How do they process high frequency signals, much high than the available uC clock speeds ?

[BradleyEnglish]

For example ADC's that can process at 200M hz ?
How do they measure radio signals that can get up to 300G hz ?

Fast microcontrollers can only work at around ~150Mhz

Regards,
Bradley.

 

[JimB]

A lot depends on what you mean by "process".

The simple answer is that you dont use ADCs and microprocessors at high RF frequencies.

Can you be a bit more specific with your question?
What do you want to process or measure?

JimB

 

[BradleyEnglish]

A lot depends on what you mean by "process".

The simple answer is that you dont use ADCs and microprocessors at high RF frequencies.

Can you be a bit more specific with your question?
What do you want to process or measure?

JimB

I want to process high frequencies such as gigahertz radio frequencies.
& audio frequencies too.

Another part of my question is, Could i use multiple microcontrollers at high clock speeds through interlinking them / sharing workload etc?
I do know that the data travel between microcontrollers is a lot slower compared to the speed that data comes from their internal registers, so i suppose the microcontroller outputs can output one after the other in a manner similar to time division multiplexing so that the slow data transfer problem wont exist.

 

[BradleyEnglish]

I also posted this on 4chans / diy and got a great answer.

To be able to process high data rates with a slower uC i can use a FIFO buffer which collects the data and then sends it in a bulk to the uC for the data to be processed. Making the process much more efficient.

I may also multiplex several uC's together to share workload and output in a multiplexed order.

The FIFO option seems cheapest and simplest.

However Im still unsure as to how high radio frequencies of around 300 GHz are measured.

 

[kubeek]

What exatly do you want to measure at those radio frequencies? If you want to measure the frequency, then either you use a counter driectly, which will measure the number of zero crossings during a given interval. If the frequency is too high then you mix it in the analog domain with a known reference and then measure the frequency of the resulting mixed frequency, which will be shifted down to a frequency that you can measure with a normal frequency counter.

There will be a lot of other parameters that you can measure on a given signal like phase, jitter, wave shape etc., so please be more concrete about what you actually want to know.

 

[ronsimpson]

how high radio frequencies of around 300 GHz are measured.

My frequency counter uses "flip flops" to divide the frequency down to a lower frequency so it can be counted. I have a optional divide by 2 and divide by10 that goes on the front end of the frequency counter. I just looked and found parts that will work up to 40ghz.

Next; you can subtract to get to a lower frequency. I built many years ago a radio that receives 145mhz.
I have a 145mhz signal and a 155.7mhz oscillator. The two signals are mixed in a "mixer". The output is (145hmz and 155.7mhz and 300.7mhz and 10.7mhz) ( I could have used 134.7mhz oscillator)
Now I filter out all frequencies and keep the 10.7mhz.
Then because the signal is still to high frequency to work with, I use another mixer and oscillator. This time I subtract to get down to 455khz using a 10.245mhz oscillaotr.
Filter again to get only the 455khz.
Here is where the audio is pulled out of the 455khz and sent to the speaker.

The signal145mhz-134.7mhz-10.245mhz=455khz, If the signal moves to 145.001mhz then the low frequency will move the same amount to 456khz.

 

[NorthGuy]

Best ADCs can go up to 4GHz, but they get really expensive

 

[BradleyEnglish]

What exatly do you want to measure at those radio frequencies? If you want to measure the frequency, then either you use a counter driectly, which will measure the number of zero crossings during a given interval. If the frequency is too high then you mix it in the analog domain with a known reference and then measure the frequency of the resulting mixed frequency, which will be shifted down to a frequency that you can measure with a normal frequency counter.

There will be a lot of other parameters that you can measure on a given signal like phase, jitter, wave shape etc., so please be more concrete about what you actually want to know.

I want to be able to process high radio frequencies so I need to measure the wave somehow whilst I program a radio transceiver. As far as I know my uC clock speed will have to be very high in order to process the signal, and in order to measure the signal i will also need something fast. I learned to use oscilloscopes but now that im dealing with a high RF i dont know what is usually used by those involved in high radio frequency signals.

 

[Nigel Goodwin]

Basically you don't use processors for RF, what exactly are you trying to build?.

You also aren't making it clear what you're trying to measure, or why?.

 

[JimB]

I want to be able to process high radio frequencies so I need to measure the wave somehow whilst I program a radio transceiver. As far as I know my uC clock speed will have to be very high in order to process the signal, and in order to measure the signal i will also need something fast. I learned to use oscilloscopes but now that im dealing with a high RF i dont know what is usually used by those involved in high radio frequency signals.

Please excuse my intemperate outburst, but this is starting to take on the feeling of a trolling post.
You are just using words here and those words do not make much sense.

PROCESS
You keep using that word.
What process are you trying to perform on this RF signal?
Demodulation perhaps? Please do tell us, we have asked several times and we are no nearer to an answer.

measure the wave somehow whilst I program a radio transceiver.
What is it about the wave you want to measure?
And what has that to do with programming a radio transceiver?

I also posted this on 4chans / diy and got a great answer.

Then why are you asking here?
Ask the experts at 4chans.

JimB

 

[BradleyEnglish]

Wow mate cool it mate you're drilling me man.

Sorry if I wasn't clear.

I'm trying to try send some data via RF from one uC (a data processor) to another uC (data processor). Now the reason I want to measure it is because I will do a lot of tweaking, trying various different modulation techniques and frequencies. I'm learning at the same time so to be able to measure the RF waves would help a lot.

 

[Nigel Goodwin]

Wow mate cool it mate you're drilling me man.

Sorry if I wasn't clear.

I'm trying to try send some data via RF from one uC (a data processor) to another uC (data processor). Now the reason I want to measure it is because I will do a lot of tweaking, trying various different modulation techniques and frequencies. I'm learning at the same time so to be able to measure the RF waves would help a lot.

Well you can't, and if you have to ask then you're not going to understand it anyway

To do what you want to do you simply use licence free radio modules, which are freely and cheaply available - and legal as well, which what you're suggesting wouldn't be (as well as impossible).

 

[ronsimpson]

Transceiver:
https://www.sparkfun.com/products/12031
https://www.sparkfun.com/products/12770

Transmitter: and Receiver:
https://www.sparkfun.com/products/10534
https://www.sparkfun.com/products/10532

 

[JimB]

Wow mate cool it mate you're drilling me man.

As I said, excuse my intemperate outburst, but as someone who is new to the forum you will not know that we have had a couple of trolls who ask nebulous questions and when asked for a bit more information all they do is trot out the same words and do not give any more information on which to base a meaningfull reply.

OK, trying to get a bit more positive and helpful...

The first most obvious thing to measure about an RF signal is the frequency.
For most practical RF frequencies (up to a few GHz), this is done with a frequency counter.
Counters which can work up to several 10s of MHz are easily made using a PIC.
The frequency range can be extended up to several GHz using a pre-scaler.

RF power can be measured using a variety of techniques, low powers are paradoxically harder to measure than high powers.

It is also useful to know the spectral purity of a transmitted signal to ensure that you are not vomiting RF noise all over your RF neighbours.
To do this a spectrum analyser is used. Generally speaking this is where you can spend lots of £££ $$$ €€€.
However, is possible to get good cheap spectrum analyser features using an "SDR Dongle".

To test a receiver, you will need a signal generator.
To get a well performing one it is necessary to spend lots of £££ $$$ €€€.

But as already suggested, the best (easiest, cheapest and legal) way for someone with no RF experience is to use the readily available modules.

JimB

 

[BradleyEnglish]

As I said, excuse my intemperate outburst, but as someone who is new to the forum you will not know that we have had a couple of trolls who ask nebulous questions and when asked for a bit more information all they do is trot out the same words and do not give any more information on which to base a meaningfull reply.

OK, trying to get a bit more positive and helpful...

The first most obvious thing to measure about an RF signal is the frequency.
For most practical RF frequencies (up to a few GHz), this is done with a frequency counter.
Counters which can work up to several 10s of MHz are easily made using a PIC.
The frequency range can be extended up to several GHz using a pre-scaler.

RF power can be measured using a variety of techniques, low powers are paradoxically harder to measure than high powers.

It is also useful to know the spectral purity of a transmitted signal to ensure that you are not vomiting RF noise all over your RF neighbours.
To do this a spectrum analyser is used. Generally speaking this is where you can spend lots of £££ $$$ €€€.
However, is possible to get good cheap spectrum analyser features using an "SDR Dongle".

To test a receiver, you will need a signal generator.

To get a well performing one it is necessary to spend lots of £££ $$$ €€€.

But as already suggested, the best (easiest, cheapest and legal) way for someone with no RF experience is to use the readily available modules.

JimB

Thankyou for this great information. But Im still lost about one more thing. How would i send digitalized data via high RF if there isnt a processor fast enough to keep up with the RF ?

Would this mean that my data transfer is limited by the maximum speed of my data processing microcontrollers ?

 

[JimB]

Radio signals comprise a "carrier" and "modulation".

The carrier is the basic signal which defines the frequency of the radio signal.

The modulation is what conveys the information, and depending on the system, it alters or modulates the carrier in some way.
We could have Amplitude Modulation (AM), Frequency Modulation (FM) or some type of Pulse Modulation (PM).
There are several variants of each type of modulation, each with their own advantages and disadvantages.

Consider a basic radio system for voice communication.
There will be an oscillator which generates the carrier frequency, that oscillator is probably controlled by a crystal to give good frequency stability.
There will be a modulator which takes an input from a microphone, and varies the output voltage (power) of the oscillator in accordance with the sound picked up by the microphone.
The carrier frequency is in no way related to the voice frequencies.

Similarly with data transmission.
There is an oscillator generating a carrier.
There is the data which may be lets say an RS232 type signal.
The frequency of the radio carrier is in no way related to the RS232 data rate, and neither is it related to the clock speed of the microprocessor which created the RS232.

The clock speed of the micro has NOTHING to do with the carrier frequency of the radio. They are completely independant.

All this is very simplified, but does it help your understanding?

JimB

 

[BradleyEnglish]

Radio signals comprise a "carrier" and "modulation".

The carrier is the basic signal which defines the frequency of the radio signal.

The modulation is what conveys the information, and depending on the system, it alters or modulates the carrier in some way.
We could have Amplitude Modulation (AM), Frequency Modulation (FM) or some type of Pulse Modulation (PM).
There are several variants of each type of modulation, each with their own advantages and disadvantages.

Consider a basic radio system for voice communication.
There will be an oscillator which generates the carrier frequency, that oscillator is probably controlled by a crystal to give good frequency stability.
There will be a modulator which takes an input from a microphone, and varies the output voltage (power) of the oscillator in accordance with the sound picked up by the microphone.
The carrier frequency is in no way related to the voice frequencies.

Similarly with data transmission.
There is an oscillator generating a carrier.
There is the data which may be lets say an RS232 type signal.
The frequency of the radio carrier is in no way related to the RS232 data rate, and neither is it related to the clock speed of the microprocessor which created the RS232.

The clock speed of the micro has NOTHING to do with the carrier frequency of the radio. They are completely independant.

All this is very simplified, but does it help your understanding?

JimB

Im trying to say that the micro speed will limit the amount of data i can send from the micro into the radio wave.

 

[kubeek]

Well obviously, RF or not, the speed of the micro governs how much data you can put through any channel, be it ethernet, radio, rs232 or anything else.
If you want to put through the channel say 1Gbps, then your micro needs to be able to process 1Gbps, there is no way around that. However it does not matter whether the carrier medium is an ethernet cable with 125MHZ signalling speed, or a 60GHz microwave link, and the micro desnt need to be concerned with that.

 

[JimB]

Im trying to say that the micro speed will limit the amount of data i can send from the micro into the radio wave.

Actually, the radio system will be the limiting factor for the system data rate.
You cannot just arbitrarily keep on increasing the modulating frequency. As the modulating frequency increases, so does the bandwidth occupied by the signal. At some point you will reach the stage where your signal is splattering all over signals on adjacent radio channels.

JimB

 

[ronsimpson]

Adding to Jim's thoughts,

With high data rates the bandwidth is wide, the receiver must have a wide receiving area and lots of noise gets in.
With slow data rates the bandwidth is small and more noise can be filtered out.
Also with slow data there is more time to average out noise.
There is a correlation between data rage and noise or distance you can transmit/receive.