DE102017110295A1 - Apparatus and method for detecting and categorizing vital data of a subject with the aid of radar radiation - Google Patents

Abstract

The present invention relates to a device (1) for acquiring and classifying vital data of a subject (2) with a radar device (3) for detecting at least one distance of the radar device (3) from at least one reflecting or scattering interface (8) of the subject (2 ), wherein the radar device (3) is set up and arranged such that in the operation of the device (1) electromagnetic radiation emitted by the radar device (3) is directed at least to a portion of the subject (2), so that from the interface ( 8) reflected or scattered electromagnetic radiation contains a time-dependent distance information about the distance of the interface (8) from the radar device (3), wherein the time-dependent distance information contains a measure of at least one vital parameter of the subject, and wherein the radar device (3) is set up in that in the operation of the device it is a time-dependent measurement The evaluation device (4) is set up in such a way that it evaluates the measurement signal in the operation of the device (1) and, based on the time-dependent measurement signal, evaluates a state of the subject in at least one dimension selected from a group of dimensions comprising cognitive load, fatigue and stress levels classified in multiple levels.

Description

The present invention relates to a device and a method for detecting and categorizing vital data of a subject with the aid of radar radiation.

For example, from the DE 21 2015 000 102 U1 It is known to use microwave radiation in a frequency range between 40 GHz and 300 GHz for non-contact detection of the heartbeat frequency or the respiratory frequency of a subject.

The measurement method is basically suitable for recording vital data of a subject. However, it has been found that its application in everyday practice was not feasible due to the relatively high cost of conventional radar systems and their large size. In addition, it has been shown that, outside of laboratory conditions with a fixed position of the subject and largely trouble-free environmental conditions, an evaluation of the measurement signal output by the radar device leads to unreliable results.

It is therefore an object of the present invention to provide a device and a method for acquiring and classifying vital data, which overcomes at least one of the aforementioned disadvantages.

At least one of the aforementioned objects is achieved by a device for acquiring and classifying vital data of a subject with a radar device for detecting at least one distance of the radar device from at least one reflective or scattering interface of the subject, wherein the radar device is set up and arranged such that in electromagnetic radiation radiated to the operation of the device by the radar device is directed at least to a portion of the subject such that electromagnetic radiation reflected or scattered by the interface includes time-dependent distance information about the distance of the interface from the radar device, the time-dependent distance information being a measure of at least one of the distance information Contains vital parameters of the subject, and wherein the radar device is arranged such that in the operation of the device, a time-dependent measuring signal, the distance information contains the evaluation signal, and the evaluating device is set up in such a way that it evaluates the measuring signal in the operation of the device and selects a condition of the subject in at least one dimension from a group of dimensions including cognitive starting from the measuring signal Stress, fatigue and stress levels classified in several stages.

The device according to the invention is based on the finding that one or more vital parameters of the subject can be derived from the distance information, which describes the distance between the radar device and at least one interface of the subject in a time-dependent manner. In particular, the radar device makes it possible to detect the heartbeat frequency (also known as heart rate) and / or the heart rate variability and / or the respiratory rate of the subject. Heart rate variability refers to the quantification of the fluctuations in the time intervals of two successive heartbeats. In the present application, the term subject is used in summary for a human or an animal, in particular a mammal.

The thorax movement of a subject, that is the raising and lowering of the breast, is proportional to the subject's breathing activity. The heartbeat, however, is directly related to the pulse, that is, a more or less periodic pulsation or dilation of the blood vessels. Both the respiratory activity and the heartbeat thus lead to changes in the position of surfaces or interfaces of the subject relative to a reference point, which is inventively formed by the radar device.

In the case of a distance measurement with the radar device, therefore, the vital parameters can be detected as time-dependent changes in the distance between the radar device and the reflective or scattering interface of the subject.

The boundary surface to be considered, which reflects or scatters the electromagnetic radar radiation, is in particular the skin surface of the subject. However, an interface that reflects or scatters the radar radiation is also formed, in particular, by a skin-near surface of a blood vessel, that is to say a vein or other boundary surfaces. Generally speaking, an interface is characterized by a refractive index change, which in turn causes scattering or reflection.

The evaluation device is set up such that it evaluates the measurement signal output by the radar device and classifies a state of the subject in at least one dimension into at least two classes. An example of a multilevel classification of the subject's condition in one dimension is the classification of the subject's fatigue in awake, tired and exhausted.

The aim of this evaluation is to reliably evaluate or predict the vital state of the subject with the aid of the measurement signal in at least one dimension.

In the simplest case, the condition of the subject for a single dimension, such as fatigue, is classified into two classes, such as awake and tired. Being able to perform this classification of the subject's state as reliably as possible is a central aspect of the present invention.

In one embodiment, the evaluation device is set up in such a way that it evaluates the measurement signal in the operation of the device and determines from the distance information contained in the measurement signal a measure for at least one vital parameter of the test person and classifies a state of the test person in multiple stages on the basis of the measure for the vital parameter ,

In one embodiment, the dimension is selected from a group of dimensions consisting of cognitive stress, fatigue and stress levels.

For example, in one embodiment, the heartbeat frequency and / or the respiratory rate are determined by suitable filtering of the measurement signal and transformation of the measurement signal, in particular by applying a Fourier transformation to the time-dependent measurement signal. In order to classify the state of the subject from one of these vital parameters derived from the measurement signal, the derived vital parameter, for example, can be compared to a threshold value. For example, if the derived measure of the vital parameter is below the threshold, then the condition of the subject is the first class of that dimension. On the other hand, if the measure for the vital parameter is above the threshold value, then the condition of the subject in the second class is for the same dimension.

In one embodiment of the invention, the evaluation device is set up in such a way that it determines from the measurement signal a measure of the heartbeat rate and a measure of the respiration rate of the subject in the operation of the device and based both on the measure of the heartbeat frequency and on the measure of the respiratory rate classifies the condition of the subject in several stages.

In one embodiment, however, the evaluation device comprises a self-learning system which, on the basis of patterns and laws learned during a learning phase, from the time-dependent measurement signal classifies the state of the subject in at least one dimension.

For this purpose, the self-learning system receives in the learning phase as input at least one time-dependent measurement signal containing a distance information about the distance of a test person's interface from the radar device and a time-dependent, independent of the measurement signal state description of the subject in the dimension to be classified, a Time dependence of the measurement signal and a time dependence of the state description are synchronized.

In this case, during the learning phase in one embodiment, the subject may be the same subject to whom the device is applied in measuring mode, that is to say in the measuring phase outside the learning phase. However, embodiments are also conceivable in which a subject is used in the learning phase, which serves as a pattern as a pattern and is not identical to the subject to whom the device is applied in the measurement phase.

The time-dependent subjective state description can either be created by the subject himself, it is then a subjective description of the condition, by a third party as an observer of the subject or by an electronic detection of vital parameters with conventional, invasive or contacting procedures with a correspondingly proven reliable evaluation of the Measurement signal and a categorization of the condition of the subject in the dimension to be categorized.

It is important that the state description read into the evaluation device by the measurement signal in the learning phase also has a time dependency that is synchronized with the time dependency of the measurement signal during the learning phase. Thus, events in the measurement signal can be clearly assigned to a subjective state description at the same time.

The core of the self-learning system is that it identifies and stores patterns and patterns from the time-dependent measurement signal and the time-dependent state description during the learning phase, these identified and stored patterns and laws then being used in the operation of the device, ie in measurement mode, to classify the Serve the subject's condition.

To identify the patterns and regularities during the learning phase and to classify the state of the subject are known from the prior art algorithms for self-learning systems, in particular Linear and Logistic Regression, Decision Tree, Support Vector Machine, Naive Bayes, artificial neural Nets, K-Means, Random Forest, Dimensionality Reduction Algorithms, Gradiant Boost and Adaboos and Neural Networks.

Particularly suitable are artificial neural networks and random forest algorithms.

In one embodiment of the invention, the radar device comprises a monolithically integrated microwave transceiver integrated circuit (MMIC), preferably based on SiGe semiconductor technology, a microwave module on which the microwave transceiver circuit is mounted, and a control module , In this case, the control module comprises a microprocessor for controlling the microwave elements both on the microwave module and in the microwave transceiver circuit, an analog-to-digital converter arranged to be digitized in the operation of the radar means an intermediate frequency signal from a receiver mixer and a USB interface, wherein the USB interface is connected to the analog-to-digital converter, wherein the USB interface is configured such that it provides the intermediate frequency signal as a digital measurement signal to the evaluation in the operation of the radar device and wherein the USB interface is configured such that it provides a power supply for the radar device in the operation of the radar adjustment.

It has been found that the use of a radar device in a device for detecting and categorizing a subject's vital data has three essential requirements for the radar device: Firstly, the size of the radar device must be so small that the radar device does not have any appreciable space in the surroundings of the subject, for example in a motor vehicle claimed. On the other hand, the radar device must have a unit price, which makes the claimed use useful from an economic point of view. In addition, the radar device must be integrated via a standardized interface with the evaluation device.

All three requirements are fulfilled by the radar device described above. The provision of transmitter and receiver on a monolithically integrated microwave transceiver circuit allows a design with the smallest possible space requirement. In addition, this design is suitable for low-cost production in higher quantities.

Finally, the conversion of the intermediate frequency signal and thus of the measurement signal to be evaluated in an analog-to-digital converter, before processing, that is an evaluation of the intermediate frequency signal, the required simple connection to the evaluation available. The USB interface allows easy transfer of the digitized intermediate frequency signal and thus its evaluation in a computer as an evaluation device outside the actual radar device.

In one embodiment of the invention, the radar device is a Doppler radar, which detects the changes in the distance between the reflective or diffusive interface and the radar device as a frequency shift in a movement of the test person's interface relative to the radar device.

In particular, however, in one embodiment, the radar device is a frequency-modulated continuous-wave radar. This is called in English Frequency Modulated Continuous Wave Radar, short FMCW radar. This radar principle enables a radar operation with a determination of the direction and the distance between an object, here the scattering or reflecting surface of the subject, and a transmitter or receiver antenna of the radar device with manageable expenditure on equipment.

The aim of the FMCW radar is to determine the transit time of a radar signal radiated from a transmitter antenna and received by a receiver antenna, and thus the distance between the interface and one of the antennas. The basis for the distance measurement is that the frequency of a monofrequency, comparatively narrow-band radar signal is varied over time. For example, over a time interval, the frequency of the radiated signal increases continuously and linearly with respect to time. If one uses now a part of the generated radar signal as a reference signal and directs this reference signal to the receiver directly, while the actual radar signal from the transmitting antenna on the object, here the interface of the subject, runs back to the receiver antenna and mixed at the receiver received by the receiver antenna Radar signal with the reference signal, the mixing process generates an intermediate frequency signal as a measurement signal in the context of the present application. The frequency of the intermediate frequency signal results from the different transit times of reference signal and radar signal. It is important that the duration of the radar signal is not greater than the predetermined time interval over which the frequency of the radiated radar signal is changed. If you determine the generated intermediate frequency at the receiver, ie behind the mixer, this is proportional to the distance between the transmitter antenna or the receiver antenna and the radar signal reflecting or scattering interface.

In other words, the timing of radiation of the radar signal is frequency-encoded within an interval in which the frequency of the radiated radar signal is varied. In one embodiment of the invention, but in particular in one embodiment of the invention in which the radar device is a frequency-modulated continuous wave radar, the radar device is designed such that the radar device in the operation of the device electromagnetic radiation with a center frequency in a frequency range of 10 GHz to 300 GHz, preferably in a range of 20 GHz to 170 GHz and radiates.

The mentioned frequency ranges are to be covered in the meantime with commercially available radar systems of small size. Such systems can also be integrated, for example, in vehicles and machines, since they do not require excessive space.

In an embodiment of the invention in which the radar device is a frequency modulated continuous wave radar, the radar device is configured to provide continuous wave electromagnetic radiation in the operation of the device with a tuning bandwidth of 40 GHz or less, preferably with a tuning bandwidth of 30 GHz or less , generates and receives.

The tuning bandwidth basically determines the resolution with which the radar device can detect changes in distance. While it is generally desirable to have the largest possible resolution, the tuning bandwidth is always accompanied by a considerable expenditure on equipment on the side of the radar device. For economic reasons, therefore, it is necessary to select the radar device in such a way that it fits the object to be measured or the task set. It has been found that for the acquisition of vital data of a subject, a tuning bandwidth in a range of 20 GHz to 40 GHz, preferably 20 GHz to 30 GHz is sufficient and at the same time requires an expenditure on equipment which is also acceptable for mass production.

In one embodiment of the invention, the radar device has a dynamics, viewed over the whole system (i.e., a system bandwidth) of 60 dB or more, preferably 80 dB or more.

It is possible in embodiments of the invention to combine various of the measuring methods described above and in this way to obtain the distance information.

In particular, in one embodiment of the invention, the frequency modulated continuous-wave radar is a Doppler radar. Such a combination of a Doppler radar and a frequency-modulated continuous-wave radar is also known in the literature as micro Doppler radar.

In a further embodiment of the invention, the radar device is designed bistatically, that is, both for the transmitter and for the receiver each has its own antenna ready. This requires for embodiments of the device according to the invention the least equipment and thus financial expense.

In one embodiment of the invention, the radar device is a multi-frequency radar device which is adapted to emit and receive electromagnetic radiation having at least two center frequencies in at least two distinct frequency ranges, wherein preferably the at least two frequency ranges are selected from a frequency range of twenty GHz to 28 GHz, a frequency range of 60 GHz to 90 GHz and a frequency range of 120 GHz to 170 GHz.

It has been shown that with distance measurements with center frequencies in at least two mutually different frequency ranges additional information can be obtained, which facilitate the classification of the condition of the subject.

It has proved to be suitable if the radar device is set up such that it radiates and receives electromagnetic radiation having a center frequency in a frequency range from 20 GHz to 28 GHz and electromagnetic radiation in a frequency range from 120 GHz to 170 GHz in the operation of the device wherein a measure of the heartbeat rate as a vital parameter from a measurement signal is determined, the distance information is obtained with the electromagnetic radiation in the frequency range of 20 GHz to 28 GHz, and wherein a measure of the respiratory rate as a vital parameter from a measurement signal is determined whose distance information is obtained with the electromagnetic radiation in the frequency range of 120 GHz to 170 GHz.

The device for recording and categorizing vital data of a subject is particularly suitable for categorizing vital data of the driver of a land vehicle, an aircraft or a maritime vehicle. Applications in vehicles have the advantage that there the test person, that is the driver, typically has an approximately defined position on a seat or behind a steering wheel. In particular, in a land vehicle in which the subject is strapped as a driver of the vehicle, it is possible to perform a reference measurement on the seatbelt, which has a defined position over the skin of the subject.

In addition, the inventive device for detecting and categorizing vital data of a subject but also for installation in a stretcher, a hospital bed, a control station or a machine is suitable to capture the vital data, for example, a patient, a machine operator or an air traffic controller. In all of these applications, it is in one embodiment, to prevent operator error due to tiredness or even diseases of the subject.

• Bestrahlen zumindest eines Abschnitts des Probanden mit elektromagnetischer Strahlung aus einer Radareinrichtung,
• Empfangen der an einer Grenzfläche des Probanden reflektierten oder gestreuten elektromagnetische Strahlung mit der Radareinrichtung,
• Erfassen einer zeitabhängigen Abstandsinformation über den Abstand der Grenzfläche von der Radareinrichtung, wobei die zeitabhängige Abstandsinformation ein Maß für mindestens einen Vitalparameter des Probanden enthält, und
• Ausgeben eines zeitabhängigen Messsignals, das die Abstandsinformation enthält, aus der Radareinrichtung,
• Einlesen des Messsignals in eine Auswerteeinrichtung,
• Auswerten des Messsignals in der Auswerteeinrichtung und
• mehrstufiges Klassifizieren eines Zustands des Probanden in mindestens einer Dimension ausgewählt aus einer Gruppe von Dimensionen umfassend kognitive Belastung, Ermüdung und Stresslevel ausgehend von dem zeitabhängigen Messsignal.

At least one of the aforementioned objects is also solved by a method for acquiring and classifying vital data of a subject with the steps:

• Irradiating at least a portion of the subject with electromagnetic radiation from a radar device,
• Receiving the reflected or scattered at an interface of the subject electromagnetic radiation with the radar device,
• Detecting a time-dependent distance information about the distance of the interface from the radar device, wherein the time-dependent distance information contains a measure of at least one vital parameter of the subject, and
• Outputting a time-dependent measurement signal containing the distance information from the radar device,
• Reading the measuring signal into an evaluation device,
• Evaluation of the measurement signal in the evaluation and
• Multi-level classifying a subject's condition in at least one dimension selected from a group of dimensions including cognitive stress, fatigue and stress levels based on the time-dependent measurement signal.

As far as aspects of the invention have been described with respect to the device, they also apply to the corresponding method for detecting and categorizing a subject's vital data and vice versa. As far as the method is carried out with a device according to this invention, so this has the appropriate facilities for this. In particular, embodiments of the apparatus are suitable for carrying out embodiments of the method.

Further advantages, features and applications of the present invention will become apparent from the following description of an embodiment and the associated figures.

1 shows a schematic representation of the arrangement of a device according to the invention for detecting and categorizing vital data with a subject.

In the figures, like elements are designated by like reference numerals.

1 shows a schematic representation of a device according to the invention 1 to capture and categorize vital data of a subject 2 , The subject 2 In the illustrated embodiment, an air traffic controller sitting on a chair and the device 1 is integrated in the pilot's tableau.

The device 1 includes a radar device 3 and an evaluation device 4 , The radar device is a two-color Doppler radar whose radar radiation 5 on the chest of the subject 2 is directed. The radar radiation 5 is used by the subject's skin as an interface 8th reflected in the context of the present application. The radar emits electromagnetic radar radiation 5 with a center frequency of 25 GHz to determine a measure of heart rate as a vital parameter from the measurement signal and radar radiation 5 with a center frequency of 150 GHz to determine a measure of the respiratory rate as a vital parameter from the measurement signal. The measuring signal is from the radar device 3 a digital signal via a USB interface to the evaluation 4 to hand over.

In the in 1 the embodiment shown is the evaluation device 4 via an interface with the internet 6 and about this with a control center 7 connected. It is important in the control center 7 to capture when the subject 2 , a condition with reduced attention must be achieved and replaced with a rested subject.

On the basis of the radar device 3 provided measurement signal, which to the evaluation 4 is transmitted to make a classification of the vital state of the subject in one dimension, includes the evaluation 4 a self-learning system 9 ,

This self-learning system 9 is prior to the actual measurement phase on the specific subject 2 learned. This is indicated by the self-learning system 9 an algorithm with an artificial neural network. In the learning phase the subject becomes 2 exposed to the same influences as during the measuring phase. That is, the subject 2 carries out his work as an air traffic controller, but subjectively records his condition, in this case exemplary with regard to fatigue, with exact time information. At the same time during the learning phase, the time-dependent measurement signal from the radar device 3 captured and both the subjective, from the subject 2 created state description as well as the time-dependent measurement signal from the radar device 3 be in the self-learning system 9 the evaluation device 4 evaluated. An artificial neural network is used to identify and store patterns and laws.

These patterns and laws identified during the learning phase serve in the measurement phase to classify the state of the subject with regard to his tiredness as a dimension within the meaning of the present application from the information of the time-dependent measurement signal. With the help of the self-taught patterns and laws, the condition of the test person with regard to his attention is then derived or predicted from the measurement signal. In this case, states with increased fatigue can even be predicted from the measurement signal before the fatigue is categorized into a category identified as critical. When this condition occurs, will be on the control center 7 A warning signal is issued and the air traffic controller is replaced by a rested, alert colleague.

For purposes of the original disclosure, it is to be understood that all such features as will become apparent to those skilled in the art from the present description, drawings, and claims, even though they have been specifically described in connection with certain further features, may be individually hereto in any manner Compilations with other features or feature groups disclosed herein can be combined, unless this has been expressly excluded or technical conditions make such combinations impossible or pointless. On the comprehensive, explicit representation of all conceivable combinations of features is omitted here only for the sake of brevity and readability of the description.

While the invention has been illustrated and described in detail in the drawings and the foregoing description, such illustration and description is exemplary only and is not intended to limit the scope of the protection as defined by the claims. The invention is not limited to the disclosed embodiments.

Variations of the disclosed embodiments will be apparent to those skilled in the art from the drawings, the description 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. The mere fact that certain features are claimed in different claims does not exclude their combination. The reference signs in the claims are not intended to be limiting of the scope.

LIST OF REFERENCE NUMBERS

1

contraption

2

Family

3

radar device

4

evaluation

5

radar radiation

6

Internet

7

control center

8th

Interface of the subject

9

self-learning system

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 212015000102 U1 [0002]

Claims (15)

Device (1) for acquiring and classifying vital data of a subject (2) with a radar device (3) for detecting at least one distance of the radar device (3) from at least one reflective or scattering interface (8) of the subject (2), wherein the radar device (3) is set up and arranged such that in the operation of the device (1) electromagnetic radiation radiated by the radar device (3) is directed at least to a portion of the subject (2), so that from the interface (8) reflected or scattered electromagnetic radiation contains a time-dependent distance information about the distance of the interface (8) from the radar device (3), wherein the time-dependent distance information contains a measure of at least one vital parameter of the subject, and wherein the radar device (3) is arranged such that in the operation of the device, it outputs a time-dependent measurement signal containing the distance information, and an evaluation device (4) reading in the measurement signal, wherein the evaluation device (4) is set up in such a way that it evaluates the measurement signal in the operation of the device (1) and selects from the time-dependent measurement signal a state of the subject in at least one dimension selected from a group of dimensions including cognitive load, fatigue and stress level classified in several stages. Device after Claim 1 , characterized in that the evaluation device (4) is set up in such a way that it evaluates the measurement signal in the operation of the device (1) and determines a measure for at least one vital parameter of the test person from the distance information contained in the measurement signal, and based on the measurement for the vital parameter, a condition of the subject classified in several stages. Device (1) according to one of the preceding claims, characterized in that the vital parameter is the heartbeat rate or the heart rate variability or the respiratory rate. Device (1) according to one of the preceding claims, characterized in that the evaluation device (12) is set up in such a way that it measures a measure of the heartbeat rate and a measure of the respiratory rate of the subject (2) from the measurement signal in the operation of the device (1). determined and classified on the basis of the measure of the heart rate and the measure of the respiratory rate, the state of the subject in several stages. Device (1) according to one of the preceding claims, characterized in that the radar device (3) comprises a monolithically integrated microwave transceiver circuit (MMIC), a microwave module on which the microwave transceiver circuit is mounted, and a control module with a microprocessor, a An analog-to-digital converter arranged and arranged to digitize an intermediate frequency signal from a receiver mixer in the operation of the radar device (3), and a USB interface, the USB interface to the analog to digital converter is connected, wherein the USB interface is configured such that it provides in the operation of the radar device (3) the intermediate frequency signal as a measurement signal for the evaluation, and wherein the USB interface is configured such that in the operation of the radar device (3 ) provides a power supply to the radar device (3). Device (1) according to one of the preceding claims, characterized in that the radar device (3) has a system dynamics of 60 dB or more, preferably 80 dB or more. Device (1) according to one of the preceding claims, characterized in that the radar device (3) is a Doppler radar, wherein a Doppler shift forms the distance information. Device (1) according to one of the preceding claims, characterized in that the radar device (3) is a frequency-modulated continuous wave radar. Device (1) according to one of the preceding claims, characterized in that the radar device (3) is designed such that the radar device (3) in the operation of the device (1) electromagnetic radiation having a center frequency in a frequency range of 10 GHz to 300 GHz, preferably radiates and receives in a range of 20 GHz to 170 GHz. Device (1) according to one of the preceding claims, characterized in that the radar device is a multi-frequency radar device which is adapted to emit and receive electromagnetic radiation having at least two center frequencies in at least two mutually different frequency ranges, preferably the at least two Frequency ranges are selected from a frequency range of 20 GHz to 28 GHz, one Frequency range from 60 GHz to 90 GHz and a frequency range from 120 GHz to 170 GHz. Device (1) according to any one of the preceding claims, characterized in that the radar device is arranged such that it in the operation of the device electromagnetic radiation having a center frequency in a frequency range of 20 GHz to 28 GHz and electromagnetic radiation in a frequency range of 120 GHz radiates and receives up to 170 GHz, wherein a measure of the heartbeat rate as a vital parameter from a measurement signal is determined, the distance information is obtained with the electromagnetic radiation in the frequency range of 20 GHz to 28 GHz, and wherein a measure of the respiratory rate as a vital parameter of a Measurement signal is determined, the distance information is obtained with the electromagnetic radiation in the frequency range of 120 GHz to 170 GHz. Device (1) according to one of the preceding claims, characterized in that the evaluation device (4) comprises a self-learning system (5), wherein the self-learning system (5) in a learning phase as input at least one time-dependent measurement signal, which is a distance information about the distance an interface (8) of a subject of the radar device (3), and a time-dependent, independent of the measurement signal state description of the subject in the dimension to be classified, wherein a time dependence of the measurement signal and a time dependence of the state description are synchronized, the self-learning system such is set up that it identifies and stores patterns and regularities from the time-dependent measurement signal and the time-dependent state description, and wherein the self-learning system is set up so that it can be used in the operation of the device from the time-dependent measurement signal with the aid of the Learning phase identified and stored patterns and laws classified the condition of the subject (2). Device (1) according to Claim 14 , characterized in that the patterns and laws are identified with an algorithm and the state of the subject is classified with the algorithm, the algorithm being selected from a group of algorithms consisting of Linear and Logistic Regression, Decision Tree, Support Vector Machine, Naive Bayes, an Artificial Neural Network, K-Means, Random Forest, Dinemsionality Reduction Algorithms, Gradient Boost and Adaboost, and a Neural Network. Land vehicle, aircraft, maritime vehicle, stretcher, hospital bed, control station or machine with a device (1) according to one of the preceding claims. Method for recording and classifying vital data of a subject (2) with the steps Irradiating at least a portion of the subject (2) with electromagnetic radiation from a radar device (3), Receiving the electromagnetic radiation reflected or scattered at an interface (8) of the subject with the radar device (3), Acquiring time-dependent distance information about the distance of the interface (8) from the radar device (3), wherein the time-dependent distance information contains a measure of at least one vital parameter of the subject (2), and outputting a time-dependent measurement signal containing the distance information from the radar device (3) Reading the measuring signal into an evaluation device (4), Evaluation of the measuring signal in the evaluation device (4) and multilevel classifying a subject's condition (2) in at least one dimension selected from a group of dimensions including cognitive stress, fatigue, and stress levels based on the time-dependent measurement signal.

Images