Sensor system and method for calibrating a sensor system

Abstract

The invention relates to a sensor system for sensing electromagnetic radiation and a method for calibrating such a sensor system. The system includes a sensing element having a plurality of resonances, where the sensing element is configured to receive a plurality of calibration signals or signal components having different frequencies; a recording device for recording a response of the sensing element to at least two of the plurality of calibration signals or signal components, where a respective response to a calibration signal of the at least two calibration signals or signal components depends on an excitation of the calibration signal to one or more of the resonances; a data memory for storing at least a portion of the recorded response and / or information derived from at least a portion of the recorded response in a model associating the response with the frequency and / or signal level of the respective calibration signal or signal component; and a processor for converting the response of the sensing element to the EM signal to be analyzed into a spectrum of the EM signal using the model.

Description

Technical Field

[0001] The present invention relates to a sensor system for sensing EM (electromagnetic) radiation, in particular a quantum sensor system. The present invention also relates to a method for calibrating such a sensor system. Background Art

[0002] Quantum systems can be used to analyze the spectral components of electromagnetic (EM) radiation, particularly in the RF or optical range, through its energy levels. Such quantum systems may include ground state atoms, excited atoms (particularly Rydberg states), or atom-like systems such as NV defects in diamond. If the EM radiation resonates (or nearly resonates) to two energy levels in the quantum system, it will interact with the quantum system and change its state. The state of the atoms can then be read out by other means (e.g. optically).

[0003] Typically, only a single frequency component of EM can be analyzed by examining the interaction with a specific pair of energy levels. However, multiple frequency components can be analyzed to perform a spectral analysis of EM radiation. For example, multiple discrete frequency components can be analyzed by shifting the resonant frequency with a specific transition via an external field, or by intentionally examining the interaction with multiple energy levels (or a combination of both).

[0004] However, this approach is challenging because the estimation of the frequency and / or power of the EM radiation requires precise knowledge about the external fields. For example, the magnetic field distribution within the setup should be precisely known to predict the transition frequencies of atoms within the magnetic field.

[0005] Furthermore, EM radiation may interact with multiple energy levels. Depending on how the readout process is implemented, transitions to different energy states may not be distinguished, leading to ambiguity in the sensor output signal. If there are transitions at, for example, 3 GHz and 4 GHz that appear similar to those in the state readout, the user cannot unambiguously determine the frequency of the EM radiation.

[0006] Additionally, the atoms of a quantum sensor may exhibit nonlinear behavior at high power levels of EM radiation. Depending on the type of interaction, the transitions may saturate so that the sensor response no longer changes significantly when analyzing a single frequency component. Instead, "power broadening" may occur, meaning that transitions at nearby frequencies will also respond. With power broadening, it may be impossible to distinguish between a narrowband high-power signal and a broadband low-power signal. Summary of the invention

[0007] Therefore, there is a need to provide an improved sensor system and an improved method for calibrating a sensor system that can achieve frequency-resolved measurement of EM signals over a wide frequency spectrum.

[0008] This is achieved by the embodiments provided in the present invention. Advantageous implementations are further provided in the present invention.

[0009] According to a first aspect, the invention relates to a sensor system for sensing electromagnetic (EM) radiation, comprising: a sensing element having a plurality of resonances, wherein the sensing element is arranged to receive a plurality of calibration signals or signal components having different frequencies; and a recording device configured to record a response of the sensing element to at least two of the plurality of calibration signals or signal components, wherein each response depends on the excitation of one or more of the resonances by the respective calibration signal or signal component. The sensor system further comprises: a data memory configured to store at least a portion of the recorded responses and / or information derived therefrom in a model relating the responses to the frequencies and / or signal levels of the respective calibration signals or signal components; and a processor configured to convert a further response of the sensing element to the EM signal to be analyzed into a frequency spectrum of the EM signal using the model.

[0010] This achieves the following advantages: sensor systems based on multiple different resonances, in particular quantum sensor systems, can effectively measure the spectrum of EM signals with high accuracy. Since multiple signals of different frequencies are used for calibration, the entire spectrum of EM waves can be sensed over a high span, and the ambiguity problem of multiple resonances can be resolved. Calibration with real-world calibration signals takes into account individual sensor characteristics (e.g., precise electric / magnetic fields, or other characteristics of the sensing element) and compares favorably to less accurate techniques based purely on simulation or calculation.

[0011] For example, the individual responses to each calibration signal depend on whether one or more resonances are excited and the strength of the excitation. Multiple different resonances in the sensing element can be excited by the same calibration signal. For example, if the calibration signal does not excite a resonance, a weak response or no response at all is recorded.

[0012] Preferably, the calibration signal or signal component has a known signal level.

[0013] The calibration signal may be a continuous wave (CW) signal.

[0014] However, the calibration signal may also have a higher-level signal form (e.g., a rectangular pulse in the time domain). With such higher-level signal forms, the calibration process can be significantly accelerated because they contain multiple signals or signal components (e.g., peaks in the frequency space) that can be used for calibration.

[0015] For example, a single calibration signal may include multiple calibration signal components having different frequencies (eg, different narrow frequency peaks) used for calibration.

[0016] For example, the calibration signal is generated by a frequency comb.

[0017] The sensing element may comprise a sensing volume or may be formed by a sensing volume.

[0018] The sensing volume may include a gas stored in a gas chamber. Thus, the sensing volume may include a spatial distribution of mobile atoms. The gas may be a quantum gas. The excitation of resonance in the sensing volume may refer to the excitation of resonant transitions in atoms of the sensing volume.

[0019] Alternatively, the sensing element may comprise a single ion or a string of ions in an ion trap, or neutral atoms in an optical lattice.In general, the sensing element may comprise a single or multiple "receiving elements".

[0020] The model may be a mathematical model. The model may form or may include calibration data for the sensor system. The calibration data may be stored in a data memory of the sensor system and may be used during each measurement of an unknown EM signal to invert the raw sensor response into spectral information about the EM signal.

[0021] The EM signal to be analyzed may be an RF signal or an optical signal, preferably an RF signal or an optical signal in the frequency range covered by the calibration signal or signal component.

[0022] In one implementation form, the system also includes a magnetic field and / or electric field generator, which is configured to generate a temporally and / or spatially varying magnetic field and / or electric field in the sensing element; wherein the magnetic field and / or electric field generator is configured to change the strength of the magnetic field and / or electric field within a determined range during recording of respective responses to the at least two calibration signals or signal components.

[0023] For example, the magnetic field and / or electric field generator can be configured to generate a magnetic field gradient and / or a different magnetic field in the sensing element. The resonance in the sensing element can be shifted in frequency by the magnetic field and / or electric field. The magnetic field and / or electric field generator can include capacitor plates on two opposite sides of the sensing element.

[0024] For example, the magnetic and / or electric fields may be varied over time (ie in a scanning fashion).

[0025] In one implementation form, the system further comprises: an RF field generator configured to generate an RF field in the sensing element; wherein the frequency and / or intensity of the RF field varies over time, and / or wherein the RF field has a spatially varying intensity distribution.

[0026] For example, the energy level in the sensing element can be changed and / or shifted by an RF field (e.g., a microwave field). The tuning of the resonance is thereby achieved by varying the frequency or intensity of the microwave field / RF field over time, or by creating intensity variations in space in the sensing element (e.g., a sensing volume). The RF field can be generated as an RF signal.

[0027] In an implementation form, the recording device is configured to record the response of the sensing element to the at least two calibration signals or signal components as a function of the strength of the temporally and / or spatially varying magnetic and / or electric fields.

[0028] In this way, a unique response pattern for each calibration signal (and therefore each frequency) can be stored in the model.For example, the sensor response values ​​are recorded for a number of predetermined electric or magnetic field strengths applied to the sensing element.

[0029] In one implementation form, the model comprises at least one matrix or at least one look-up table relating the responses of the sensing elements to the frequencies and / or signal levels of corresponding calibration signals or signal components.

[0030] In one implementation form, the model is a neural network.

[0031] For example, the model is a machine learning (ML) model or a deep learning (DL) model.

[0032] The neural network may be trained on the responses of the sensing element to a calibration signal or signal components. In this case, the recorded responses may not be stored directly in the model, but some quantity may be derived from the recorded responses.

[0033] In one implementation form, the processor is configured to adjust the model such that the stored response exhibits a linear relationship with the signal level of the calibration signal or signal component.

[0034] In this way, a matrix or lookup table can define the linear relationship between the sensor response and the spectrum of the input signal. For example, a 1D spectrum matrix can be formed for each calibration signal. Measurements with 1000 calibration signals and 1000 recording points per calibration measurement can produce a 1000 x 1000 matrix. Linearization can be performed by rescaling the stored responses (values).

[0035] In one implementation form, in order to calculate the frequency spectrum of the EM signal, the processor is configured to linearize the further response of the sensing element to the EM signal and to apply the thus linearized further response to the model.

[0036] For example, the model defines a calibration curve for the sensor system. Applying the linearized response to the model may refer to multiplying it with the calibration curve (or a matrix defining the calibration curve).

[0037] For example, linearization of calibration data stored in the model and / or recorded data (otherwise responses) eliminates non-linear effects that may arise from power broadening due to high powers in the EM signal.

[0038] In one implementation form, the sensor system is a quantum sensor system.

[0039] The sensing element of a quantum sensor system may include multiple different resonances that produce multiple simultaneous responses to a single signal. The atoms in the sensing element may be in a ground state or an excited quantum state. The multiple resonances may include different types of resonant transitions.

[0040] For example, the sensing element includes Rydberg atoms. Rydberg atoms may include various resonances in the MHz-THz range, which may be shifted by an electrostatic field. The sensing element may include Rubidium 87 atoms.

[0041] In one example, the system includes a light source that excites atoms in the sensing element 102 to an excited quantum state.

[0042] Where the calibration signal excites one or more resonances, the response of the sensing element may be indicative of the strength of the corresponding resonant transitions in the sensing element.

[0043] In one implementation form, the system includes a signal source configured to send a plurality of calibration signals or signal components to the sensing element.

[0044] The signal source may include a signal generator configured to generate a calibration signal or signal component and / or an antenna configured to transmit the generated calibration signal or signal component.

[0045] Alternatively, the signal source may be a component of an external system (eg, a calibration system).

[0046] According to a second aspect, the invention relates to a method for calibrating a sensor system for sensing electromagnetic (EM) radiation, wherein the sensor system comprises a sensing element having a plurality of resonances. The method comprises the following steps: sending a plurality of calibration signals or signal components having different frequencies to the sensing element; recording a response of the sensing element to at least two of the plurality of calibration signals or signal components, wherein the respective response depends on the excitation of one or more of the resonances by the calibration signals or signal components; storing at least a portion of the recorded responses and / or information derived from at least a portion of the recorded responses in a model relating the responses to the frequencies and / or signal levels of the respective calibration signals or signal components; wherein the model enables the sensor system to convert a further response of the sensing element to the EM signal to be analyzed into a frequency spectrum of the EM signal.

[0047] In an implementation form, the method comprises the further step of converting the further response of the sensing element into a spectrum of the EM signal using the model.

[0048] In one implementation form, the sensor system comprises a magnetic field and / or electric field generator configured to generate a temporally and / or spatially varying magnetic field and / or electric field in the sensing element; wherein the method comprises the further step of varying the strength of the magnetic field and / or electric field within a determined range during recording of respective responses to the at least two calibration signals or signal components.

[0049] In an implementation form, the response of the sensing element to the at least two calibration signals or signal components is recorded as a function of the strength of the temporally and / or spatially varying magnetic and / or electric fields.

[0050] In one implementation form, the model comprises at least one matrix or at least one look-up table relating the responses of the sensing elements to the frequencies and / or signal levels of corresponding calibration signals or signal components.

[0051] In one implementation form, the model is adjusted so that the stored response exhibits a linear relationship with the signal level of the calibration signal or signal component.

[0052] In one implementation form, in order to calculate the spectrum of the EM signal, the further response of the sensing element to the EM signal is linearized and the thus linearized further response is applied to the model.

[0053] In one implementation form, the sensor system is a quantum sensor system.

[0054] For example, where the calibration signal excites one or more resonances, the response of the sensing element may be indicative of the strength of the corresponding resonant transitions in the sensing element.

[0055] The above description regarding the sensor system according to the first aspect of the invention applies correspondingly to the method according to the second aspect of the invention.

[0056] In particular, the method according to the second aspect of the invention may be used to calibrate a sensor system according to the first aspect of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The above aspects and implementation forms of the present invention will be explained in the following description of specific embodiments with reference to the accompanying drawings, in which:

[0058] Figure 1 shows a schematic diagram of a sensor system according to one embodiment;

[0059] Figure 2 shows the energy levels of a rubidium-87 atom according to one embodiment;

[0060] Figure 3 shows a recorded response of a sensing element according to one embodiment;

[0061] Figure 4 A flow chart showing a method for calibrating a sensor system according to one embodiment; and

[0062] Figure 5 A flow chart of a calibration and measurement process according to one embodiment is shown. DETAILED DESCRIPTION

[0063] Figure 1 A schematic diagram of a sensor system 100 according to one embodiment is shown.

[0064] The sensor system 100 comprises: a sensing element 102 having a plurality of resonances, wherein the sensing element 102 is arranged to receive a plurality of calibration signals or signal components having different frequencies; a recording device 103 configured to record a response of the sensing element 102 to at least two of the plurality of calibration signals or signal components, wherein each response depends on the excitation of one or more of the resonances by the respective calibration signals or signal components; a data storage 104 configured to store at least a portion of the recorded responses and / or information derived from at least a portion of the recorded responses in a model relating the responses to frequencies and / or signal levels of the respective calibration signals or signal components; and a processor 106 configured to convert a further response of the sensing element to the EM signal to be analyzed into a frequency spectrum of the EM signal to be analyzed using the model.

[0065] Each calibration signal may be a narrowband signal of a different frequency within a determined frequency range (eg, a calibration range). The calibration signal may include a radio frequency (RF) signal and / or an optical signal.

[0066] In particular, a "corresponding calibration signal / signal component" of a recorded response is the calibration signal / signal component whose response is recorded.

[0067] The recording device 103 may be configured to record a response of the sensing element 102 to each of the calibration signal(s), wherein the response to each calibration signal depends on the excitation of one or more resonances by the calibration signal.

[0068] The recording device 103 may include an optical recording device, such as a camera or a photodiode. The response and / or the additional response may be an optical signal and / or an electrical signal. The recorded response of each calibration signal may represent the intensity of the optical response of the sensing element 102 to the calibration signal.

[0069] The calibration signal may be a CW (continuous wave) signal. Alternatively, the calibration signal may have a higher order signal form (eg the calibration signal comprises rectangular pulses in the time domain).

[0070] Instead of multiple calibration signals, several or even a single calibration signal with multiple signal components may be used, wherein the multiple signal components have different frequencies and wherein the responses to these signal components are recorded.

[0071] The calibration signal can be generated by a signal source 101. The signal source 101 can be a component of the sensor system 100 or an external system (e.g., a calibration system). For example, the signal source 101 includes a signal generator, such as a comb generator. The signal generator can be configured to generate the calibration signal as a step signal or a swept signal. The signal generator can generate the calibration signal in the form of (several) frequency lines or a series of harmonics. The signal source 101 can also include an antenna.

[0072] The data storage 104 may be a memory such as a flash memory. The processor 106 may be a microprocessor or a controller.

[0073] The data store 104 may store the individual responses to each (recorded) calibration signal (or a portion thereof), or information derived from the individual responses, in a model that relates the responses to the frequency and / or signal level of the corresponding calibration signal.

[0074] Preferably, the sensing element 102 is formed of or comprises a sensing volume, wherein the sensing volume comprises one or more atoms (eg, atoms in a gaseous or solid state) exhibiting resonance.

[0075] The model 105 may include at least one matrix or at least one lookup table that relates the response of the sensing element 102 to the frequency and / or signal level of the respective calibration signals.

[0076] The model 105 may be a mathematical model. The model 105 may form or may include calibration data and / or may define a calibration curve for the sensor system 100. The calibration data and / or calibration curve may be stored in a data memory of the sensor system and may be used during each measurement of the (unknown) EM signal to be analyzed to convert the raw sensor response (i.e., the further response) into spectral information about the EM signal. For example, the raw sensor response may include a plurality of response values ​​recorded using different magnetic and / or electric fields applied to the sensing element.

[0077] Alternatively, the model may be a neural network. The neural network may be trained based on the response of the sensing element to the calibration signal / signal components.

[0078] The model and / or instructions for executing the model may be stored in the data store 104. In one example, the recording device 103 and / or the processor 106 may be configured to forward at least a portion of the recorded response (and / or information about the response) to the data store to be stored / inserted into the model.

[0079] Preferably, the sensor system 100 is a quantum sensor or a quantum sensor system. In particular, the sensor system 100 is a quantum spectrum analyzer. The sensing element 102 may include a quantum gas in a ground state or an excited quantum state exhibiting multiple resonances. For example, the sensing element 102 includes rubidium atoms in gaseous form.

[0080] Preferably, multiple frequency components of the EM radiation can be analyzed by the quantum sensor system 100. This can be achieved by analyzing the frequency components simultaneously, for example by using different energy levels of multiple atoms simultaneously, by applying an electric or magnetic field gradient, by analyzing the frequency components quickly enough so that the signal does not change during the measurement (for example by sweeping the external field, or by performing consecutive measurements of different energy levels).

[0081] Figure 1 The illustrated exemplary system 100 includes a magnetic and / or electric field generator 107 configured to generate a temporally and / or spatially varying magnetic and / or electric field in the sensing element 102 .

[0082] For example, the magnetic and / or electric field generator 107 is configured to vary the strength of the magnetic and / or electric field in the sensing element 102 within a specific range during each recording of a response to a corresponding calibration signal.

[0083] For example, the magnetic and / or electric field generator 107 is configured to vary the magnetic and / or electric field in time (ie, by sweeping the magnetic and / or electric field).

[0084] The system 100 may also include an RF field generator instead of, or in addition to, the M / E field (magnetic field and / or electric field) generator 107. The RF field generator may be configured to generate an RF field in the sensing element; wherein the frequency and / or intensity of the RF field varies over time, and / or wherein the RF field has a spatially varying intensity distribution.

[0085] Figure 2 The ground state energy level of a rubidium 87 atom according to the applied magnetic field is shown. As can be seen, rubidium 87 has eight different energy levels. Resonance can be excited by an EM wave whose frequency corresponds to the transition frequency between two energy levels (e.g. Figure 2 By varying the magnitude of the applied magnetic field, the energy level, and therefore the corresponding resonant frequency, can be shifted.

[0086] Where the calibration signal excites one or more resonances, the response of the sensing element may be indicative of the strength of the corresponding resonant transitions in the sensing element 102 .

[0087] Due to the large number of transitions, it may not be possible to distinguish which exact transitions are responsible for a particular response of the sensor to a signal. For example, if a 10 GHz signal is applied to a sensing element 102 comprising rubidium 87, thereby applying a magnetic field gradient, there are multiple magnetic field strengths at which a signal of this frequency can cause transitions. Typically, the detection scheme cannot distinguish between different transitions; it only checks whether there is a transition from the ground state (bottom line) to an excited state (top line).

[0088] Figure 3 The recorded response of a sensing element comprising rubidium 87 atoms to a calibration signal of known frequency according to one embodiment is shown. The sensor response is recorded with a varying magnetic field. It can be seen that the response signal comprises three peaks caused by the multiple energy levels of the rubidium ground state.

[0089] The measurement may be repeated with multiple calibration signals or signal components, each calibration signal / signal component having a different frequency. The response to each calibration signal has a unique signature of multiple peaks of the sensor response / static field.

[0090] The recording device 103 of the sensor system 100 can be configured to record such a unique response pattern based on the changing magnetic and / or electric fields of each calibration signal or signal component. To calibrate the sensor device, the recording device 103 can store at least a portion of the response (e.g., sensor response "values" for multiple magnetic field values) in a model together with the frequency and / or signal power of the corresponding calibration signal or signal component.

[0091] The processor 106 may also be configured to linearize the model so generated, for example by rescaling the stored response values ​​so that they exhibit a linear relationship with the input signal level for all recorded frequencies. Thus, for example, if the power level of the input signal exceeds a threshold, the nonlinear effects caused by power broadening may be removed from the model.

[0092] When measuring unknown EM signals, the processor 106 may also rescale the recorded raw sensor response (additional response) and apply this rescaled (normalized) data to the model, in particular to the calibration data.

[0093] In particular, the sensor system 100 may perform a linearization step in which nonlinear effects (eg, due to power broadening at high signal levels) are removed from the sensor response and / or the model.

[0094] Furthermore, by focusing the EM signal onto a small element with an antenna, and taking into account the properties of the sensor system 100 and the calibration mentioned above, increased sensitivity may be achieved.

[0095] Due to the calibration, the quantum sensor system 100 described above allows for spectral analysis of EM signals with high frequency and power accuracy, resolving the ambiguity of multiple resonances.

[0096] Figure 4 A method for calibrating a sensor system 100 (eg, Figure 1 The sensor system 100 is preferably a quantum sensor system.

[0097] The method 40 comprises the following steps: sending 41 a plurality of calibration signals or signal components having different frequencies to a sensing element 102 of the sensor system 100; recording 43 a response of the sensing element 102 to at least two of the plurality of calibration signals or signal components, wherein the respective responses depend on the excitation of one or more resonances by the calibration signals or signal components; and storing 44 at least a portion of the recorded responses and / or information derived from at least a portion of the recorded responses in a model 105, the model 105 associating the responses with the frequencies and / or signal levels of the respective calibration signals or signal components; wherein the model 105 enables the sensor system 100 to convert an additional response of the sensing element 102 to the EM signal to be analyzed into a spectrum of the EM signal.

[0098] The sensor system 100 may include a magnetic field and / or electric field generator 107 configured to generate a temporally and / or spatially varying magnetic field and / or electric field in the sensing element. Figure 4 As shown, the method 40 may also comprise the further step of varying 42 the strength of the magnetic and / or electric field within a determined range during each recording of a response to a calibration signal or signal component.

[0099] The response of the sensing element to each calibration signal or signal component may be recorded 43 as a function of the varying magnetic and / or electric field strength.

[0100] Where the calibration signal or signal component excites one or more resonances, the response of the sensing element 102 is indicative of the strength of the corresponding resonant transitions in the sensing element.

[0101] The model may include at least one matrix or at least one look-up table relating the response of the sensing element to the frequency of the calibration signal.

[0102] In a further step of method 40, the model (or data stored in the model) may be adjusted so that the stored response exhibits a linear relationship with the signal level of the calibration signal or signal component. In addition, in order to calculate the spectrum of the EM signal, the additional response of the sensing element to the EM signal may also be linearized and applied to the model. In this way, non-linear effects due to power broadening at higher input signal levels may be mitigated.

[0103] The spectrum of the EM signal so determined may be the RF spectrum, ie the EM input signal in the frequency domain.

[0104] Figure 5 A flow chart of a calibration process 60 and a measurement process 70 of a quantum sensor system 100 according to one embodiment is shown.

[0105] The calibration process 60 may be substantially equivalent to Figure 4The method 40 is shown and may be used to calibrate a quantum sensor system 100 including a sensing element 102 having, for example, rubidium atoms.

[0106] The calibration process 60 may include the following sequence:

[0107] 1) First, the calibration signal source 61 generates a signal, such as a CW signal, whose signal frequency changes in small steps. These signals are forwarded 62 to the sensing element 102 of the quantum sensor system 100, and the spectrum of each signal frequency (the relationship between the signal response and the magnetic field strength, such as Figure 3 ). Therefore, it can be assumed that the power is small enough so that saturation and / or power broadening do not play a role. In this way, calibration data can be recorded.

[0108] 2) In a next step, the measured calibration data 63 may be rescaled so that the sensor response is linear with respect to the input power of the (CW) signal. This may be done by theoretical calculations or with the help of additional calibration data (sensor responses measured for different CW signal powers). Preferably the rescaled calibration data 63 may be fed to an intermediate model 64 which relates the sensor response to the input signal frequency.

[0109] 3) In the linear range of small signal powers, the sensor response is usually a linear transformation of the input data spectrum. Therefore, the calibration data can be represented by a 2D matrix, where the first dimension is given by the static magnetic field and the second dimension is given by the input signal frequency.

[0110] For example, the calibration sequence 60 may be run at the factory before the sensor system 100 is shipped to a customer, and / or run in the field by a customer or a service technician.

[0111] The calibration process 60 can be applied to any given further sensor response of the EM signal to be analyzed, for example, by first rescaling the measured data (similar to step 2 in the calibration sequence 60) and solving the linear equation "sensor_response = calibration_matrix * input_spectrum" for input_spectrum. This solution can reconstruct the input spectrum of the EM signal to be analyzed and resolve any ambiguity caused by multiple resonances of atoms in the quantum sensor.

[0112] One possibility to speed up the calibration sequence 60 is not to measure the spectrum of a large number of frequencies of the calibration signal, but to measure only a few frequencies and interpolate the results between them. By applying a frequency comb, this can even be done in a single measurement.

[0113] For large signal powers, the sensor response at resonance may saturate and the line(s) in the sensor response caused by the resonant transmission may broaden, i.e. nearby frequencies also respond. This effect can be compensated for in step 2) of the calibration procedure above by theoretical calculation or by calibration data at high signal powers.

[0114] Due to power broadening, it may not always be possible to distinguish between a narrowband high power signal and a broadband low power signal. This ambiguity can be resolved by measuring different field amplitudes of the same EM radiation. In this case, additional measurements can be included in the calibration routine 60.

[0115] The measurement process 70 that can be performed after calibration may include: detecting a test signal 71 (i.e., a signal to be analyzed) with a quantum sensor system 72, processing the sensor response in a processor 73 by comparing the sensor response with calibration data and / or an intermediate model 74, and calculating a signal spectrum 75 of the test signal 71 based on the comparison.

[0116] Prior to the calibration process 70, a pre-calibration may be performed by theoretical calculations and / or simulations. For example, depending on the specific atomic species in the sensing element 102, the frequency and intensity of the transitions may be accurately simulated and / or calculated by solving the Schrödinger equation for these atoms. Thus, in principle, the response of the atoms may be accurately predicted. This information may be used to infer information about an unknown EM signal to be analyzed from the sensor response.

[0117] However, to infer this information by calculation / simulation alone, precise knowledge of the magnetic and / or electric fields in the sensor setup is necessary. For example, the magnetic field distribution within the sensing element 102 must be precisely known in order to predict the transition frequencies of atoms within the magnetic field.

[0118] In order to enhance the intensity of the electromagnetic radiation (i.e. to increase the sensitivity of the sensor system 100), the quantum sensor 100 can be placed in an antenna or resonator that focuses the incident EM radiation onto a small element. However, in order to calculate the intensity of the initial EM radiation, in this case the field distribution generated by the antenna / resonator should be known. In this way, an RF spectrum analyzer that measures the spectrum of the RF signal contained in the cable / waveguide (instead of measuring the free space electromagnetic field) can be implemented. For example, in order to access the signal with atoms, the signal can be converted by an antenna into a known electric or magnetic field distribution, and / or the atoms can be brought close to the field of the waveguide.

[0119] In addition, the field distribution of static magnetic and / or electric fields and RF signals may be affected by thermal drift, aging and / or manufacturing uncertainties, which are difficult to consider in theoretical calculations or simulations. In order to take these uncertainties into account, the calibration methods and routines 40, 60 described above allow the quantum sensor 100 to be calibrated by a well-known reference signal.

[0120] All features mentioned above or shown in the drawings can be combined with one another in any advantageous manner within the scope of the invention.

Claims

1. A sensor system for sensing electromagnetic (EM) radiation, comprising: a sensing element having multiple resonances, wherein the sensing element is arranged to receive multiple calibration signals or signal components having different frequencies; a recording device configured to record a response of the sensing element to at least two of the plurality of calibration signals or signal components, wherein each response depends on the excitation of one or more of the resonances by the respective calibration signal or signal component; a data store configured to store at least a portion of the recorded responses and / or information derived from at least a portion of the recorded responses in a model relating the responses to frequencies and / or signal levels of the corresponding calibration signals or signal components; and A processor is configured to convert a further response of the sensing element to the EM signal to be analyzed into a frequency spectrum of the EM signal using the model.

2. The system according to claim 1, further comprising: a magnetic field and / or electric field generator configured to generate a temporally and / or spatially varying magnetic field and / or electric field in the sensing element; Therein, the magnetic field and / or electric field generator is configured to change the strength of the magnetic field and / or electric field within a determined range during recording of respective responses to the at least two calibration signals or signal components.

3. The system according to claim 2, wherein: The recording device is configured to record a response of the sensing element to the at least two calibration signals or signal components as a function of the strength of the temporally and / or spatially varying magnetic and / or electric field.

4. The system according to claim 1, further comprising: an RF field generator configured to generate an RF field in the sensing element; wherein the frequency and / or the intensity of the RF field varies over time, and / or wherein the RF field has a spatially varying intensity distribution.

5. The system according to claim 1, wherein: The model comprises at least one matrix or at least one look-up table relating the response of the sensing element to the frequency and / or signal level of the corresponding calibration signal or signal component.

6. The system according to claim 1, wherein: The model is a neural network.

7. The system according to claim 1, wherein: The processor is configured to adjust the model such that the stored response exhibits a linear relationship with the signal level of the calibration signal or signal component.

8. The system according to claim 1, wherein: In order to calculate the frequency spectrum of the EM signal, the processor is configured to linearize the further response of the sensing element and to apply the thus linearized further response to the model.

9. The system according to claim 1, wherein: The sensor system is a quantum sensor system.

10. The system of claim 1, further comprising: A signal source is configured to send the plurality of calibration signals or signal components to the sensing element.

11. A method of calibrating a sensor system for sensing electromagnetic (EM) radiation, wherein: The sensor system comprises a sensing element having a plurality of resonances, the method comprising the steps of: sending a plurality of calibration signals or signal components having different frequencies to the sensing element; recording a response of the sensing element to at least two of the plurality of calibration signals or signal components, wherein each response depends on excitation of one or more of the resonances by the respective calibration signal or signal component; and storing at least a portion of the recorded responses and / or information derived from at least a portion of the recorded responses in a model relating the responses to frequencies and / or signal levels of the corresponding calibration signals or signal components; Therein, the model enables the sensor system to convert the further response of the sensing element to the EM signal to be analyzed into the frequency spectrum of the EM signal.

12. The method according to claim 11, in, The sensor system comprises a magnetic field and / or electric field generator, wherein the magnetic field and / or electric field generator is configured to generate a temporally and / or spatially varying magnetic field and / or electric field in the sensing element; wherein the method further comprises the following steps: During the recording of the respective responses to the at least two calibration signals or signal components, the strength of the magnetic and / or electric field is varied within a defined range.

13. The method according to claim 12, wherein: The response of the sensing element to the at least two calibration signals or signal components is recorded as a function of the strength of the temporally and / or spatially varying magnetic and / or electric field.

14. The method according to claim 11, wherein: The model comprises at least one matrix or at least one look-up table relating the response of the sensing element to the frequency and / or signal level of the corresponding calibration signal or signal component.

15. The method according to claim 11, wherein: The model is adjusted so that the stored response exhibits a linear relationship with the signal level of the calibration signal or signal component.

16. The method according to claim 11, wherein: In order to calculate the frequency spectrum of the EM signal, the further response of the sensing element to the EM signal is linearized and the thus linearized further response is applied to the model.

17. The method according to claim 11, wherein: The sensor system is a quantum sensor system.

Images