HackRF One, ADALM-PLUTO comparison, or other possibilities

Talk about other SDR products like the FunCube, HackRF, BladeRF etc.
Post Reply
KevinJones
Posts: 21
Joined: Mon May 04, 2020 4:38 pm

HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by KevinJones » Tue Nov 24, 2020 5:27 pm

Hi,

I'm looking at using an SDR device as a spectrum analyzer to detect and analyze very weak cellular signals. I am feeling that I should get something better than a $20 dongle. As I am needing to look at both the 850 MHz and 1700 - 2100 MHz bands, Airspy R2 and SDRPlay RSP-1A don't qualify.

So ADALM-PLUTO costs $150 from AD. A genuine HackRF One costs around $300+. Is there anything special about the HackRF that would make it worth spending the extra money?

The ADALM-PLUTO claims a NF or 2.5db. Some[1] claim the HackRF to be around 3.4 db with LNA in place. I also hear that the HackRF has no input overpower protection, and -5 dBm will destroy the front-end; I don't know if the PLUTO has this or not. (While I said I want to use it to detect weak signals, it is also conceivable that I would be near higher power devices such as when testing out cellular amplifiers.) I'm not stating these things necessarily as a dis to the HackRF; just reporting some of the information I've gathered. The additional higher upper BW figure for the HackRF (6 GHz vs 3.8 GHz for the PLUTO) isn't really an advantage to me.

Or does anyone have any other input or suggestions on this matter? As the signals I'm looking at are close to the noise floor, NF would be important as well as the tendency to create spurs. Good spectrum analyzer software would be important as well, if anyone has input on this. (I've toyed a little bit with SDRs in the past, and observed that some of the software does not report values in absolute power (eg, dbM) but as "relative" db (relative to what?). Knowing absolute power would be essential.)

Thanks...

[1] https://www.rtl-sdr.com/measuring-the-n ... real-time/

KevinJones
Posts: 21
Joined: Mon May 04, 2020 4:38 pm

Re: HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by KevinJones » Thu Nov 26, 2020 1:47 pm

I'd like to expand my question. I know in my original post I "disqualified" the Airspy R2 and SDRPlay RSP-1A based on their lower upper-range spectrum limit. I am removing that disqualification, and would be interested to know, just based on Rx capabilities, would the HackRF offer anything that would make it a better choice as a spectrum analyzer, in particular, for analyzing very weak signals, than either the Airspy R2 and SDRPlay RSP-1A? Or does anyone have any input they could offer comparing the Airspy R2 and SDRPlay RSP-1A?

Thanks. The lack of replies from my original post made me wonder if folks here just aren't as familiar with the Pluto.

snn47
Posts: 239
Joined: Tue Dec 27, 2016 11:00 pm

Re: HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by snn47 » Tue Dec 01, 2020 5:51 pm

There is no simple answer to your question, it depends on the frequency range and bandwidth you want to monitor. The intermodulation free dynamic range (ratio of strong to weak signals received) limits the sensitivity you can achieve. How strong would undesired signals be at your receiver input?

Even in comercial SA the intermodulaion free range dynamic range will limit the sensitivity you can achieve, because without filter to reduce strong signals, you have to sitch on attenuators which reduces the sensitivity.

KevinJones
Posts: 21
Joined: Mon May 04, 2020 4:38 pm

Re: HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by KevinJones » Mon Dec 07, 2020 3:28 pm

@snn47: Thanks for your reply:
snn47 wrote:
Tue Dec 01, 2020 5:51 pm
Even in comercial SA the intermodulaion free range dynamic range will limit the sensitivity you can achieve, because without filter to reduce strong signals, you have to sitch on attenuators which reduces the sensitivity.
Would you mind expounding on this? Would placing a bandpass filter for the frequencies of interest ahead of the SDR increase sensitivity for those frequencies? Is that what you are meaning?

ps, I'm sorry I didn't acknowledge your reply sooner, I was waiting for email notifications, but never received any (not in spam folder either) even though my site preferences are set to receive them.

snn47
Posts: 239
Joined: Tue Dec 27, 2016 11:00 pm

Re: HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by snn47 » Tue Dec 08, 2020 2:18 pm

Would placing a bandpass filter for the frequencies of interest ahead of the SDR increase sensitivity for those frequencies? Is that what you are meaning?
No, I know it is not the simple answer you have hoped for, but it depends on the number of interfering signals and their strength and how many MHz the strong signals are separated from the frequency range you want to monitor, if a filter can improve reception at all.

Placing filter in front of a receiver does not increase sensitivity, but reduces sensitivity and increases the receiver noise factor. A filter allows the receiver to operate in presence of strong signal, which otherwise would lead to the condition, that the receiver will generate IM (intermodulation).

Any receiver that is operating within the dynamic range will amplify all received signals by the same factor. Once signals are to strong that the receiver cannot amplify those strong signals sufficiently, compression occurs and in consequence receiver intermodulation occurs.

Intermodulation products e.g. IM 3rd order, occurs when two strong signals generated wo additional signals in the receiver. The first IM is 2*f1 -f2 the second is 2*f2 -f1. The more strong signals you have the higher the IM order is the more IM products will be generated.

In presence of strong noisy signals or signals with a large bandwidth the receiver noise floor may increase.
You can separate real signals from IM products, when you place (adjustable) attenuators in front of the receiver input and increase the attenuation. A signal should reduce in strength by the same factor that you insert attenuation, e.g. if 10 dB of attenuation are added all signals should be 10 dB weaker. If signals get more than 10 dB weaker they are IM products generated in your receiver.

If you have an attenuator in 1 dB and increase the attenuation in 1 dB steps you can optimize the receiver for the min. attenuation required that your receiver is IM free. Unfortunately every dB in attenuation added will reduce the sensitivity by the same amount of dB.

Simplified filter are frequency selective attenuator, meaning they attenuate only certain frequencies or frequency ranges, e.g. a LP- and HP-filter have only a few dB of insertion loss in the pass band, an increasing attenuation in the filter skirt and <60 dB in the stop band. Band-stop Band-reject or Notch filter block a frequency or frequency range, while a Band-pass is the opposite filter type.

Unless filter provide enough rejection in the stop-band and filter skirts to attenuate the strong signals that create IM, so that the receiver in IM free operation, adding filter won’t help.

Filter properties vary with operating frequency, meaning the higher the frequency the wider the filter skirt and pass band, while the attenuation in the pass-/stop-band and the width of filter skirts depend on filter design. If your interferer are to close to the receive frequency, e.g. FM-BC transmitter which can exceed 100 000 kW, two on 107.75 MHz and 107.900 MHz would generate IM on 108.05 MHz. In those cases only a better receiver with a sufficiently large dynamic range will help.

KevinJones
Posts: 21
Joined: Mon May 04, 2020 4:38 pm

Re: HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by KevinJones » Tue Dec 08, 2020 4:43 pm

snn47 wrote:
Tue Dec 08, 2020 2:18 pm
Would placing a bandpass filter for the frequencies of interest ahead of the SDR increase sensitivity for those frequencies? Is that what you are meaning?
No, I know it is not the simple answer you have hoped for...etc
Thank you very much for the detailed explanation.

(Just when I thought understanding RF was overwhelming enough, they pile on more...)

So I've been reading up on IMD. From what I understand this is a result of non-linearity in processing the signal (either in the analog or digital realm.) So the IMD is being created in either the SDR hardware (which sounds more likely to which you are referring) or the algorithms processing the output in the computer. Although it seems that you are referring to IMD that results from stronger signals/noise/interference close to the frequencies of interest.

It seems then that both components are needed for IMD to occur – both non-linearity of the system and strong interference at nearby frequencies – is that true?

Is sounds like you are saying that:

1) By adding linear (eg, near totally resistive) attenuation, the IM distortion will be attenuated more than the signal of interest (though it is not immediately intuitive to me why that would be) and thus, while the signal may be at a reduced level, it will be more distinctive in the sea of noise. No?

2) Bandpass filtering (frequency selective attenuation) the input to the radio may or may not help, depending on the Q of the filter (and perhaps other factors), no?

3) That a receiver with larger dynamic range is more immune to the effects of IMD, and thus would be a better detector of very weak signals. No?

4) Perhaps the best solution for all of this is to look for a receiver with the highest dynamic range. No?

(This is all I have for now.)

Thanks again, Kevin

snn47
Posts: 239
Joined: Tue Dec 27, 2016 11:00 pm

Re: HackRF One, ADALM-PLUTO comparison, or other possibilities

Post by snn47 » Tue Dec 15, 2020 11:11 am

Hello Kevin,
1. If the IM distorion is within your receiver attenuating all signals will also reduce the burden on your RX.
e.g. take a 20 dB preamp it will amplify a single -90 dBm to -70 dBm, or a single -20 dBm signal to 0 dBm. If it cannot amplify the -20 dBm by 20 dB to 0 dBm it is in compression and that’s when the blocking and IM start.
Therefore unless you add a filter that will limit the bandwidth and reduce the number of signals that the preamp has to amplify,
or add attenuator to bring your preamp in linear mode, the strength and number of undesired signals will determine your IM and blocking.

2. Yes, if your filter can reduce the signals that can cause you problems. It all depends on how much you can suppress the undesired strong signals that cause you the problems. Often those transmitters are to close in frequency and produce to strong signals, that it is hard to impossible to suppress them.

3. No, larger dynamic range will only increase the range between strongest to weakest signal that you can receive. It will not automatically improve the robustness of an Rx to interference, e.g. PSR Radar Rx have a large dynamic range to allow reception of signals down to ~-130 dBm, but they are also very sensitive to interference.

4. No, dynamic range is only one factor.

IM or increase in noisefloor can also occur in transmitter or between transmitters, e.g. those located at the same mast or using the same antenna. Unless the Tx have a continious power output, interference in base station tower may occur at one time when the Tx operate at max power and no interference when they reduce their power. and spurious

Post Reply