OP25 OP25 Hackrf RX issue

[MTL_Emergencies]

Hi,

I'm trying to monitor a VHF system in my region with a hackrf one but I am able to decode signals.
I have it connected to a VHF antenna and am able to get a clean decode with a regular rtl-sdr device (NESDR Smart 5).
My gains settings are "gains": "RF:0,IF:40,BB:36".
We can clearly see that the datascope signals are very faint.

I was able to get a good decode via sdrtrunk on windows with a measured error of 0.2 ppm (28 Hz).

Does anyone know why that is happening?

Thanks

config file for multi_rx.py

{
    "channels": [
        {
            "name": "RENIR",
            "device": "sdr0",
            "trunking_sysname": "RENIR",
            "meta_stream_name": "metastream_0",
            "demod_type": "fsk4",
            "cqpsk_tracking": true,
            "tracking_threshold": 30,
            "tracking_feedback": 0.85,
            "destination": "udp://127.0.0.1:23466",
            "excess_bw": 0.2,
            "filter_type": "rc",
            "if_rate": 24000,
            "plot": "",
            "symbol_rate": 4800,
            "enable_analog": "off",
            "blacklist": "",
            "whitelist": "renir.wlist",
            "crypt_keys": "key.json"
        }
    ],
    "devices": [
        {
            "args": "hackrf=0",
            "gains": "RF:0,IF:40,BB:36",
            "gain_mode": false,
            "name": "sdr0",
            "offset": 0,
            "ppm": 0.0,
            "rate": 10000000,
            "usable_bw_pct": 0.65,
            "tunable": true
        }
    ],
    "trunking": {
        "module": "tk_p25.py",
        "chans": [
            {
                "nac": "0x14e",
                "sysname": "RENIR",
                "control_channel_list": "143.385",
                "whitelist": "renir.wlist",
                "blacklist": "",
                "tgid_tags_file": "renir.csv",
                "rid_tags_file": "renir-rid.txt",
                "tdma_cc": false,
                "crypt_behavior": 2
            }
        ]
    },
    "audio": {
        "module": "sockaudio.py",
        "instances": [
            {
                "instance_name": "audio0",
                "device_name": "default",
                "udp_port": 23466,
                "audio_gain": 1.0,
                "number_channels": 1
            }
        ]
    },
    "terminal": {
        "module": "terminal.py",
        "terminal_type": "curses",
        "#terminal_type": "http:0.0.0.0:8080",
        "curses_plot_interval": 0.1,
        "http_plot_interval": 0.5,
        "http_plot_directory": "../www/images",
        "tuning_step_large": 1200,
        "tuning_step_small": 100
    }
}

 

[boatbod]

If you're trying for a P25 system you generally want to be set up for cqpsk not fsk4.

            "demod_type": "fsk4",

Try this instead:

           "demod_type": "cqpsk",

What branch of boatbod op25 code are you running from and has it been updated recently?

 

[MTL_Emergencies]

Yeah but the site I'm monitoring is a single tower site thus C4FM. Nevertheless I did try to change it to cqpsk and I still wasn't able to get a proper decode. I'm using gr310 branch and have updated it a few weeks ago.

I remember from the HackRF training that in the middle of the rx spectrum of the hackrf there is some sort of DC spike/hardware error. Do you think it's possible to not center the spectrum on the rx frequency so it's not on that DC spike? (i.e. mid-spectrum frequency is 142.385 and rx frequency is 143.385)?
Also do you think if_rate, tracking_threshold, tracking_feedback and excess_bw should be changed for the hackrf?

Thanks.

 

[boatbod]

Yeah but the site I'm monitoring is a single tower site thus C4FM. Nevertheless I did try to change it to cqpsk and I still wasn't able to get a proper decode. I'm using gr310 branch and have updated it a few weeks ago.

I remember from the HackRF training that in the middle of the rx spectrum of the hackrf there is some sort of DC spike/hardware error. Do you think it's possible to not center the spectrum on the rx frequency so it's not on that DC spike? (i.e. mid-spectrum frequency is 142.385 and rx frequency is 143.385)?
Also do you think if_rate, tracking_threshold, tracking_feedback and excess_bw should be changed for the hackrf?

Thanks.

You can de-center a DC spike by setting the "offset" parameter in the devices section. In the example below I've set the offset to 20kHz.

    "devices": [
        {
            "args": "rtl=0", 
            "frequency": 773843750, 
            "gains": "LNA:36", 
            "name": "sdr0", 
            "offset": 20000, 
            "ppm": 0.0, 
            "rate": 1000000, 
            "tunable": true
        },

What does your mixer plot (#5) look like?

 

[MTL_Emergencies]

Thank you, an offset of 20000 did the trick!
The mixer plot looks like this:

 

[boatbod]

Thank you, an offset of 20000 did the trick!
The mixer plot looks like this:

View attachment 176733

That looks pretty normal for a 4800sps phase 1 channel. If you do anything with 6000sps phase 2 you'll find that it's a bit wider.
Glad the offset got things working for you.

 

[MTL_Emergencies]

Looking back I'm fairly certain that the transmit issue explained here was caused by the offset mentioned in this thread.
Oh well

 

[boatbod]

Looking back I'm fairly certain that the transmit issue explained here was caused by the offset mentioned in this thread.
Oh well

That could be the case. I was reading an article about HackRF and the dc spike and it definitely did say the spike affected transmit as well as receive, but was harder to work around on the transmit side since it's actually caused by hardware.