Phase noise compensation system, and method

Abstract

A system for compensating for phase noise, with particular application in lidar, includes a compensation interferometer that receives a signal from a source, and splits it into a first and second path, with a path length difference Δτ between them. Typically the path length is significantly less than that of the return distance to a target. The output of the compensation interferometer, which consists of phase noise generated in time Δτ is vectorially summed during a time similar to a signal flight time to a target, and the result used to reduce phase noise present on measurements of a target. It further includes means for selecting Δτ such that competing noise elements are reduced or optimised.

Claims

1. A system comprising: a) a signal generator for providing a signal, b) a splitter for splitting the signal from the signal generator and directing it to a measurement path and a compensation path, wherein: i) the measurement path comprises a measurement interferometer having a splitter for splitting the signal into a first and a second part, the measurement interferometer being arranged to transmit the first part to a remote target, and to receive reflections therefrom, and to combine the received reflections with the second part in a measurement interferometer mixer, and ii) the compensation path comprises a compensation interferometer arranged to split its input signal into a first, delayed path, and a second, undelayed path, and to combine the first and second paths in a compensation interferometer mixer to produce an output bearing information pertaining to a phase difference between the two paths, the relative delay between the first and second paths being a predetermined time Δτ; c) a processor configured to calculate a phase compensation figure by digitally measuring a phase change across n successive passes through the compensation interferometer over a time nΔτ, where nΔτ is equal to n times the predetermined time Δτ, and subtracting the calculated phase compensation figure from the output of the measurement interferometer, wherein the system is arranged to calculate a range L/2 to the remote target based upon the compensated output from the measurement interferometer, said range giving a time delay τ to the received reflections as compared to the second part, and wherein the phase compensation figure is calculated under the condition that |τ−nΔτ|≤Δτ/2.

2. A system as claimed in claim 1 wherein the system is arranged to select a value for n a based on an estimate of the range to the target.

3. A system as claimed in claim 1 wherein the delay time Δτ of the compensation interferometer is chosen based upon a combination of errors associated with multiple measurement of the phase error from the compensation path, and the size of Δτ/2.

4. A system as claimed in claim 1 wherein the values of n and Δτ are chosen to reduce residual phase noise following the processing carried out in step (c).

5. A system as claimed in claim 4 wherein the length of delay coil ΔL in the compensation interferometer, and hence the values of n and Δτ are chosen, to produce a minimum value for σ.sub.phase in the equation: ${\sigma }_{\mathrm{phase}}=\sqrt{{\left({\left[{\int }_{f_{\mathrm{Low}}}^{f_{\mathrm{High}}}{.Math.\Phi \left(f\right).Math.}^2\mathrm{df}\right]}^{1/2}Y\right)}^2+\frac{L}{\Delta L}{\sigma }_{\mathrm{meas}}^2}$ where σ.sub.phase is the RMS residual phase error, Φ(f) is the laser phase noise spectrum expressed in radians per root Hertz referred to 1 m optical path difference which is integrated over the relevant frequency band f.sub.Low.fwdarw.f.sub.High, ΔL is the optical path length of the delay coil (with time equivalent Δτ), Y represents the maximum uncompensated path length, (with a maximum value of ΔL/2), a σ.sub.meas is the RMS phase error associated with a single differential phase measurement in the compensator interferometer, and L is the target return optical path length.

6. A system as claimed in claim 1 wherein the compensation interferometer is arranged to have a delay time Δτ of less than 10% of the flight time of the signal traversing twice an expected in-use target range.

7. A system as claimed in claim 1 wherein the compensation interferometer is arranged to have a delay time Δτ of less than 1% of the flight time of the measurement signal traversing twice an expected in-use target range.

8. A system as claimed in claim 1 wherein the compensation interferometer mixer provides a complex output allowing in-phase (I) and quadrature (Q) signals to be extracted.

9. A system as claimed in claim 1 wherein the measurement interferometer mixer provides a complex output allowing in-phase (I) and quadrature (Q) signals to be extracted.

10. A system as claimed in claim 1 wherein the measurement path contains a modulator for modulating the signal before it is transmitted to the target.

11. A system as claimed in claim 1 wherein the signal generator is a laser.

12. A system as claimed in claim 11, wherein the measurement path contains a modulator for modulating the signal before it is transmitted to the target, wherein the modulator is an acousto-optic modulator.

13. A system as claimed in claim 1 wherein the signal generator is a radio frequency signal generator.

14. A system as claimed in claim 1 wherein the delay path of the compensation interferometer comprises an optical fibre.

15. A system as claimed in claim 11 wherein the delay path comprises of a coaxial cable.

16. A system as claimed in claim 1 wherein the system is a LIDAR system or a radar system.

17. A system as claimed in claim 16 wherein the system is arranged to have a slant range of at least 1 km.

18. A system as claimed in claim 1 wherein values of phase noise compensation are calculated for different target ranges, and are used to correct the output of the measurement interferometer at different ranges.

19. A method for compensating for phase noise in a LIDAR or radar, comprising the steps of: i) generating a signal; ii) splitting the signal into a measurement signal and a compensation signal and directing the measurement signal to a measurement path and directing the compensation signal to a compensation path; iii) transmitting a first part of the measurement signal in the measurement path to a remote target, and mixing returns from the remote target with a second part of the measurement signal in a measurement interferometer mixer to down-convert the returns; iv) passing the compensation signal into a compensation interferometer arranged to split its input signal into a first, delayed path, and a second, undelayed path, wherein a relative delay between the first and second paths being a predetermined time Δτ; v) calculating a phase compensation figure by measuring a phase change across n successive passes through the compensation interferometer over a time nΔτ, where nΔτ is equal to n times the predetermined time Δτ; vi) subtracting the calculated phase compensation figure from the down-converted output of the measurement interferometer mixer of step (iii); and vii) calculating a range L/2 to the target based upon the compensated output from the measurement interferometer, wherein said range giving a time delay τ to the received reflections as compared to the second part, and wherein the phase compensation figure is calculated under the condition that |τ−nΔτ|≤Δτ/2.

Description

(1) The invention will now be explained in more detail, by way of example only, with reference to the following figures:

(2) FIG. 1 shows a high level block diagram of a system according to an embodiment of the present invention; and

(3) FIG. 2 diagrammatically illustrates an example phase error signal, being tracked by multiple samples.

(4) FIG. 1 shows a simplified block diagram of a system according to one embodiment of the present invention. The system is a long range LIDAR. A laser 1 feeds a splitter 2, which in turn supplies EM energy from the laser to a measurement interferometer 3 and a compensation interferometer 4.

(5) The measurement interferometer 4 has a splitter 5, for splitting the laser light into two paths—a transmission path and a reference path. The transmission path comprises an acousto-optic modulator 6, a circulator 7, and transmission/reception optics 8, which transmit EM energy to a target scene 9 and receive reflections back therefrom. The returning energy passes through the circulator to an output path into a mixer 10, which also has as an input the energy from the splitter 5 that is directed to the reference path. Thus, the mixer 10 mixes energy that is returned from a distant target, and has a delay τ with respect to energy from the laser that has no effective delay applied thereto. The output of the mixer includes phase noise due to the incoherence of the laser that accumulates over the time τ. This output is digitised, in digitiser 11, and sent to processor 12.

(6) The compensation interferometer 4 comprises of a splitter 13 that splits its input EM energy into two paths, with the first path going straight to one input of a phase sensitive detector 14, and the other input feeding a fibre delay line 15 of 20 m length (generating a delay Δτ), before going to a second input of the phase sensitive detector. The phase sensitive detector comprises a 3×3 splitter, with the 3 outputs feeding individual detectors, the outputs of which are digitised in digitiser 16. These measurements can be resolved to provide the differential phase noise that occurred in the time period Δτ. An algebraic summation of successive differential phase measurements provides a measure of the differential phase noise over the equivalent summed length (in distance or time, as appropriate). The processor is able to calculate the differential phase noise values over any desired length, and to then subtract this from the phase information derived from mixer 10.

(7) This embodiment is intended for implementation in a LIDAR having a slant range of approximately 30 km. Thus, it will be seen that the 20 m delay line is significantly smaller than one that would conventionally be used to balance a 60 km free-space time delay.

(8) The compensation interferometer is, in this embodiment, implemented in polarisation maintaining fibre. It is beneficial to have means, such as this, to control polarisation, as it helps to maintain an adequate and stable interference signal.

(9) FIG. 2 shows a simulated example of absolute phase noise, as may be present on a laser, and any other component up-stream of splitter 2 that might impart phase noise. The horizontal axis represents time, and the vertical axis represents phase instability over time. It can be seen that the phase noise varies randomly, typically exhibiting ‘1/f’ or ‘flicker’ noise characteristics. Four measurements of differential phase noise over optical path difference Δτ are marked on the graph, spanning t.sub.0.fwdarw.t.sub.0+Δτ, t.sub.0+Δτ.fwdarw.t.sub.0+2Δτ etc.

(10) Each measurement gives the differential phase change since the previous one. The algebraic sum of the four differential phases Δϕ.sub.1, Δϕ.sub.2, Δϕ.sub.3 and Δϕ.sub.4 gives the total phase change Δϕ.sub.n corresponding to differential path length 4Δτ. In practice, the range to a target (or more accurately, the relative delay in the measurement interferometer caused by the distance of the target, and associated systematic path differences in the interferometer) is unlikely to be an exact multiple of Δτ. Assuming, for example's sake, that the two-way time of flight is slightly greater than 4Δτ, then the unknown differential phase noise, taken from time t.sub.0 is as indicated by the question mark. Subtraction of the value Δϕ.sub.n from the output of the measurement interferometer therefore leaves this error still present on the signal. This error however is small, and as shown above is no greater than the phase noise due to a delay of Δτ/2 in the measurement interferometer, and can be controlled within limits by appropriate choice of the size Δτ and the inherent phase noise of the laser.

(11) The uncompensated phase error cannot be discounted. Neither can the phase noise introduced through measurement error on the compensation value. It will be appreciated that the smaller the value of Δτ the smaller the uncompensated differential path length, but the higher the value of n results in higher overall measurement noise, and vice-versa.

(12) The residual phase noise can be written as the quadrature sum of these two independent sources of error:

(13) ${\sigma }_{\mathrm{phase}}=\sqrt{{\left({\left[{\int }_{f_{\mathrm{Low}}}^{f_{\mathrm{High}}}{.Math.\Phi \left(f\right).Math.}^2\mathrm{df}\right]}^{1/2}Y\right)}^2+\frac{L}{\Delta L}{\sigma }_{\mathrm{meas}}^2}$
where σ.sub.phase is the RMS residual phase error, Φ(f) is the laser phase noise spectrum expressed in radians per root Hertz referred to 1 m optical path difference which is integrated over the relevant frequency band f.sub.Low.fwdarw.f.sub.High, ΔL is the optical path length of the delay coil (with time equivalent Δτ), Y represents the maximum uncompensated path length, a σ.sub.meas is the RMS phase error associated with a single differential phase measurement in the compensator interferometer, and L is the target return optical path length (i.e. to the target and back). In the worst case the value of Y will be equal to ΔL/2, but will in general be less than this. An optimum delay length ΔL can be found by solving this expression for minimum residual phase noise.

(14) The invention has particular utility in radar or LIDAR systems, although it will be appreciated that it may be used in other areas, such as fibre sensors, or other areas where a compact size and coherent operation at long ranges (compared to the coherence length of the signal generator source) are desired. This includes some types of radar, for example, as described above.

(15) It should also be noted that, whilst the description above has been for a single discrete target, the technique also applies to multiple discrete and continuous down-range scatterers. To this end, it will be appreciated that appropriately configured embodiments of the invention (i.e. those that store the measurements from the compensation and measurement interferometers in suitable longer term memory) allow for post-processing of the returns, to apply the phase noise reduction at multiple ranges if required. Some embodiments may be configured to implement this in real-time also, of course.