SDS100/SDS200: Global Auto Filter Vs. Global Attenuation

[JASII]

I was reading through some previous posts and I am unclear on the difference between Global Auto Filter and Global Attenuation.

What I have been doing recently is scanning groups of frequencies in sequential order to determine what DMR Color Codes are being used. On a few UHF frequencies in the 450 and 460 Mhz area I have had multiple Color Codes logged. While it is possible that they are all legitimate frequencies using those Color Codes, it is also possible that some erroneous things being logged. Anyway, as I was programming some frequencies, I ran across the Global Auto Filter and Global Attenuation settings. Now that the SDS100 has been around for a while, should I try either the Global Auto Filter or Global Attenuation on my SDS100 when searching/scanning several MHz at a time?

 

[dougjgray]

Global means it will apply the filter and/or attuanation to all. Though auto filter is kinda slow

 

[n1chu]

As far as global attenuation is concerned I don’t think I’d be applying that anytime soon... not globally anyways. If I have a single signal that is so strong as to affect the audio, I would consider attenuation, but for that agency only, not the entire programming.

As for filtering and when to apply it, in any form, I just apply it and compare my RSSI readings. If I see a better RSSI, fine. But if you are mobile, it’s going to change, making any filter a personal choice... because not every setup is the same and your results WILL vary. At home I just pick an agency and look at the different RSSI readings the radio sees and set it there for a while, recheck it. There’s also a concern that some of these filters will cause your scanner to have a slower scan rate. Read the threads on filtering to identify which ones do this. This group has some knowledgeable folks that have written extensively on the subject. My hats off to them. But my takeaway on the subject is if you want to try the filters, go ahead. Do some comparisons and see what results you get... keeping in mind your results mean nothing to anyone else. That’s why I don’t post mine. They may work for me but Uniden has stated that’s not saying they will be effective for others.

 

[Ubbe]

As for filtering and when to apply it, in any form, I just apply it and compare my RSSI readings. If I see a better RSSI, fine.

It's probably the opposite, as the RSSI signal are displaying the signal level from all adjacent frequencies that interfere. In digitial mode the d-error are the best value to monitor for lowest number of data errors and for analog it is probably better to monitor the NOISE level for as low value as possible. RSSI will probably only fool you that you think you have the best filter in use when in fact you have the most interferenceis possible recieved with that filter setting.

/Ubbe

 

[n1chu]

Opposite? It’s a simple comparison. Choose the better reading, or not. I happen to like using the RSSI values to determine if a filter is beneficial, as suggested in a previous post. As for the reason for the filters, I doubt I have need for any filters since I never missed them before they became available. But it’s nice to have choices. I thank Uniden for the filter offerings. They may prove to be advantageous or not. You don’t have to use them. What’s “opposite” about that?

 

[Ubbe]

The RSSI reading are taking from a wide frequency spectrum that includes any adjacent frequencies as well as the one you are monitoring. If you select a filter that includes more adjacent frequencies that have transmitters on them it will give a higher reading but the other two readings I suggest to use are more true to the actaual signal you are receiving, excluding any interfering adjacent frequencies.

RSSI are useful indicator in a standard type of scanner that gets the value after very narrow filters that only look at a very small frequency spectrum that only include your monitored frequency. Not so with the type of receiver used in the SDS scanners and the RSSI value are more often than not an untrue value of the frequency received.

/Ubbe

 

[n1chu]

“An untrue value more often than not“? So, Uniden has supplied us with a Received Signal Strength Indicator that provides untrue values? Is it generous or stingy with its readings? Both?

I have not re-read this in some time but it sounds like it’s pretty much the defining word on the subject. Id be interested in what you think after reviewing it.

The following is from Paul Opitz, (RIP):


From UPMAN
Here is a great write up of RSSI values coming out of an SDR. Different chip, but should be very close to the same situation with the SDS100.

https://dsp.stackexchange.com/questions/30889/possible-to-derive-receive-signal-power-from-rssi

Marcus Miller is an undeniable genius for all things SDR. Here is text from the above, and nearly all his words and absolutely none of them mine:
[The] Received Signal Strength [Indicator is] always relative to some signal model, incorporating considered bandwidth, assumptions on the modulation scheme, duration of transmission, generally: It's a estimation of received signal strength based on some property of the received signal.

Yes, I wrote that, and I still think it applies, so:

What is the difference between RSSI measurements and the dBm values that I record from the spectrum analyzer?

Well, the RSSI is calculated based on some signal-model based estimator, whereas the dBm value on your spectrum analyzer is calculated with another metric – namely, the power measured going through a filter with the filter bandwidth you configured at your spectrum analyzer.

These two might, or might not be related, taking into account that even a signal with very low power spectral density values might be sufficiently good for something that has e.g. matched filters to extract the signal from noise, or any other form of processing gain.

So, your only option is to understand both estimators, with the Spectrum Analyzer estimator being relatively simple, and the flex estimator being free and open source (so you can just read the code ), and relate the two; so basically, you'll have to find, given a (class of) signal(s) s(t), the functions fSA, fflex and finally g from

Pspectrum analyzer(s(t))Pflex frame sync RSSI(s(t))⇝P′flex(Pspectrum analyzer)=fSA(s(t))=fflex(s(t))=g(fSA(s(t))) .
I've just skimmed the liquiddsp code for the RSSI estimation, and the interesting part seems to happen in qdetector_cccf_execute_align.

However, this can never work without making a lot of assumptions. For example, if your signal happens to have a very narrow, but very strong single tone interferer, it will probably not break flex reception, and if that RSSI estimator is any good, not change RSSI very much, but it will drastically change the dBm value displayed by your spectrum analyzer. So if the question was

Can I find an estimator that always gives me the same reading as the liquidDSP flex RSSI based on the dBm value displayed by a spectrum analyzer?

then the answer is a

No. See above proof by counterexample (single tone interferer).

However:

If I cannot relate these two measurements by a correction factor, is there a way to obtain the receive signal power measurements (in dBm) programmatically (GNUradio perhaps)?

Sure. Why not? You got the raw samples, don't you? So do a magnitude squared of them, and find a linear correction factor. That will work (have done so multiple times), because now you're actually measuring to things that relate to each other (signal amplitude as seen by your SDR frontend's ADC and signal power as seen by the spectrum analyzer simply have a quadratic relation).

Note that you will have to re-calibrate for different frequencies, sampling rates, if settable, analog frontend widths (only applies to B2xx, E3xx, and a few very old daughterboards for your N2xx currently; by the way, the TVRX* boards have always-on AGC, so they can't be used for this).

The relationship between input voltage and digital number amplitude is linear for quite a large range for the Ettus frontends – however, as soon as your input amplifier and mixers start reaching saturation, this will no longer apply, so you did good to use a few dB of attenuation.

If you happen to have a WBX, SBX, CBX or UBX daughterboard, there's IP2/IP3 measurements (along with effective vs set gain) available that will help you estimate the region of linearity on files.ettus.com:/performance_data/.

shareimprove this answer
edited May 21 '16 at 14:03
answered May 19 '16 at 10:08

Marcus Müller

 

[Ubbe]

If you have a signal generator, set it to a nearby frequency to the one you are receiving a signal from. Increase the level from the signal generator and you will see that the RSSI value increase as if the frequency you are monitoring are getting stronger.

I will do some tests to see how far off in frequency the RSSI indicator are sampling it's signal from.

/Ubbe

 

[n1chu]

I was reading through some previous posts and I am unclear on the difference between Global Auto Filter and Global Attenuation.

What I have been doing recently is scanning groups of frequencies in sequential order to determine what DMR Color Codes are being used. On a few UHF frequencies in the 450 and 460 Mhz area I have had multiple Color Codes logged. While it is possible that they are all legitimate frequencies using those Color Codes, it is also possible that some erroneous things being logged. Anyway, as I was programming some frequencies, I ran across the Global Auto Filter and Global Attenuation settings. Now that the SDS100 has been around for a while, should I try either the Global Auto Filter or Global Attenuation on my SDS100 when searching/scanning several MHz at a time?

JASII, The RSSI numbers come with a “-“ (minus) preceding the value. -60 or -95 for example. If you see something around -60 you are looking at a better RSSI than -105. So look for the lesser numerical value. I’m not sure if you were guessing the higher numerical values were better... they are not. And it’s possible UBBE was referring to this when he used “opposite” in his post... he may clarify this for you.

 

[Ubbe]

OK, so I made some measurements on a SDS100 and it's plauged with intermod issues.

First the RSSI indicator.

I monitored a frequency in the 400Mhz band with a signal of -100dBm, just a little noise in it so it can be heard if the sensitivity changes. I used the filter set to Off, which is the standard symetrical filter used in any other scanner.
I injected a -50dBm signal and at a frequency 7MHz lower than the monitored one. The RSSI changed from showing -100dBm to -60dBm.
When I got within 3MHz it started to show -50dBm.

The desensening of the receiver, how the RF AGC adjust the gain by the result of signals off frequency, where that the reciever lost 5dB of the signal when injecting -80dBm signal 100KHz from the monitored frequency.
Injecting a -60dBm signal gave 20dB loss and reception where almost completly lost and could only be heard faintly with the squelch open.

With the -80dBm injected signal it gave intermod at two frequencies 300Khz and 600KHz from the monitored frequency and totally killed the reception.

When using different kind of filters everything changes and moves around. But one thing that seemed to make the receiver perform much better where when IFX where used. In Unidens other scanners there's two filters, one around the 250Mhz band and one around 350Mhz that IFX switches between, and I suppose it is the same in the SDS100 and one of the filters might be of a much higher quality, or are better suited to be used in the circuit. Couldn't find any intermod when IFX where enabled.

General reception are good at a 0,35uV value in all bands, even VHF air. But it seems that the gain could had been set too high. If an air pilot raise his voice or his microphone are more sensitive, the modulation goes into the roof and clips and create gun shoots in the audio. It can be heard sometimes even in other frequency bands in FM mode.

It sounds as if a very narrow demodulation filter has been used for VHF air as the noise and voice are cut lower in audio frequency. In Uniden standard scanners they for some reason use the FM filter in AM and not the NFM filter that would be more suitable. But in SDS100 the DSP has been set to something that sounds like a 6KHz wide filter. VHF airband have mostly an excellent quality and are better than any other scanners I have, but could be that the RF AGC that have been set a bit too optimistic. Usually AM mode isn't as sensitive as FM mode in a scanner.


Conclusion are that the intermod issues that kills reception appear at different frequencies and are not something that slowly dissapears as the offending frequency are further away from the monitored frequency. So it is a guessing game with the filter settings but IFX seems to help the most and makes the receiver behave much better, at least in the 400Mhz band.

Nearby frequencies will reduce the receivers gain and sensitivity. The RSSI indicator starts to react to signals as far away as 7MHz from the monitored frequency. It's difficult to have a good RSSI meter when the signal starts to go up in the -60dBm strenght, as the measured voltage from the receiver chip has very small differences at those signal levels. It sometimes jumps from -60 to -40 and the -40 got stuck a couple of times even when lowering signal to -90dBm. It's a very unreliable indicator. When it works, with a single pure signal source, it shows about 5dB too much but seems to be usefull up to -60dBm.

The pre-filters in the SDS100 are different for each frequency band but seems to be very flat with bad return loss and high swr. It's probably not the same type and of same high standards as in Unidens other scanners. These cheap SDR receiver chips needs better pre-filters than other scanners to help with their poor performance of intermod and desense issues.

/Ubbe

 

[kruser]

Thanks for sharing your test results Ubbe.

 

[Ubbe]

I didn't look close enough using finer frequency steps on the signal generator. It's the same amount of intermod also when using IFX.

I now also tested with only the signal generator connected directly and used a -60dBm signal and there's a huge amount of intermod going on. I compared to my RTL-SDR and at the same signal level and it had actually less intermod.

I actually didn't had the gain adjusted high enough on the RTL-SDR. It had a 0,6uV sensitivity and when increasing the gain to make it 0,35uV it probably had about the same amount of intermod issues as the SDS100, a lot. The SDR chips are from the same family of circuits from the same manufacturer so the SDS100 and RTL-SDR probably perform pretty much the same when it comes to intermod and overload issues.

/Ubbe

 

[Ubbe]

I had one system change it's color code from 1 to 0 and the programming in SDS100 remained at a fixed 1 for the system. It still opened up and monitored the system in about 1/3 of the occasions when it was transmitting a 0, as verified on other scanners. So I advice against using fixed color codes, or NACs or any kind of restrictions as the decode accuracy seems to be a bit off in Unidens scanners. If some other system use the same frequency then it probably anyhow will create havoc with your monitoring.

Uniden manage to use the wrong trigger for backlight in DMR systems. It turns the backlight on when someone stops talking and not when they start to transmit.

A problem with digital modulation are that the DSP seems to be programmed to use a more narrow filter than conventional receivers, probably to try and filter out some of the interferencies, so that when the modulation reaches near maximum it will create lots of noise in the signal. That's the reason why FM in most cases works better than the NFM setting with digital modulation, as reported from some users, if the signal are not fully saturated at a higher signal level.

/Ubbe

 

[Ubbe]

Deleted

 

[Ubbe]

I did a test by generating a -60dBm signal from a signal generator modulated with a tone and opened the squelch and stepped away from the frequency in 5KHz steps.

I used 850MHz as a frequency and made a recording of the audio. After a while the scanner didn't liked the interference signals and rebooted and settings where restored to what was used at last power off.

In the 800Mhz band with a search with squelch set at 2 the interferencies doesn't seems to affect the search rate at all. In UHF it starts to hesitate at certain frequencencies and at VHF it really starts to slow down search at about 50% of the time. The interferencies are of such high distorsion and squeling high pitch noises that the squelch doesn't open but only stops briefly and then continues to search.

The audio interference where exactly the same in all bands VHF,UHF and 800Mhz, so it's probably something that happens in the receiver after the first mixer in the 1:st IF. If you only scan digital or mostly use 800Mhz band you'll probably never be aware of the problems with the receiver. These interferencies will blend with all signals it recieves and create mixing products that spread all over the frequency band.

Make a live test at home and open the squelch and listen to the noise while stepping thru the frequency range to see how much or little interferencies you have at your location. Compare to the clean noise when the antenna are removed.

For some reason the signal strenght where showing -45dBm on the display during the whole test. When the RF signal where removed it displayed dashes and went back to -45dBm when signal where applied again.

Recorded audio: SDS100_800MHz.wav

/Ubbe

 

[Kaleier1]

Good recording. Interesting how on some frequencies you don't hear the tone but on ones further away you do. I guess that's why I can't even listen to FRS frequencies on my SDS200. On FRS channel 7 462.7125 MHz an ambulance company on 462.975 MHz comes through loud and clear and that is a difference of .2625 MHz. The rest of FRS channels get interfered with by DMR control channels from other frequencies.

 

[Saint]

Good recording. Interesting how on some frequencies you don't hear the tone but on ones further away you do. I guess that's why I can't even listen to FRS frequencies on my SDS200. On FRS channel 7 462.7125 MHz an ambulance company on 462.975 MHz comes through loud and clear and that is a difference of .2625 MHz. The rest of FRS channels get interfered with by DMR control channels from other frequencies.

You could try an FM TRAP inline and see if that will help It made a difference for me on some of the bands but I still get get interfered with by DMR control channel on some FRS frequencies.
Steve

 

[Kaleier1]

You could try an FM TRAP inline and see if that will help It made a difference for me on some of the bands but I still get get interfered with by DMR control channel on some FRS frequencies.
Steve

Thank you. Do you have a suggestion of which one to get? `I see most are for 88 to 108 Mhz broadcast band. That's not my problem. My problem is frequencies 262.5 KHz away from the one I am listening to.

 

[Ubbe]

It's not much you can do than trying different filter settings and enable IFX on the frequency.

All my signals are -80dBm down to -105dBm and it's a lot of mixing and blending of different frequencies going on with those weak signal levels, despite using FM trap filters.

With the -80dBm injected signal it gave intermod at two frequencies 300Khz and 600KHz from the monitored frequency and totally killed the reception.

There's a lot of birdies in the scanner, like any other SDR dongle receiver. It makes it impossible to monitor some even frequencies like 156.000 and 142.000 and 468.000 The most ones appear in the 800Mhz band but some of them dissapears when the display turns off.

/Ubbe