[Image appears of Benjamin Dix-Matthews wearing headphones and talking to the camera]
Benjamin Dix-Matthews: Hi. My name is Benjamin Dix-Matthews. I’d like to start off by thanking you for awarding me this scholarship and I have to apologise that I’m not able to make this speech in person as I am already over in Toulouse getting started on some experiments. It’s a huge honour to be awarded with this scholarship and it makes me very proud to have been awarded it. So far, this scholarship has allowed me to do one week of research with CNES, the French Space Agency and already it’s been the highlight of my PhD. So, thank you very much and I’ll get started on talking about what my PhD is actually about.
[Image changes to show a black screen showing the University of Western Australia, The International Centre for Radio Astronomy Research, and EQUS logos and text appears: 2020 CSIRO Alumni Scholarship in Physics, Coherent Free-Space Frequency Transfer by Benjamin Dix-Matthews]
So, my PhD is related to coherent free-space frequency transfer. Also, I’m a student of UWA and ICRAR and EQUS.
[Image changes to show a slide showing a photograph of a space station and background information to Benjamin’s PhD and text appears: Background, Many scientific and commercial applications can benefit from improvements in the distribution of stable frequencies, Radio Astronomy, Geodesy and Doppler ranging, Tests of fundamental Physics, High speed optical communications, Research focused on transfer through optical fibre, AIM – develop a system capable of similar stabilized transfer through free space]
So, a bit of background. There are many scientific and commercial applications that can benefit from improvements in the distribution of stabilised optical frequencies. These include the radio astronomy, Geodesy and Doppler ranging, tests of fundamental physics and high-speed communications. Up until now research has been focused on optical fibre implementations and the aim of my PhD is to develop systems capable of similar types of stabilised transfer but through free space.
[Image changes to show a photo of a symbol on the right of the screen and a text heading and text appears: Atmospheric turbulence, Fluctuations in refractive index, Alter the propagation of the optical wave, Zeroth-order “piston mode”, Causes variations in the optical time-of-flight, Introduces phase-noise, First-order “beam-wander”, Spatial variations in the beam centroid position, Leads to intensity fluctuations in the received optical signal]
Now, as soon as you jump over into free space you encounter a number of other challenges and in this talk I’m going to focus on the challenges associated with atmospheric turbulence. So, in the atmosphere there are constantly fluctuating refractive index which are caused by variations in air temperature. As our optical beam propagates through these fluctuating refractive indexes it distorts the propagation of our optical wave. This Zeroth-order distortion or piston mode effect causes variations in the optical time-of-flight and this introduces phase noise on to our optical signal decreasing it’s stability. A First-order effect is beam wander which is spatial variations in the beam’s centroid. What this does is lead to intensity fluctuations as we try to couple our incoming beam into our optical fibre. There are also higher order effects that lead to scintillation but dealing with these first two effects is the most important one that I’m going to talk about here.
[Image changes to show a photograph of the Phase Noise Stabilisation System and the Tip Tilt Stabilisation System and text heading and text appears: Experimental system – Overview, Phase Noise Stabilisation system compensates the Zeroth-order effects, Tip-Tilt Stabilisation system compensates the First-order effects]
So, in order to overcome those two effects we’ve developed an experimental system that’s comprised of two stabilisation sub systems. We have the Phase Noise Stabilisation system that compensates the Zeroth-order effect, that piston effect and we a Tip-Tilt Stabilisation system that compensates for that First-order beam wandering effect.
[Image changes to show the Phase Noise system inset and a diagram appears above to show how it works and a text heading and text appears: Experimental system, Phase Noise Stabilisation, Portion of receiver light is reflected, used at local site to actively suppress the Zeroth-order phase-noise]
So, our Phase Noise system is down here and I’ve drawn a block diagram to show how it actually works. So, our input signal, our input optical reference signal comes from our input laser and we transmit this through an acoustic optic modulator which causes a frequency shift in one direction and we transmit it over our free-space link to our remote terminal where we then pass it on to the user.
Now, the Phase Stabilisation system works by reflecting a portion of the light back over to that link to our local site. Now over the, at the time of flight, as the optical time of flight scales the refracted, the phase noise in this link is essentially reciprocal. So, what that means is that the phase noise stream of forward transmission is going to be approximately equal to the phase noise encountered during the reverse transmission.
So, we can use that in order to stabilise against the phase noise in the forward transmission. So, our incoming light will be mea… we perform a heterodyne measurement of our light coming back into our local site and we beat that against a phototype. After down mixing we get a lower frequency signal that contains the information, that contains information about the phase noise encountered during the transmission and we use this through a servo route in order to actuate this transmission AOM in order to actively suppress the phase noise encountered during the forward transmission. What that means is that the signal received by the end user will have any of the phase noise contributions encountered during the transmission actively suppressed. So, the stability of the signal here will be approximately equal to the stability of the signal here.
[Image changes to show a photo of the Tip-Tilt Stabilisation system and a diagram of an Optical Terminal appears below and a text heading and text appears: Experimental system, Tip-Tilt Stabilisation, Quad Photo Detector to measure the position of the incoming optical beam, Two piezo electric actuators are then used to steer a flat mirror, Suppresses intensity fluctuations caused by First order beam wander]
We also have a tip-tilt stabilisation system that’s used to correct for that first order beam wandering. The way that works down here in this block diagram is we come from our fibre and couple into free space using a collimator. We pass through beam splits, a Gaussian beam expander and reflect off this active mirror into free space. The actual Tip-Tilt system works on incoming light, however. So, we get our light from our link reflected off this active tip-tilt mirror, through our Gaussian beam expander and then it arrives at this beam splitter. Half the light is then sent on to the collimator which then couples it back into the fibre and the other half is received by this quad photo detector.
This quad photo detector is able to measure the spatial variations in the centrode of that beam and those spatial variations are then sent to, through a servo loop, and used to control the two Piezo actuators on our tip-tilt mirror here. And what that does is actively suppress the movement of the beam on this QPD and because this is the same light as is being received by the collimator it also suppresses the spatial deviations of the beam at the collimator and it reduces the, suppresses the intensity in fluctuations.
[Image of a photo of the optics appears in the bottom right corner of the slide]
I also have a, an image of the actual optics that we use so you can see how they work here. We have our light coming into our terminal being reflected off our tip-tilt mirror, through this Gaussian beam expander to our beam splitter where half of it is coupled into fibre and the other half is sent over to this quad photo detector. The quad photo detector measures the variations in the beam and then uses these two high-voltage amplifiers in order to control the two piezo actuators on the tip-tilt mirror suppressing those fluctuations.
[Image changes to show three diagrams showing the system on the Local site, the system on the Remote site and the Optical Terminal and a text heading and text appears: Experimental system, Performance Measure, Performance is obtained through an out of loop measurement between the “Input Laser” and the signal sent “To User”, Noise added during the out of loop measurement degrades the performance measurement]
So, when we’ve built this stabilisation system that comprises two stabilisation sub-systems we want to be able to measure the performance of our transfer. So, in order to measure the performance of a transfer we have to see how much our input signal has degraded by the point we send it to our end user. And in order to do this we need to perform an out of loop measurement between this input signal and the signal being received by the user. Any noise added into the, into either of those signals during that measurement process will make it appear like the transfer was degrading the performance of the system even if it wasn’t. So, we want to make sure that we’re minimising the noise added in during that out of loop measurement.
[Image changes to show the Local site diagram and the Remote site diagram with arrows moving between the two and then beams converging in a mirror on the right of the diagram and a text heading and text appears: Experimental system – Performance measure, Out of loop measurement achieved using a folded link, The “Input Laser” and the signal sent “To User” are physically close together, Minimises the noise added by during the out of loop measurement, This is no longer a practical optical transfer link]
The way that people traditionally do this is by sending the free space component of the optical beam over a folded link and what that means, allows is the input laser and the signal sent to the user to be located physically close to each other. And that means that there’s, it minimises the amount of noise added during that out of loop measurement. It’s important to note however that this is no longer a practical optical transfer link because our local site and our remote site are located in the same location.
[Image changes to show a Time Domain Stability line graph and a Frequency Domain Stability line graph below the text heading: Experimental system – Preliminary results]
Nevertheless, we built this system with a folded link and a flat mirror at UWA and these are the results we got. So, for our unstabilised system we were up here and when we turned on our stabilisation system we see our fractional frequency stability improve by about three orders of magnitude which means that the stabilisation system is indeed working and improving the quality of the signal that’s been sent to the user. Over here we have a frequency domain stability metric, our phase noise measured in dBc/Hz, and we can see that we’re getting around 70 decibels of improvement at 1Hz. Also, interesting is to note that in this unstabilised plot here the vast majority of the phase noise is at very low frequencies. So, the low sort of 100 or 1000 Hz.
[Image changes to show a diagram showing the local site with a transmitter and the remote site with a receiver joined overhead by a free space link and beneath by a stabilised fibre link and a text heading appears: CNES Collaboration – Overview]
Now onto the research that this scholarship has actually paid for, is some collaborative work with CNES, the French Space Agency. So, CNES is interested in free-space communications as well and we developed an experiment that utilises some of the unique advantages that both we at the UWA Astrophotonics Group have and that CNES has at their site here in Toulouse. So, they have this interesting, useful, architecture where they have two buildings separated by two… 300 metres with a fibre going from the roof of each, under the ground, and then up to the roof of the other. And they also have line of sight between these two buildings.
So, what we’ve proposed is to do a true point to point optical transfer and then use this fibre to do that out of loop measurement to actually measure the performance. So, we have the remote, the local part of our phase stabilisation system located at our local site. We then pass it to our tip-tilt terminal and launch it over the 300 metre free space link. It’s then received over here and passed to our, the remote part of our phase stabilisation, phase noise stabilisation terminal here.
The plan is to build, what we’ve done is build a second fibre based phase stabilisation system that transfers our optical reference signal from our local site, through this fibre, to our remote site. Because fibre based stabilisation systems have had a lot more research done on them and they are a very high quality transfer we’re expecting the transfer of our, the stabilised transfer of our signal through the free-space link to be, to have more noise, to be what’s limiting our out-of-loop measurement here. And then we’re going to, form an out-of-loop measurement between the fibre based transfer and the free-space transfer at our remote site. So, I got here about a week ago. So, we have already started setting up this equipment.
[Image changes to show a photograph of Benjamin setting up the equipment on the left of the screen and the Luke Skywalker optical terminal on the right of the screen and a text heading appears: CNES Collaboration – Local site]
So, on this day, this picture on the left you can see me setting up the optical terminal at our local site and this diag… this picture on the right is another image of that optical terminal which we’ve called Luke Skywalker. It’s called Luke Skywalker because the guide laser is green. The guide laser for the terminal on the other side is red so naturally we’ve dubbed that optical terminal Darth Vader.
[Image shows a red pointer circling a building on the city in the background of the photo]
So, our Luke Skywalker terminal is located on our local site and it’s used to transmit our beam over this 300 metre link to our remote site which is in our telescope dome on this building over here. Our phase stabilisation system is located inside this building and we have a fibre coming out to the terminal. The phase stabilisation system hasn’t actually changed at all so I didn’t bother taking another photo of it.
[Image changes to show a photo of a telescope dome on the left of the screen and a tip-tilt mirror on the right of the screen]
At our remote site we have a nice telescope dome which is shown in this picture on the right. Our Darth Vader terminal here receives the light from our Luke Skywalker terminal at the local site, sends the signal through this optical fibre to the remote part of our phase stabilisation system which is located here. The fibre link comes up behind this rack here and we perform our out-of-loop measurement between the stabilised free space and the stabilised fibre measurement inside this opt… inside this telescope dome.
On the right here, we have a picture of the tip-tilt mirror located in our Darth Vader terminal at the front and you can see over here the small green dot. That is the guide laser, the green guide laser from our Luke Skywalker terminal at the local site 300 metres away. So, the fact that you can see that green dot means that the link is relatively well aligned.
[Image changes to show a Time Domain Stability line graph and a Frequency Domain Stability line graph below the text heading: CNES Collaboration – Preliminary Results]
Now, this set-up took a while but we still did manage to get a few preliminary results during the first week which I have on the next slide, oops, located here. So, this is our unstabilised deviation and we can see that we’re still getting that three order of magnitude improvement when we turn on our stabilisation system. So, despite it being a slightly further distance at CNES, 300 metres as opposed to at UWA we have only a 500 metre link. The performance of these are actually still very similar. I’d like to bring your attention down to this trace here.
What that trace represents is the most stable optical transmission ever performed in a point to point link. So, that is, this is currently the world record which is pretty impressive. It’s actually more stable than what we were expecting when we came over. So, already that is an incredible result. It only goes up to integration times of about 80 seconds due to technical reasons.
Next week we’re hoping to extend this out a little bit further so that we can see how it integrates down at longer time spent. In terms of our frequency domain stability metric we’re seeing about a 60 decibel improvement at 1Hz and what’s interesting is you can see that almost the vast, vast majority of the phase noise from the atmosphere, so in our unstabilised system is below about 100 Hz. So, that is useful information in itself and also this plot is, is some very nice results.
So, next week we’re hoping to extend these plots out a little bit more and see how this phase noise of the unstabilised version changes due to different atmospheric conditions. So, that’s all that we’ve done so far. Hopefully next week we’ll get some really nice new results that I can report back on.
[Image changes to show a photograph of Benjamin smiling at the camera outside the CNES Space and Rocketry Museum]
And I’d like to leave you with this picture of me on the weekend which is in CNES they have a Space and Rocketry Museum. So, this is me at that Space and Rocketry Museum on Saturday, which just happened to be my birthday. So, I’d like to thank you for allowing me, giving me the scholarship and allowing me to do such fun science and be in such a fun place on my birthday and just in general. It’s been a fantastic experience during the first week and I’m sure that next week is going to be just as amazing. So, thanks for listening.