Method and device for determining and calculating a scattered radiation spectrum and method for compressing data

Abstract

The invention relates to a device (1) and a method (20) for determining a spectrum (X) of scattered radiation (S). The invention further relates to a method (70) for calculating the spectrum (X) and a method for compressing unstructured data (60) of known distribution. To be able to determine the spectrum (X) as precisely as possible and to derive from this the characteristics of materials that scatter laser pulses (P), the invention proposes that at least one characteristic of the laser pulse (P) is determined and that a spectrum analyzer (5) is used for this. Frequencies (F) of laser pulses (P) and volumes (M) of backscattered radiation (S) are combined into frequency and volume values (F, M) to calculate the spectrum (X). The most frequent data values are deleted from the data to compress the data (60).

Claims

1. The method for determining a spectrum of scattered radiation, wherein several laser pulses are successively scattered, the method comprising the steps of: measuring the scattered radiations per height channel and per laser pulse; determining at least one characteristic of each laser pulse; and linking the at least one characteristic of each laser pulse with the scattered radiation per height channel and per laser pulse for determining the spectrum of scattered radiation; wherein the method comprises the step of determining the frequency of each laser pulse as the characteristic.

2. The method according to claim 1, further comprising: detecting a portion of the scattered radiation for each laser pulse; and linking the quantity of detected scattered radiation with the determined characteristic of the scattered laser pulse.

3. The method according to claim 1, further comprising: determining a quality feature of the laser pulses and using only those laser pulses and scattered radiation associated with those laser pulses to determine the spectrum whose quality feature satisfies a quality criterion.

4. The method according to claim 3, further comprising determining the spectrum of the laser pulses for determining the quality feature.

5. The method according to claim 1, further comprising determining a frequency of a reference beam before or after at least one of the laser pulses.

6. The method according to claim 5, further comprising determining the frequency of the reference beam between several of the laser pulses.

7. The method according to claim 1, further comprising scattering laser pulses having different frequencies in succession.

8. The method according to claim 7, further comprising setting the frequency of the laser pulses based on the quantity of the detected scattered radiation.

9. The method according to claim 1, further comprising determining absolute frequencies of the spectrum of the scattered radiation from their quantity distribution in relation to the at least one determined characteristic of the laser pulses.

10. A device for determining a spectrum of scattered radiation, comprising: a pulsed laser whose laser pulses emitted during operation of the device are scattered for generating the scattered radiation; a radiation sensor that is configured to at least partially receive the scattered radiation, and a spectrum analyzer which is connected with the pulsed laser for receiving laser radiation and which is further configured to determine a characteristic of the laser pulses; wherein the characteristic is the frequency of the laser pulses; and wherein the device comprises a computing unit that links the determined frequency of the laser pulses with the quantity of detected scattered radiation per height channel and per laser pulse.

11. The device according to claim 10, wherein the radiation sensor is configured to generate a quantity signal that is dependent on the quantity of the received scattered radiation.

12. The device according to claim 10, wherein the spectrum analyzer comprises an interferometer that superimposes incident laser light with itself, a measurement signal converter, and a lens, wherein the lens is arranged between the interferometer and the measurement signal converter in a beam path of laser light exiting from the interferometer and propagating to the measurement signal converter.

13. The device according to claim 10, further comprising a reference laser, wherein the spectrum analyzer is arranged in the beam path of the reference laser.

14. A method for calculating a scattered radiation spectrum, wherein frequencies of laser radiation to be scattered are linked with the quantity of backscattered radiation per height channel and per laser pulse, the method comprising the steps of: measuring frequencies of the scattered laser radiation associated with frequency intervals; and in each case combining quantities of scattered radiation associated with the measured frequencies into frequency values and quantity values; wherein the frequency values and quantity values are determined so as to approximate values of an expected theoretical spectrum.

Description

(1) It is shown in:

(2) FIG. 1 a schematic diagram of an exemplary embodiment of a device according to the present invention for determining a scattered radiation spectrum;

(3) FIG. 2 a schematic diagram of an exemplary embodiment of a method according to the present invention for determining a scattered radiation spectrum in form of a flow diagram;

(4) FIG. 3 a schematic diagram of an exemplary embodiment of a method according to the present invention for data compression in form of a flow diagram;

(5) FIG. 4 a schematic diagram of data;

(6) FIG. 5 a schematic diagram of an exemplary embodiment of a method according to the present invention for calculating a scattered radiation spectrum in form of a flow diagram; and

(7) FIG. 6 a schematic diagram of a scattered radiation spectrum.

(8) The structure and function of a device according to the invention for determining a scattered radiation spectrum will be described first with reference to the exemplary embodiment of FIG. 1.

(9) FIG. 1 shows schematically the device 1 for determining a scattered radiation spectrum with a pulsed laser 2 and a radiation sensor 3. The pulsed laser 2 is constructed to emit laser pulses P, wherein the emitted laser pulses P are scattered and scattered radiation S from the scattered laser pulses P is at least partially received by the radiation sensor 3. The laser pulses P are directed, for example, to a gas volume V in order to derive from the scattered radiation S at least one property of the gas volume V. Alternatively, the laser pulses P can also be directed to another volume, for example a volume of a liquid, or other structures for determining at least one property of these volumes or structures and can be scattered by these volumes or structures.

(10) The laser pulse P emitted from the pulsed laser 2 propagates along an optical axis O2 in the direction of the gas volume V to be examined. A beam splitter 4, on which the laser pulse P is incident and by which the laser pulse P is scattered, is disposed along the optical axis O2 between the pulsed laser 2 and the gas volume V. A first portion P of the laser pulse P exits the beam splitter 4 towards the gas volume V, where it is scattered. A second portion P of the laser pulse P is guided by the beam splitter 4 along an optical axis O4 of the beam splitter 4 to a spectrum analyzer 5. At least one characteristic of the laser pulse P can be determined with the spectrum analyzer 5 based on the second portion P. This characteristic is, for example, a selected frequency F of the laser pulse P and, in particular, the frequency F, where a frequency spectrum of the laser pulse P has its maximum or its center of gravity. Furthermore, a quality feature Q of the laser pulse P can be determined with the spectrum analyzer 5. The quality feature Q may include, for example, characteristics of laser modes or broadband emissions of the pulsed laser 2.

(11) The spectrum analyzer 5 includes, for example, a laser radiation and in particular the second portion P of the laser pulse P with an interferometer superimposed on itself, a measurement signal converter, and a lens. The lens is preferably a convex lens arranged between the interferometer and the measurement signal converter in an optical path of laser light exiting the interferometer and guided to the measurement signal converter. The convex lens images onto the measurement signal converter an interference pattern of the laser radiation that is superimposed on itself.

(12) The measurement signal converter generates a frequency signal Fs and/or a quality signal Qs, which can be outputted by the spectrum analyzer 5 to a computing unit 6 of the device 1. The computing unit 6 compares, for example, the quality signal Qs with a quality criterion. If the laser pulse P does not satisfy the quality criterion, then the data of the laser pulse P and its scattered radiation S are not considered in the determination of the scattered radiation spectrum and are for example discarded. If the laser pulse P satisfies the quality criterion, then its frequency F represented by the frequency signal Fs is at least temporarily stored in the computing unit 6. The computing unit 6 is also connected with the radiation sensor 3 for receiving the measurement signal. The radiation sensor 3 is in the exemplary embodiment of FIG. 1 configured to generate as a measurement signal a quantity signal Ms representative of the quantity (amount) M of received scattered radiation S and to transmit this signal to the computing unit 6. In the computing unit 6, at least the frequency F of the laser pulse P and the quantity M of the received scattered radiation S are linked together and outputted, for example as a data packet FM, to a storage device 8.

(13) The computing unit 6 can determine, based on the quantity signal MS representative of the quantity M of scattered radiation S received by the radiation sensor 3 in combination with the frequency signal Fs representative of the frequency F of the laser pulse P, whether the frequency F of the laser pulses P generated by the pulsed laser 2 should be changed. For this purpose, a decision rule may be stored in the computing unit 6. The computing unit 6 can be connected with the pulsed laser 2 via a control line 8 for transmitting control signals and thereby affect the frequency F of the laser pulses P generated by the pulsed laser 2. If the pulsed lasers 2 is excited by a seeder laser 9 (shown by dashed lines), then the control line 8 may also be routed as a control line 8 (shown by dashed lines) from the computing unit 6 to the seeder laser 9, so that the computing unit 6 influences the frequency of the excitation radiation A of the seeder laser 9 that excites the laser pulses P.

(14) The frequency F of the laser pulses P cannot readily be determined absolutely with the spectrum analyzer 5, but only relatively. If the spectrum analyzer 5 changes over time, for example due to temperature changes, then the frequencies F of the laser pulses P cannot be repeatedly accurately measured. In order to recognize a change of the spectrum analyzer 5, the device 1 may include a reference laser 10. The reference laser 10 emits during the operation of the device 1 reference radiation R with a reference frequency Fr to the spectrum analyzer 5. The reference frequency Fr was previously determined with the spectrum analyzer 5 at least once under controlled conditions. Since the reference frequency Fr is stable even over longer periods of time, changes of the spectrum analyzer 5 can be determined by re-measuring the reference frequency Fr. If the determined reference frequency Fr deviates too much from a previously determined reference frequency Fr, then this indicates a change in the spectrum analyzer 5. The reference frequency Fr needs thereby not be absolutely known. Measurements of the frequencies F of the laser pulses P can even be corrected based on the measurements of the reference frequency Fr performed at different times and on measurements producing mutually different measurement results.

(15) In order to determine longer-term changes in the spectrum analyzer 5, it is sufficient to determine the reference frequency Fr with the spectrum analyzer 5 occasionally, for example every hour. In order to identify also short-term changes of the spectrum analyzer 5, the reference frequency Fr can be determined with the spectrum analyzer 5, for example, between two laser pulses P. Preferably, the reference frequency Fr is measured between all laser pulses P.

(16) As an alternative to coupling the second portion P out of the laser pulse P into the beam splitter 4, the spectrum analyzer 5 can also be connected directly with the pulsed lasers 2. A portion of the laser pulses P may enter directly from the pulsed laser 2 into the spectrum analyzer 5, where the frequency F or the quality feature Q of the laser pulses P can then be determined. Such a spectrum analyzer is illustrated in the exemplary embodiment of FIG. 5 by dashed lines and provided with the reference numeral 5.

(17) The gas volume V is, for example, a gas volume in the upper atmosphere, and in particular in the mesosphere. In order to be able to determine properties of a gas at a low pressure, the gas should be irradiated with many laser pulses P. In particular, when the gas volume V is located in the upper atmosphere and, for example, in the mesosphere, at best only a few photons of scattered radiation S per laser pulse P reach the radiation sensor 3. However, the scattered radiation spectrum can nevertheless be determined due to the large number of employed laser pulses P. To determine the gas properties, the gas must be irradiated with laser pulses P of different frequencies F, so that laser pulses P of different frequencies F are consequently also scattered.

(18) FIG. 2 shows a first exemplary embodiment of a method according to the invention. The same reference numerals are used for elements that correspond in function and/or structure to the elements of the exemplary embodiment of FIG. 1.

(19) The method 20 for determining a spectrum of scattered radiation is shown schematically in FIG. 2 in form of a flowchart. In a first method step 21, a laser pulse P is emitted. In the following step 22, the frequency F of the laser pulse P is determined.

(20) The laser pulse P, or at least the first portion P of the laser pulse P, is scattered in method step 23 and the quantity M of generated scattered radiation S and received by the radiation sensor 3 is determined in method step 24. In a subsequent method step 25, the frequency F determined in method step 22 and the quantity M of scattered radiation S measured in step 25 are linked with one another. For this purpose, for example the frequency signal Fs and the quantity signal Ms are used. The frequency F associated with the quantity M is stored as a data packet FM in method step 26.

(21) Before or after method step 21 and in particular between method steps 22 and 23, a reference frequency Fr can be measured in optional process steps 27, 28 in order to detect changes in the frequency measurement.

(22) As indicated by the arrow 29, the method 20 may be executed repeatedly and for example up to 1000 times per second, up to 10,000 times per second or even up to 100,000 times per second or even more frequently. The frequency F of the laser pulse P can be changed in each pass of the method 20, in order to determine the scattered radiation spectrum with the largest possible number and/or widely spaced frequencies F.

(23) In method step 22, the quality feature Q of the laser pulse P can also be determined and the signal Qs representing the quality feature Q can be outputted and used in method step 25. If the quality feature Q of the laser pulse P does not correspond to a quality criterion, then neither the frequency F of the laser pulse P nor the quantity M of received scattered radiation S is stored in method step 26.

(24) The method 20 of the exemplary embodiment of FIG. 2 thus generates a large amount of data, for example, one terabyte for each day of measurements. Even modern data storage devices are unable to handle this amount of data generated every day when the measurements are performed daily. Conventional compression method for data are incapable of compressing the data generated by the method 20, since these data have little or no structure, i.e. their structure corresponds to white noise.

(25) In particular, quantity data, i.e. count data, are generated when determining properties of gases in the upper atmosphere that correspond to the number of backscattered photons per laser pulse P. The expected distribution of the quantity M and the number of backscattered photons is known. For example, no photon is received by the radiation sensor 3 for many or even for most of the laser pulses P. As the second-most frequent event, a single photon is received. As the third-most frequent event, two scattered photons are received. Three or more scattered photons are received only in exceptional cases.

(26) Because of the lack of data structures, these quantity data or count data can be compressed by conventional methods only poorly or not at all, because known compression methods compress data based on repetitive structures which, however, the measurement data lack. In particular, conventional compression methods are often incapable of significantly compressing the measurement data loss-free. In a lossy compression, however, the scattered radiation spectrum might be so greatly distorted that it cannot be determined.

(27) FIG. 3 shows schematically a compression method 40, with the unstructured count data having a structure that corresponds to for example white noise, but has a known distribution, for loss-free compression. The compression method 40 is shown highly schematically in FIG. 3 as a flowchart.

(28) In a first method step 41, count data is generated and, for example, the number or the quantity M of backscattered photons of a laser pulse P are counted in a channel, i.e. a gas volume with a predetermined thickness. When the laser pulses P are scattered by a gas volume V in the upper atmosphere, for example in the mesosphere, then most frequently zero photons, less frequently a single photon, even more rarely two photons and only in exceptional cases three or more photons per laser pulse P and height channel are received and counted. In method step 42 following method step 41, it is checked whether no photon at all was counted. If no photon is received by the radiation sensor 3 and if this is the most frequent event, then it is decided in method step 42 not to store at least the measured number of zero photons and possibly also not the data for the height channel as data, but to instead discard these data.

(29) In method step 43, it is checked whether for example exactly one scattered photon has been received by the radiation sensor 3 as the second-most frequent result. For example, if just for one photon was received for one of the channels, then these data representative of a measurement, e.g. the number of height channels, are stored in method step 44. In particular, the distance to the height channel, where the same number of photons was previously received, is stored.

(30) When exactly two photons were received, it is decided in method step 45 to store data for this measurement. However, it is not necessary to store the value two. To save memory space, it is sufficient to store the distance zero.

(31) Because larger data sets with count data must usually be compressed, the compression method 40 is carried out repeatedly, as indicated by the arrow 46.

(32) FIG. 4 shows schematically uncompressed data 60 and compressed data 61. The same reference numerals are used for elements corresponding in function and/or structure to the elements of the exemplary embodiments of the previous figures.

(33) The uncompressed data 60, and the compressed data 61 include at least one source data field S1 and, for example, three source data fields S1, S2, S3, wherein source data can be stored. Source data are, for example, data of a first laser pulse P and, in particular, its determined frequency F1, its intensity I1, and possibly its quality Q1. Several address data fields K are associated with the at least one source data field S1. The number of address data fields K may correspond to the number of height channels and one address in the address field may be assigned to each of the height channels. In the illustrated exemplary embodiment, the uncompressed data 60 include address data fields K1 to K8000 for one of the laser pulses P 8000. The address data fields have a minimum length of 13 bits to allow the digital display of 8000 different address data values.

(34) A target data field Z, where the quantity M of scattered radiation S for the associated height channel is registered for one of the scattered laser pulses P, is assigned to each of the address data fields K. In the illustrated exemplary embodiment, the most-frequent target data value is zero. This means that zero photons are received from the scattered laser pulse P for each channel. On the other hand, instead of zero photons, one photon or two photons were received for the respective address data K4 and K5.

(35) The uncompressed data 60 furthermore include data for more scattered laser pulses P. For example, the uncompressed data 60 include data for 50,000 laser pulses P.

(36) The compressed data 61 have, instead of the address data K1 to K8000 and the target data, only distance data D to a first data field or to a previous field that does not include the most frequently occurring target data value zero. For example, the target data value of one photon for the address data field K containing the address data K4 is represented by the distance data value three, which indicates that a target data value corresponding to the second-most frequent target data and for example the value of one photon exists only in the address data field K1 having a distance of three address data. The value of one following the value of three in the next field expresses that a target data value deviating from the most-frequent target data is contained already in the next target data field associated with the address data field K. However, the encoded target data does not match the value of one photon, but the value of two photons. To be able to encode the target data value of two photons, the smallest possible distance data value null follows the distance data value one. The smallest possible distance data indicates that the target data value needs to be incremented, e.g. by the value of one.

(37) The compressed data 61 for the remaining laser pulses two to 50,000 are comparably encoded and compressed in this way.

(38) FIG. 5 shows schematically a method 70 for determining the scattered radiation spectrum in form of a flowchart. The same reference numerals are used for elements corresponding in function and/or structure to the elements of the exemplary embodiments of the previous figures.

(39) The method 70 uses the data generated with the device 1 or with the method 20 for determining the scattered radiation spectrum. The data represent discrete support points, which however cannot be readily assembled into the scattered radiation spectrum. The method 70 starts at a first method step 71. In method step 71, the method 20 is for example repeatedly performed to determine frequencies of F of scattered laser pulses P and quantities M of backscattered radiation S. A plurality of frequencies F and quantities M are forwarded to method steps 72 and 73 that follow the first method step 71. Frequencies F located within a frequency interval are combined in method step 72 to a combined frequency value F. In method step 73, quantities M of scattered radiation S associated with the frequencies F of the frequency interval are combined to a combined quantity value M.

(40) The values of the combined frequency F and the combined quantity M are determined or calculated so that these values approximate as closely as possible expected values of a theoretical spectrum. For example, the combined frequency value F and the combined quantity value M can be calculated by a linear approximation or higher-order approximation and matched to the expected theoretical spectrum. Other calculation methods can also be used. For example, the frequencies F and the quantities M can be weighted based on the number of frequencies F in the frequency interval, or the intensity of the laser pulses P can be weighted with the frequencies F in the frequency interval. It may also be necessary to determine and store at least the relative intensity of the laser pulse P in addition to the frequency F.

(41) If the frequencies F and the quantities M cannot be combined so that their combined values F, M approximate closely enough values of the expected theoretical spectrum, then this may indicate a malfunction of the device 1 or a fault in the theoretical model.

(42) FIG. 6 shows schematically a scattered radiation spectrum X determined with the device 1 and/or the method 20 and calculated with the method 70. The scattered radiation spectrum X is shown greatly simplified.

(43) Values of the frequencies F of the laser pulses P are depicted on an x-axis 80 (abscissa). The values of the quantity M and the number of scattered photons measured by the radiation sensor 3 are depicted on the y-axis 81 (ordinate). The scattered radiation spectrum X of FIG. 6 shows combined quantity values M only for twenty combined frequency values F of the laser pulses P. However, many thousands of combined frequency values F and associated quantity values M are often necessary for determining properties of the matter scattering the photons.

(44) The spacings and/or widths of the frequency intervals may be different from the distances shown. A realistic scattered radiation spectrum X covering all possible variants cannot be readily represented graphically, so that only the twenty frequency values F with the associated quantity values M are shown in the illustrated exemplary embodiment for clarity.

(45) The scattered radiation spectrum X has a maximum at a frequency f0. For example, many thousands or even millions of photons are counted over a period of, say, 24 hours at this frequency f0. Laser pulses P with the frequency f0 or an at least similar frequency F were scattered multiple times to receive that many scattered photons. Laser pulses P with frequencies F deviating from the frequency f0 were perhaps emitted at a similar rate. However, fewer photons were backscattered, which is evident from a smaller number of quantity values M or count rate for frequency values F of such laser pulses P, as depicted on the axis 81.

(46) An absolute frequency value can be assigned to the measured frequency f0 based on theoretical models. Furthermore, the scattered radiation spectrum X has a width B. The shape of the scattered radiation spectrum X can also be determined based on the measurement. The absolute value of the frequency f0, the width B of the scattered radiation spectrum X and the shape of the scattered radiation spectrum X allow conclusions about the properties of the matter scattering the laser pulses P. For example, the temperature of the gas scattering the laser pulses P in the mesosphere can be determined. Likewise, flow rates of gases in the mesosphere can be accurately determined to within 1 m/s or even 0.1 m/s. When suitable models are available, other properties of the matter scattering the laser pulses P can also be highly accurately determined with the method 20 according to the invention and the device 1 according to the invention.

LIST OF REFERENCE SYMBOLS

(47) 1 Device 2 Pulsed laser 3 Radiation sensor 4 Beam splitter 5, 5 Spectrum analyzer 6 Computing unit 8, 8 Storage device 9 Seeder laser 10 Reference laser 20 Method 21, 22, 23, 24 Method step 25, 26, 27, 28 Method step 29 Arrow 40 Compression method 41, 42 Method step 43, 44, 45 Method step 46 Arrow 60 Uncompressed data 61 Compressed data 70 Evaluation method 71 Start 72, 73 Combine 80 Abscissa 81 Ordinate A Excitation radiation B Width D Distance data F Frequency F Combined frequency value FM Data packet Fr Reference frequency Fs Frequency signal f0 Frequency F1, I1, Q1 Source data K Address data field K1 . . . K8000 Address data M Quantity M Aggregate quantity value Ms Quantity signal O2 Optical axis O4 Optical axis of the beam splitter P Laser pulse P First portion of the laser pulse P Second portion of the laser pulse Q Quality feature Qs Quality signal R Reference radiation S Scattered radiation S1, S2, S3 Source data field V Gas volume X Scattered radiation spectrum Z Target data fields