meauring phase difference between two 2.4 GHz radio signals

Are there any micro controlers that can accept 2.4 GHz radio frequency inputs from an antenna and that can be programmed to measure the time delay between the 2 waves and accept mathematica or matlab programing for a voltage vs phase curve that outputs a voltage.

In a word … no.

I assume there’s some reason for not using the normal test equipment ?

You might marry an AD8302 to a micro or SoC to measure relative phase and report it.

I was going to suggest the AD8302 as well, but it needs 2 distinct inputs which you can’t get from a single antenna.

Once you split into two antenna systems, then you need a way to match the systems to have identical phase delay. Narda makes (made?) a small phase trimmer with SMA connectors that would allow for precise matching.

I believe you can also use a double balanced mixer as a phase detector for two signals close in frequency, as long as the IF port goes down to DC.

What’s the path length? Unless very short and line of sight, the phase difference could be from propagation as well as differing distances.

Dave Mueller:
I was going to suggest the AD8302 as well, but it needs 2 distinct inputs which you can’t get from a single antenna…

Well if the OP has only 1 antenna and there are 2 signals identical in frequency input to it, then all there is at that antenna's output is a single signal w/some "combined" phase. There's no way to separate them once added together. So I assume he's got 2 antennae.

Once again we’re given insufficient info to give a good answer. :frowning:

http://blockyourid.com/~gbpprorg/mil/in … index.html

This one does sound like a university physics/engineering project.

But until he/she states exactly what they want to accomplish we can only guess at solutions.

I just remembered it is possible, but probably not for this application. I worked on a distributed HDTV network in NYC. It was multiple transmitters all on the same frequency, everything GPS locked, called an SFN - single frequency network. Each transmitter had a unique ID embedded in its data stream. We had a box (sorry, can’t remember what make/model) that you connected to a receive antenna. It would analyze all of the signals and provide a graph of time delay between each stations. Since it was able to decode each transmit stream separately and measure the time difference between them, it could probably figure out phase as well since it’s related. I doubt the OP has a 6 or 7 figure budget for this problem though :slight_smile:

Sounds like something close to GPS-RTK?

Then there is this wild leap of the imagination:

accept mathematica or matlab programing

Yeah, obviously the OP doesn’t recognize the difference between a micro-controller (MCU) and an system on a chip (SoC) or a PC. I’ll opine that the OP has some grand intention in mind that he’s yet to tell us about (Ssssh, Top Secret). Still ignorance isn’t arrogance … and we may yet hear what he really wants to do.

Thank you so much for the replys, my first degree is in chemical engineering and I am working on a BS/MS in EE now. This is my first attempt at undertaking a real life project and not just doing math and matlab models. I have half a block diagram sketched out and will finish soon and post it. I have matlab code that allows me to do imedance matching and I would accomplish that by just verying the length of cable to the antenna (which is the simplest way, and just coil it up with a zip tie). There will be 2 antennas, one that is constantly transmitting from an osscilator from a DC power supply and the other will be reciving (the recieving antenna will be recieving the signal that I am broad casting plus any other signals out there. As I understand antenna operate on frequency so one freqeuncy will be from what I am broadcasting and the other from where ever (but the same frequency) so I guess that would only be one input with one frequency of 2 different phases, I appologize for that. Then the phase shifter will always be working to ensure that the phases are 90 out of phase but I have to have a way to meause the difference in order to send the proper signal to the phase shifter.

I am sure this is possible the issue is cost, I am now willing to spend 3 grand on a spectrum analyser so this one block may kill the project. The other angle is I could perhaps build a spectrum analyser but the entire device has to be somewhat small.

rppearso:
There will be 2 antennas, one that is constantly transmitting from an osscilator from a DC power supply and the other will be reciving (the recieving antenna will be recieving the signal that I am broad casting plus any other signals out there. As I understand antenna operate on frequency so one freqeuncy will be from what I am broadcasting and the other from where ever (but the same frequency) so I guess that would only be one input with one frequency of 2 different phases, I appologize for that. Then the phase shifter will always be working to ensure that the phases are 90 out of phase but I have to have a way to meause the difference in order to send the proper signal to the phase shifter.

Hmmm, what ? I’m afraid I’m not understanding your experimental setup. Nor do I understand what the purpose of your measurements is ? Perhaps a block diagram of the setup or some further description of what you’re trying to do will help.

Antennae have both a frequency range (over which they work well) and a spatial range (over which they will transmit and/or receive w/o much loss). Are you transmitting a signal (of some sort) from 1 antenna to another and then trying to measure … what, exactly … ??? Where does the phase shifter (mentioned above) figure into this all ? Are you trying to create a PLL ?? If so, why ? Is the transmitted signal just an un-modulated carrier or ?? Are you trying to detect it’s presence given that there’s other interfering signals, or demodulate some information impressed onto it ??

I have a gut sense you’re trying very hard to complicate something that, for an “RF person”, is actually pretty simple. If you can describe what you’re trying to do, instead of asking how to do it your way, you may get better answers.

If the two signals are exactly the same frequency, and both received by one antenna, it is impossible to separate them to measure their phase difference. They would appear as a single signal on a spectrum analyzer. The amplitude of the signal will depend on the strength of the two signal plus their relative phase. Also, a spectrum analyzer measures amplitude vs frequency, it can’t measure phase.

for low cost/home electronics, and at 2.4GHz, it’s impractically expensive to determine the relative phase difference between two antennas for one signal.

An interferometer for three antennas is more complex yet but can get you the direction of arrival if you have analog processing at a few nanoseconds speed. You also always have to have a way to remove the ambiguities of multiples of 180 or 360 degrees in phase difference.

To determine bearing angle at 2.4GHz, and from combinations of that, using time difference of arrival, one can get location. Companies like Ekahau make such for 2.4GHz, but they’re costly.

So what im hearing is that if there are 2 different antennas somewhere in space tranmitting at the same frequency but different phase it would be very difficult to measure that phase difference? I guess I should define low cost, I would be willing spend a few hundred bucks or maybe a bit more on this but the only reason I brought up cost is becasue when I started looking for 2.4 ghz range osillascopes they were like 30 grand. The company that makes them gave me the information and permission to build one if I wanted too but I will have to make time (lots of time). Another individual from a different company also makes a frequency analyzer but those are 3 grand which is pricy but not outside the relm of possibility.

In theory when those 2 signals hit the antenna they are in fact traveling at different phase down the “transmission line” into the circuit but it sounds like capturing that difference will be a task.

There has to be a way to do it, it cant be impossible. Or cost tens of thousands, espeically if the cirucit was built from scratch or broke down into multiple 40$ components. All the other parts to the project I have found for about 40$ a “block”. phase shifter, trangle wave generator, ossilator, amplifier, antennas, filters. I might have to have a custom made board to connect the modules but thats not too big of a deal as long as im mindful of matching but I dont think I need to worry about matching for super short runs in the radio frequency.

it looks like the AD8302 might do the trick. I could design the cirucit so that the tranmitting antenna does not turn on unitl the recieving antenna gets a signal from the band pass filter the AD 8302 could then measure the signal between the bandpass filter and the ossilator. It would not be active phase shifting but I could get decent signal cancelation from the first incoming signal. Once I started transmitting then the reciever antenna would be getting 2 signals unless I was able to design the antennas so that they did not interfere but that would be tough and likely impractical.

rppearso:
So what im hearing is that if there are 2 different antennas somewhere in space tranmitting at the same frequency but different phase it would be very difficult to measure that phase difference?

No, not at all. One way would be to have 2 receiving antennae, each pointed at one of the 2 antennae above, and each having sufficient directivity that only it's intended signal is received. The phase difference between these 2 separate channels would be simple to measure.

If you could switch (in time) between the 2 signals, you could measure the phase of 1 vs some reference signal and then then measure the other vs that reference signal and compute their relative phase difference.

If just one of the 2 signals had some modulation on it, so it could be distinguished from the other, then it would be simple to measure their relative phase difference. Depending on the angular separation of those 2 antennae above, you might be able to introduce that modulation at the receiving end w/an antenna (a variation of the 1’st method above).

rppearso:
In theory when those 2 signals hit the antenna they are in fact traveling at different phase down the “transmission line” into the circuit but it sounds like capturing that difference will be a task.

But you don't have 2 signals, you have 1 signal that's the vector addition of the prior 2 signals. Unless you know their relative amplitudes or their relative phases, you won't be able to separate them again.

In a simple analogy let’s say you had 2 DC current sources that you “pumped” into a single resistor. The resulting voltage across that resistor is the addition of those 2 currents x the resistance. Knowing only the voltage, how would you go about getting back the specifics of the 2 current sources ?

Make those current sources vary vs time at audio frequencies. You have Asin(Wt) and Bsin(Wt + phi) as inputs. The resulting voltage is also a sinewave with an amplitude (C) related to A and B and phi. It’s relative phase, let’s call it rho, is also related to A and B and phi. You can look up the math if you want but you have 3 unknowns; A, B and phi and only 1 (measured) known; C. You can’t solve that.

I’ve asked that you give us some more info re: your “experiment” and you keep resisting. Frankly this is now a waste of time so I’m done until I can do something useful.

Here is a link to the basic design I am trying to accomplish. At the bottom of the link there is an extended module for filtering based on different frequencies. The end goal is to create destructive interference patterns in relation to an incoming signal.

http://www.ece.gatech.edu/academic/cour … gn/jammer/

http://www.ece.gatech.edu/academic/cour … index.html

So the only thing different I am trying to add to this design is phase modulation so that I can get clean destructive interference.

If you feel this is a waste of time there is nothing compelling you to respond. This project has both practical application as well as intellectual interest to some.

I emailed mini-circuits to see if they have a circuit that would work.

OK, what you’re trying to do now is clearer. Let me restate what I think is the problem you’re trying to solve. You have a transmitter at point A which is received at point B. For the moment let’s simplify the signal transmitted to be a single constant frequency w/o modulation. You wish to transmit a signal from point C which will also be received at B and which will destructively add to signal A, ideally completely cancelling it. So far so good ??

To this end you thought you had to measure the phase (and presumably amplitude) of the 2 signals (A and C). In fact, if the situation is just this simple, you don’t have to. You simply have to be able to control the phasing and amplitude of signal C, something you needed to be able to do in any case.

Imagine that signal C was the proper frequency but wrong phase and slightly too low in amplitude. What is to composite signal (A + C) received ? It’s a sinewave. Now what happens as you vary the phase of C over a range of 0-360 degreees ? You’ll get a sinewave at B that will vary in phase and amplitude. The amplitude will be at it’s maximum when A and C are in phase and at a minimum when A and C are 180 deg out of phase. So to tune C’s phase, all you need to do is look for a minimum. Then, assuming the phase doesn’t drift, you adjust the amplitude of C to get the cancellation you desire.

In reality if you’re trying to cancel real life WiFi and BT signals, you’ll need control loops to do the above automatically. And that requires a slightly different approach. And as a practical matter you should be able to “pipe” a small portion of signal C (and uncancelled signal B) to the circuitry. Then you can do the relative phase and amplitude measurements and adjust C accordingly.

Of course given the wavelength at 2.4 GHz your cancellation, if it can be done, will only work over a small area.

Why would it only work over a small area? I have heard that some jammers can go out as far as 3 miles, is that not true, wouldnt that all be in your amplification module?

As far as scanning phases would you need a micro controler for that so that it automaticly cycles through different voltages to feed the phase shifter until the reciever antenna measures 0 or near 0 input?

If I cant get decent range then I might scrap the project. I would be running off of a 12 V DC source. I am also not sure if I would need to add an ossilator on the front end before the triangle wave generator since triangle wave generators are op amp modules I dont think a DC source world work. I have a 60 amp alternator but its still only 12 volts I believe so that will be the limit of my power amplification. I could do a phased array but thats a level of complication that I dont want to do which I would think would involve micro controlers to varry the current to each element to “steer” the beam.

rppearso:
Why would it only work over a small area? I have heard that some jammers can go out as far as 3 miles, is that not true, wouldnt that all be in your amplification module?

The method originally used to jam was to add noise in the bandwidth of the signal so as to degrade the SNR to the point of making the signal unusable. Depending on how much power you have and the victim's signal strength, this can work over "large" distances.

Your method is a lot more complicated. In order for you to achieve perfect cancellation you need your jamming signal to match the amplitude of the victim signal and be off by 180 deg at all points where the jamming is to be effective. Now you probably have some range of amplitude and phase matching that, while not perfectly cancelling, is “good enough” to work.

Now look at the radiation pattern of your antenna vs the victim’s, for signal strength and phase. Your dipole antenna will radiate a lot like a stone dropped into water, a circular pattern for the phase front and strength decreasing by 1/r^2. A cell phone near by might act the same. I’ll leave it to you to do the geometry and find the locus of points where the amplitudes are equal and the phases are 180 out. It’s only at small radii, perhaps a cm or 2 (the wavelength of a 2.45 GHz signal is a tad over 12 cm) about those points that jamming will be effective.

A cell phone tower will be “far” away. It’s radiation pattern is likely to be a planewave when it reaches your location. Now you’re trying to match phase of a planewave with a circular pattern. The latter resembles a planewave only when the transmitter is “far” away. Meaning your jammer will need to be far away and have great strength or use multiple emitters to create a non-circular pattern.