling frequency based on the Doppler frequency (relative to the sampling frequency) of previously detected motions.

[0085] The radar sensor then uses the label results from a number of historical motion events (e.g.. 20 past motion detection results) to determine whether the current sampling frequency needs to be adjusted, by using a predefined strategy. Once the need for adjustment has been established, the radar sensor will automatically increase or decrease its sampling frequency step by step. As explained above, this makes adjustment to both the radar control signal and the ADC sampling frequency.

[0086] In this way, the radar sensor will self-adapt its sampling frequency to an optimal setting that can best fit the application while keeping the frequency as low as possible in order to minimize its power consumption.

[0087] The table below shows an example of the possible strategy to be applied.

[0088] Table 2

[0089] In this table, detector 1 is the first, time domain, detector, detector 2 is the second, high frequency, detector and detector 3 is the third, low frequency, detector. Row 1 of the table covers a situation where there is motion detection only from detector 1, or when there is detection by detector 1 before detectors 2 and / or 3. If detector 1 makes a detection earlier or even alone, it means the triggering is most likely due to the folding and aliasing from frequency components that are higher than the current Nyquist frequency (’A F_sampling). This means the current sampling frequency is too low to capture the details about a high-speed object.

[0090] Row 2 of the table covers a situation where there is motion detection only from detectors 2 and 3, or when there is detection by detectors 2 and 3 before detector 1. As mentioned above, the folding and aliasing of high-frequency components to the low- frequency range is due to the sampling frequency being too low. This folding and aliasing can also result in detection by detectors 2 and 3, simultaneously. Thus, detection by detectors 2 and 3 almost simultaneously is indicative of the presence of folding and aliasing, and thus too low sampling frequency.

[0091] Row 3 of the table covers a situation where there is motion detection from detector 2 then detector 3, or from detector 2 then detector 3 then detector 1. For motion detection of objects with almost consistent speed and with a suitable sampling frequency, the signal pattern will always appears in the high-frequency range first, and then the low- frequency range. The relative distance will only impact on the signal energy but not the pattern itself.

[0092] If the objects only move at a far distance with smaller energy, the time domain signal processing may not be able to be triggered based on a certain energy threshold. If the motion happens close enough to reach the signal energy threshold, then detector 1 may also be triggered. Thus, the detection by detector 2 then detector 3 is indicative of a motion pattern indicating a suitable sampling frequency. Whether or not there is detection by detector 1 depends on the signal energy rather than the motion pattern.

[0093] Row 4 of the table covers a situation where there is motion detection from detector 3 alone or from detector 3 then detector 1. This means the object stays at a low moving speed and less that 1 / 4 F sampling of Doppler shift can be generated. This is an indicator that the current sampling frequency setting is too high. If the sampling frequency too high, a motion signal will also not be present in the high-frequency range (due to low Doppler frequency shift) so that detector 2 will be not able to be triggered.

[0094] In this way, every motion detection event is allocated a label, e.g., Low, Good or High. The radar sensor uses the labels from a number of motion events in the past (e.g.. 20 of the past labels) to build a histogram of label distribution. Figure 6 shows one example of such histogram.

[0095] This distribution is fed into a sampling frequency adjustment decider of the controller which will determine whether the current sampling frequency setting needs to be adjusted by using a predefined strategy or a combination of strategies, which can be freely chosen depending on the target application where the radar sensor to be installed.

[0096] There are various options for optimizing the strategy:

[0097] A conservative strategy will have a low tolerance on missing the detection of high-speed objects such as road vehicles. As soon as there is one Low motion label among the preceding 20 motion labels, the decider will conclude that the sampling frequency is too low.

[0098] An aggressive strategy will minimize power consumption as the top priority and hence allow for some missed detections of high-speed moving objects (e.g. a running pet in the home). Once there is one High motion label among the preceding 20 motion labels, the decider conclude that the sampling frequency is too high.

[0099] A balanced strategy aims to provide a balance between power consumption and detection performance, such as in the garden. Only if more than 50% of the preceding motion labels are High or Low will the decider conclude that the sampling frequency is too high or too low, respectively.

[0100] The decider will eventually decide whether the current sampling frequency needs to be adjusted or not, according to the conclusions it has made. Once it concludes the sampling frequency is too low, it will increase the current sampling frequency to a higher setting. Equivalently, it will decrease the current sampling frequency to a lower setting when it concludes it is too high.

[0101] The radar sensor for example starts to work under an initial sampling frequency which is predetermined by the target applications and the adaptive preference. Eventually, the radar sensor will self-adapt its sampling frequency to an optimal setting that can best fit the application while keeping the frequency as low as possible to minimize power consumption.

[0102] Figure 7 shows a method of operating a radar motion sensor, comprising:

[0103] In step 70, controlling the radar transmitter to transmit radar pulses with a given pulse repetition time and receiving the radar pulses after reflection;

[0104] In step 72, detecting motion using a currently set sampling frequency, the detected motion being associated with a Doppler frequency; In step 74 determining where the Doppler frequency of the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency; and

[0105] In step 76, adapting the sampling frequency based on the Doppler frequency of previously detected motions.

[0106] The example above is based time domain detection and multiple frequency domain detections. The use of time domain analysis enables multiple scenarios to be detected as explained above with reference to Table. 2. However, the invention may be applied with only frequency domain detection, with the sampling frequency adapted to maintain the detected motions at the most suitable frequency relative to the sampling frequency.

[0107] Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0108] Functions implemented by a processor may be implemented by a single processor or by multiple separate processing units which may together be considered to constitute a "processor". Such processing units may in some cases be remote from each other and communicate with each other in a wired or wireless manner.

[0109] The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0110] A computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

[0111] If the term "adapted to" is used in the claims or description, it is noted the term "adapted to" is intended to be equivalent to the term "configured to". If the term "arrangement" is used in the claims or description, it is noted the term "arrangement" is intended to be equivalent to the term "system", and vice versa.

[0112] Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A radar motion sensor comprising: a radar transmitter for transmitting radar pulses with a given pulse repetition time; a radar receiver for receiving the radar pulses after reflection from a monitoring area of the radar sensor; and a controller (32); wherein the controller (32) is configured to: detect motion using a currently set sampling frequency, the detected motion being associated with a Doppler frequency; determine where the Doppler frequency associated with the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency; and adapt the sampling frequency based on the Doppler frequency of previously detected motions.

2. The motion sensor of claim 1, wherein the controller (32) is configured to label each detected motion to indicate the suitability of the sampling frequency, based on the Doppler frequency, and adapt the sampling frequency based on the labels, wherein the controller (32) is configured to label each detected motion with: a first label to indicate a correct sampling frequency; a second label to indicate the sampling frequency is too low; or a third label to indicate the sampling frequency is too high.

3. The motion sensor of claim 2, wherein the controller (32) is configured to adapt the sampling frequency based on the labels of a set of previously detected motions.

4. The motion sensor of claim 3, wherein the controller (32) is configured to adapt the sampling frequency in a step by step manner.

5. The motion sensor of any one of claims 1 to 4, comprising a time domain processor (51) and associated first, time domain detector (52) for detecting movements from the received radar pulses.

6. The motion sensor of claim 5, further comprising: a frequency domain processor (54) for performing time to frequency conversion and for splitting the frequency domains signal into a set of frequency ranges; and a set of frequency domain detectors (58, 62) for detecting movements, each frequency domain detector being configured to process the frequency domain signal in respective frequency range.

7. The motion sensor of claim 6, wherein the set of frequency detectors comprises a second detector (58) for the frequency range from one quarter of the sampling frequency to half of the sampling frequency and a third detector (62) for the frequency range from zero to one quarter of the sampling frequency.

8. The motion sensor of claim 7, wherein the controller is configured to label a detection motion with: a first label to indicate a correct sampling frequency if motion is detected by the second detector then the third detector; a second label to indicate the sampling frequency is too low if motion is detected by the first detector before the second and / or third detectors, or if the motion is detected by the second and third detectors simultaneously; or a third label to indicate the sampling frequency is too high if motion is detected by the third detector but not the second detector.

9. A lighting system, comprising: a lighting arrangement; a lighting controller; and the motion sensor of any one of claims 1 to 8 for providing a motion detection signal to the lighting controller.

10. A method of operating a radar motion sensor, comprising:(70) controlling a radar transmitter of the radar motion sensor to transmit radarpulses with a given pulse repetition time and receiving the radar pulses after reflection from a monitoring area of the radar motion sensor using a radar receiver;(72) detecting motion using a currently set sampling frequency, the detected motion being associated with a Doppler frequency;(74) determining where the Doppler frequency associated with the detected motion lies within the possible range of detectable frequencies corresponding to the currently set sampling frequency; and(76) adapting the sampling frequency based on the Doppler frequency of previously detected motions.

11. The method of claim 10, comprising labeling each detected motion to indicate the suitability of the sampling frequency, based on the Doppler frequency, and adapting the sampling frequency based on the labels; wherein labeling each detected motion with: a first label to indicate a correct sampling frequency; a second label to indicate the sampling frequency is too low; or a third label to indicate the sampling frequency is too high.

12. The method of any one of claims 10 to 11, comprising detecting motion using a time domain processor and associated time domain detector and also using a frequency domain processor and a set of frequency domain detectors, each frequency domain detector being configured to process the frequency domain signal in a respective frequency range.

13. A computer program comprising computer program code which is adapted, when said program is run on a computer, to implement the method of any one of claims 10 to

Images