However, when mentioning the estimated cost ($1250M - $1410M) it is pointed out that 4400 antenna and receiver systems comprise the bulk of the expenses. It's possible to use smaller dishes, but that requires more receivers. So Dish@Home may have some merit along these lines.
Here are some basic numbers, quoted from the paper:
Quote:
We propose that the individual antenna elements be 12-m diameter fully steerable paraboloids, which give a one degree FOV at 20 cm and broadly minimize the cost curve. In order to meet the design goal of A/T = 20,000 m^2 / K and assuming system temperatures of 18 K, we need a total effective collecting area of 360,000 square meters or a geometric area equal to 500,000 m^2 for an aperture efficiency of 72%. Each antenna has a geometric area of 113 square meters so that 4400 antennas are required.”
I need to point out that run of the mill satellite receiver hardware is not of much use for radio astronomy, except for some class of incredibly brilliant sources.
The fundamental problem is signal to noise ratio - current dishes are less than one meter in diameter, and the receivers operate at ambient temperature. This works fine for getting a signal from a satellite with ~100 watts power at ~40,000 km distance.
Radio telescopes work with sources that are much fainter - there's a reason they use large dishes (larger signal) and cool their receivers with liquid helium (lower noise). I will suggest you look at some pages at the National Radio Astronomy Observatory (www.NRAO.edu) One example page would be What is Radio Astronomy. The NRAO project that is most like what we're discussing is the VLBA, which combines the signals from many dispersed antennas.
A note about the time and frequency requirements from one of those pages:
Quote:
The successful operation of a VLBI system requires that the tape recordings be synchronized within a few millionths of a second and that the local oscillator reference signal be stable to better than one part in a trillion.
I have quite a few friends who work at NRAO, and I might be able to get someone to pipe up here. I think the fundamental problem here is that the required electronics are really expensive, and don't really cost much more/less as you scale the size of the dish. The individual antennas for the Allen Telescope project are probably about the smallest/cheapest that would be of use, and I believe even then the cost is 10's of thousands of dollars each.
Thank you, GentleGiant, for putting things into perspective. Between the Science and the Technology, there's an awful lot to consider, and so your post is most helpful.
I noticed in the “Strawman Design” (above) that their estimates included projected technological improvements over the next couple of years (e.g., better memory, faster data transfers).
Since no one doubted the need to augment an owner's dish with additional hardware, might it be reasonable to expect that hardware devised for the SKA (e.g. the dish's feed) could also be used on a smaller (TV) dish – if the smaller dish is also augmented with a phased array of even smaller antennas (like those used in cell phones), which are placed (stationary) on the same structure as the TV dish (like 6 meters away from the TV dish), so as to give the TV dish a greater effective collecting area?
– if the smaller dish is also augmented with a phased array of even smaller antennas (like those used in cell phones)
Regarding the cell phone antennas, I realize that a smaller antenna seems rather counterintuitive (if not plain loony), but they can be designed so that they're “electrically large”.
Haven't had time to look for other references, or digest the info, yet.
Maybe a TV dish wouldn't even be needed? (The “RA@Home” participant could just mount a 12 meter diameter, pin-wheel patterned, phased array of hybrid MLAs to a pole or similar?)
smaller antenna seems rather counterintuitive (if not plain loony)
Not at all, try googling "ferrite loop". This technology has been around a long time.
Quote:
The “RA@Home” participant could just mount a 12 meter diameter, pin-wheel patterned, phased array
IMHO setting up and maintaining such an installation would much too complicated to be covered by the word "just". More so if the array is spherical and if it were planar it would need to be steerable or limited in the amount of sky it could cover.
This is all good stuff, but remember this is "Dish@Home"...
Cheap is the goal.... even just the name reeks of low-budget. And it's gotta be simple enough for your average beer-bellied back-yard bar-b-que grillin' bubba to be proud of.
Actually, I was imagining that the sheer number of TV dishes available and the geographic coverage might help mitigate the some the antenna sensitivity and steering issues. Make the earth one giant spherical array.
"No, I'm not a scientist... but I did stay at a Holiday Inn Express."
I was reading up on satlite tv, frequencies, bandwidth etc in hopes of improving the timebase. I did come up with an idea that may improve the local stability to approximately 2*10^-9. It would require that the dish return the local oscilator signal to the receiver along with the tv signal. If this could be augmented with a regional super receiver with an similar absolute stability then the term local could be replaced absolute with the help of digital signal processing.
Alas, I also found another problem. If we go back to my earlier estamate based on the 20 meter accuracy of GPS we are talking about a sample rate of 15*10^6 Hz. If we further assume 8 bit encoding (wildly optimistic IMHO) and futher assume we achive a 2 to 1 compression ratio (more wildly optimistic IMHO) we are talking about a data rate of 7.5*10^6 bytes per second, 450*10^6 bytes per minute and 27*10^9 bytes per hour (5.3 4.7 GB DVDs). Multiply this by even 100 receivers and you have a serious data management challange.
If the speculative improvement I suggested could be achived then the the data rates would all be multiplied by 100/3.
I was reading up on satlite tv, frequencies, bandwidth etc in hopes of improving the timebase. I did come up with an idea that may improve the local stability to approximately 2*10^-9. It would require that the dish return the local oscilator signal to the receiver along with the tv signal. If this could be augmented with a regional super receiver with an similar absolute stability then the term local could be replaced absolute with the help of digital signal processing
Sounds like a good idea. I was thinking also that the right combination of specific signals, both terrestrial and astronomical, which are already present in the data, could also provide the necessary reference.
Quote:
Quote:
The “RA@Home” participant could just mount a 12 meter diameter, pin-wheel patterned, phased array
IMHO setting up and maintaining such an installation would much too complicated to be covered by the word "just". More so if the array is spherical and if it were planar it would need to be steerable or limited in the amount of sky it could cover
If the array is horizontal planar, I was thinking that the output from each antenna element, along with the local array observable, could be stored in parallel, in holographic media. (Ref: Holographic memory from Wikipedia: ~8 x 10^9 terabytes in a cubic inch of media is the theoretical limit for the wavelength of the helium-neon laser.)
Quote:
Cheap is the goal.... even just the name reeks of low-budget. And it's gotta be simple enough for your average beer-bellied back-yard bar-b-que grillin' bubba to be proud of.
The phased array is the cheapest way I can think of to achieve the effective collecting area of the highly specialized 12 meter dish that GentleGiant pointed out is probably about the smallest you'd want to use. Also, that way even cable subscribers could participate – the sole criteria for participation would be roof/yard space. (As good as I think your idea is, barkster, I didn't know how to tell you I ain't got no dish:)
So this array would “basically” be an EM radiation data logger, consisting of a light-weight, all-weather composite structure that holds cheap (mostly!), but sophisticated, solid-state electrical and optical components which are configured in a cheap, but very sophisticated synchronized arrangement, and its front end would be a networked database manager (whose hardware can be a single IC).
Quote:
Multiply this by even 100 receivers and you have a serious data management challange.
Upon a request from a networked participant for data pertaining to a specific direction and time interval, the database manager crunches portions of the archived raw data and churns out a file for the participant to combine with similar files from all the other networked sites, hence producing the final image that matches the request.
I think dish@home as originally conceived could get around the signal to noise ratio problem GentleGiant brought up. You can in effect add the signals from all the different receivers. As noise introduced by each receiver is random and the contributions from different receivers will largely cancel out. To do this you would need weight each receivers signal. This could be done by extracting the embedded satellite signals and comparing these to the signal that is left after the broadcast signals are subtracted (I am assuming that receivers would be grouped based on the satellite and signal they are listening to).
As I understand it the individual satellite signals have a bandwidth of 30 MHz. To accurately capture that signal you would need a sampling frequency of 60 MHz and a time base stable to a thousand or more time the sampling frequency. (I know that this does not match what I said earlier: I forgot about the effects of integration frame jitter.) The individual TV signals are less than 3 MHz so you might be able to get away with a 6 MHz sample rate. But the more of the satellite signal you capture the better.
I found this link while investigating the NRAO site SARA
Looks to be about 14
)
Looks to be about 14 different design concept white papers at the SKA site. This one, “The Square Kilometer Array Preliminary Strawman Design Large N – Small D”, proposes using 4400 12-meter dishes (large Number, small Diameter), and these aren't your run-of-the-mill TV dish.
However, when mentioning the estimated cost ($1250M - $1410M) it is pointed out that 4400 antenna and receiver systems comprise the bulk of the expenses. It's possible to use smaller dishes, but that requires more receivers. So Dish@Home may have some merit along these lines.
Here are some basic numbers, quoted from the paper:
I need to point out that run
)
I need to point out that run of the mill satellite receiver hardware is not of much use for radio astronomy, except for some class of incredibly brilliant sources.
The fundamental problem is signal to noise ratio - current dishes are less than one meter in diameter, and the receivers operate at ambient temperature. This works fine for getting a signal from a satellite with ~100 watts power at ~40,000 km distance.
Radio telescopes work with sources that are much fainter - there's a reason they use large dishes (larger signal) and cool their receivers with liquid helium (lower noise). I will suggest you look at some pages at the National Radio Astronomy Observatory (www.NRAO.edu) One example page would be What is Radio Astronomy. The NRAO project that is most like what we're discussing is the VLBA, which combines the signals from many dispersed antennas.
A note about the time and frequency requirements from one of those pages:
I have quite a few friends who work at NRAO, and I might be able to get someone to pipe up here. I think the fundamental problem here is that the required electronics are really expensive, and don't really cost much more/less as you scale the size of the dish. The individual antennas for the Allen Telescope project are probably about the smallest/cheapest that would be of use, and I believe even then the cost is 10's of thousands of dollars each.
Thank you, GentleGiant, for
)
Thank you, GentleGiant, for putting things into perspective. Between the Science and the Technology, there's an awful lot to consider, and so your post is most helpful.
I noticed in the “Strawman Design” (above) that their estimates included projected technological improvements over the next couple of years (e.g., better memory, faster data transfers).
Since no one doubted the need to augment an owner's dish with additional hardware, might it be reasonable to expect that hardware devised for the SKA (e.g. the dish's feed) could also be used on a smaller (TV) dish – if the smaller dish is also augmented with a phased array of even smaller antennas (like those used in cell phones), which are placed (stationary) on the same structure as the TV dish (like 6 meters away from the TV dish), so as to give the TV dish a greater effective collecting area?
RE: – if the smaller dish
)
Regarding the cell phone antennas, I realize that a smaller antenna seems rather counterintuitive (if not plain loony), but they can be designed so that they're “electrically large”.
Here's a white paper I ran across on Meander Line Antenna technology: MLA Antennas – Physically Small, Electrically Large
Haven't had time to look for other references, or digest the info, yet.
Maybe a TV dish wouldn't even be needed? (The “RA@Home” participant could just mount a 12 meter diameter, pin-wheel patterned, phased array of hybrid MLAs to a pole or similar?)
RE: smaller antenna seems
)
Not at all, try googling "ferrite loop". This technology has been around a long time.
IMHO setting up and maintaining such an installation would much too complicated to be covered by the word "just". More so if the array is spherical and if it were planar it would need to be steerable or limited in the amount of sky it could cover.
This is all good stuff, but
)
This is all good stuff, but remember this is "Dish@Home"...
Cheap is the goal.... even just the name reeks of low-budget. And it's gotta be simple enough for your average beer-bellied back-yard bar-b-que grillin' bubba to be proud of.
Actually, I was imagining that the sheer number of TV dishes available and the geographic coverage might help mitigate the some the antenna sensitivity and steering issues. Make the earth one giant spherical array.
"No, I'm not a scientist... but I did stay at a Holiday Inn Express."
I was reading up on satlite
)
I was reading up on satlite tv, frequencies, bandwidth etc in hopes of improving the timebase. I did come up with an idea that may improve the local stability to approximately 2*10^-9. It would require that the dish return the local oscilator signal to the receiver along with the tv signal. If this could be augmented with a regional super receiver with an similar absolute stability then the term local could be replaced absolute with the help of digital signal processing.
Alas, I also found another problem. If we go back to my earlier estamate based on the 20 meter accuracy of GPS we are talking about a sample rate of 15*10^6 Hz. If we further assume 8 bit encoding (wildly optimistic IMHO) and futher assume we achive a 2 to 1 compression ratio (more wildly optimistic IMHO) we are talking about a data rate of 7.5*10^6 bytes per second, 450*10^6 bytes per minute and 27*10^9 bytes per hour (5.3 4.7 GB DVDs). Multiply this by even 100 receivers and you have a serious data management challange.
If the speculative improvement I suggested could be achived then the the data rates would all be multiplied by 100/3.
I knew there was a reason I
)
I knew there was a reason I had all these satellite dishes in my yard!
Built my 10ft commercial grade C-Band/KU band dish back in 1982 and in the 90's I added a directv and dish network dish.
(I used to own an electronics company and set up C and KU band systems and tested lnb's and helped out the cable company when their dish crashed )
I also built Epirb's back in the late 80's
I still get lots of backhaul feeds and other free stuff on the C-band/KU band dish.
I first started watching the "NASA Select" channel in 1982
All the feeds on the big dish used to be uncut and unedited.....some still are.
*Samson*
RE: I was reading up on
)
Sounds like a good idea. I was thinking also that the right combination of specific signals, both terrestrial and astronomical, which are already present in the data, could also provide the necessary reference.
If the array is horizontal planar, I was thinking that the output from each antenna element, along with the local array observable, could be stored in parallel, in holographic media. (Ref: Holographic memory from Wikipedia: ~8 x 10^9 terabytes in a cubic inch of media is the theoretical limit for the wavelength of the helium-neon laser.)
The phased array is the cheapest way I can think of to achieve the effective collecting area of the highly specialized 12 meter dish that GentleGiant pointed out is probably about the smallest you'd want to use. Also, that way even cable subscribers could participate – the sole criteria for participation would be roof/yard space. (As good as I think your idea is, barkster, I didn't know how to tell you I ain't got no dish:)
So this array would “basically” be an EM radiation data logger, consisting of a light-weight, all-weather composite structure that holds cheap (mostly!), but sophisticated, solid-state electrical and optical components which are configured in a cheap, but very sophisticated synchronized arrangement, and its front end would be a networked database manager (whose hardware can be a single IC).
Upon a request from a networked participant for data pertaining to a specific direction and time interval, the database manager crunches portions of the archived raw data and churns out a file for the participant to combine with similar files from all the other networked sites, hence producing the final image that matches the request.
How does that sound?
I think dish@home as
)
I think dish@home as originally conceived could get around the signal to noise ratio problem GentleGiant brought up. You can in effect add the signals from all the different receivers. As noise introduced by each receiver is random and the contributions from different receivers will largely cancel out. To do this you would need weight each receivers signal. This could be done by extracting the embedded satellite signals and comparing these to the signal that is left after the broadcast signals are subtracted (I am assuming that receivers would be grouped based on the satellite and signal they are listening to).
As I understand it the individual satellite signals have a bandwidth of 30 MHz. To accurately capture that signal you would need a sampling frequency of 60 MHz and a time base stable to a thousand or more time the sampling frequency. (I know that this does not match what I said earlier: I forgot about the effects of integration frame jitter.) The individual TV signals are less than 3 MHz so you might be able to get away with a 6 MHz sample rate. But the more of the satellite signal you capture the better.
I found this link while investigating the NRAO site SARA