jks

About

Username
jks
Joined
Visits
32,340
Last Active
Roles
Member, Administrator, Moderator
Points
331
  • Demo of new user interface [now available in v1.154]

    Addition of classic LMS "denoiser" and "autonotch" filters:

    Least Mean Square (LMS) adaptive filters. The technique in ham radio is more than 25 years old now. But it remains a simple and sometimes effective solution, particularly for auto-notching interfering carriers. A simple and preliminary version of LMS filtering has been added to the new user interface demo. Better versions are being evaluated.

    The filter enable checkboxes are on the "audio" option tab. A "more" button opens an LMS_filter extension with sliders for some of the filter parameters. In particular the "beta" parameter can be adjusted to notch carriers without also notching CW signals since they are not transmitting continuously (but the notch will not be perfect). The denoiser does not work well for music and is better for voice-only AM transmissions. NBFM and IQ modes do not use the LMS filters at present.

    A few beta test Kiwis with better reception than http://kiwisdr.jks.com:8073 have been updated to run the new software:


    KA7U
  • Where is the antenna description entered in the administration page?

    For the main (land) page it's taken from the antenna field on the sdr.hu tab. The alternative was to have a separate antenna field on the webpage tab.

    rrobinet
  • Seeed KiwiSDR metal enclosure finally available!

  • Grove code for KiwiSDR

    Good point. I have not looked into the Grove stuff. But from the PCB silkscreen near the connectors it seems one of the two Grove connectors provides UART and the other SDIO/I2C. These can also likely be put into GPIO mode with the right magic. And it should be possible to do an antenna switch extension backend for them.

    Each connector also has 3.3V/GND. So if you're using the new Kiwi aluminum enclosure you can move the fan to the other connector if necessary.

    WA2ZKD
  • Demo of new user interface [now available in v1.154]

    Here is a demo of a new user interface under development. There is no antenna connected at present. http://kiwisdr.jks.com:8073/


    Changes:

    • New "option tab" bar that categorizes controls and status information.
    • Frequency history next/prev buttons. Up/down arrows to right of frequency entry field. Keyboard users: press return key when desired frequency appears. If no selection with return key is made within 3 seconds frequency field will return to current frequency. Mobile device users: this feature is not useable at the moment.
    • Added numeric value (in dBm) to S-meter.
    • Add URL parameter wf={0, off, 1, 1hz, s, slow, m, med, f, fast}. E.g. kiwisdr.local:8073/?wf=1hz
    • Removed “less buffering” and “compression” buttons until working.


    KA7U
  • Add waterfall rate to URL arguments

    No, the WSPR extension would actually still work because the DSP part is done on the Beagle. The transmission of the WSPR spectrum, decode information etc. happens separately from the audio and wf streams. You are correct that some extensions process the audio/IQ data in the browser. And for those a Mute cannot stop the audio/IQ data. That's one of the reasons handling the audio mute case is a little more complicated.

    KA7U
  • Save money on another Kiwi or tell a friend [KiwiSDR sometimes on drop.com for $100 off MSRP]

    December 14/15: Massdrop is active again for 7 days. US$230 for a KiwiSDR kit if they get 3 orders. Limited to 6 units total. Best deal out there!

    WA2ZKD
  • Important: updated assembly guide (v1.1) for the aluminum enclosure

    There is an updated assembly guide (version 1.1) for those of you who have purchased the new aluminum enclosure. Corrections to the parts list, clearer photos and an important note about removing one of the two pieces of sticky tape attached to the fan. Special thanks to Dave, PJ4VHF, from Bonaire for discovering the issue.


    KA7U
  • OV Worries

    I've explained this somewhere before (it really needs to be a FAQ so people can find it).

    The current FPGA logic just records an OV event if a single overflow is reported by the ADC. This is nonsense of course. You'll get overflows during normal operation as all those sine wave peaks from strong HF signals just happen to sum up during one ADC sample. That in itself is not a bad thing. You won't hear distortion because of it. Some sort of averaging of the ADC overflow output needs to occur and only when a threshold is exceeded should an OV event occur. 

    And that's really your question: what should the threshold for ADC overflows be before an error indication is given? You could even imagine a a couple of different threshold being shown, done e.g. with colors.

    WA2ZKD
  • Wideband pulsing noise & Noise Blanker absence: I'm in trouble! [NB added in v1.152]

    1 / 12 kHz = 83 usec. And the NB needs a couple of gate samples to work properly. So going below 100 usec doesn't get you any improvement.

    Moving the NB to a higher sampling rate in the FPGA is one option. But that will take some effort. Or increasing the audio bandwidth, which has its own problems.

    KA7U