New rtlsdr Plugin

[rgleason]

I have tried adjusting ppm to -75 and 75 and nothing seems to happen. Maybe I am missing something?

Sent from my SCH-I545 using Cruisers Sailing Forum mobile app

[rgleason]

In the bottom graphic when I hover mouse, a red line and bandwith indicator come up at 162.550. At the top do you have center. Skicky or free tuning set?

Sent from my SCH-I545 using Cruisers Sailing Forum mobile app

[rgleason]

I see now you leave the settings window open and adjust ppm up or down with volume turned way ip to hear boices inproving as the lower window band starts to align with the red setting line in the upper window.
I got two results with small antenna. At 162.500 and ppm -165. And at 162.550 and ppm +129.

These setting do NOT appear to be the same for all frequencies so this dongle is not linear and will be very difficult to tune gor AIS in my opinion.

Additionally I note that as it gets warmer the signal fades out and changes. This is a marginal device I guees.

Sent from my SCH-I545 using Cruisers Sailing Forum mobile app

[rgleason]

Ia Nooelec much better?

Sent from my SCH-I545 using Cruisers Sailing Forum mobile app

[LeoA]

The two nooElec dongles both seem to be pretty stable. One has a ppm of 55 and the other has a ppm of 74. I keep the ppm the same no matter the frequency that I tune.

[the_muck]

Hallo,
piluso, thanks i have test the software (https://github.com/dgiardini/rtl-ais/releases/tag/v0.1) and it works, have you make a performance test between rtl_ais, AISRec and PNAIS?

I use this Antennas Homemade Flower Pot Antenne für AIS 162MHz | MUCK Solutions

and this stick USB DVB T Rtl SDR Realtek RTL2832U R820T DVB T Tuner Receiver PAL IEC Input | eBay

i start with... in a *.bat file

@echo off
rtl_ais -n -R -p 19

[duichan]

I have just returned from a week's trip across the English Channel, during which I was able to test rtl_ais exhaustively (with a Newsky R820T2 dongle on a RPi2 running standard Raspbian Wheezy and OpenCPN 4.0). I was very pleased with the results. Using only a rail-mounted rubber duck antenna the range was at least 16 miles (sometimes more). I had sufficient confidence to trust it on our return journey, crossing busy shipping lanes in reduced visibility (2 NM). Even in the Solent, with 100 active targets being processed, the CPU load was not such as to cause any concern. I can confirm that the auto calibration facility really does need a lot of AIS targets to prevent it from crashing. Interestingly, the error came out at 32ppm, which is exactly the same as the value I got by eyeballing BBC Radio 4 FM with SDR#.

[rgleason]

As a result of my difficulties getting any AIS signal with two USB dongles, I am going to try to purchase one than works and make a simple 162mhz antenna to go with it.

I have decided to try the Nooelec with the Muck's very simple antenna or Arudale's.
Any suggestions before I order it. for this Wednesday?

NooElec NESDR Mini 2 USB RTL-SDR & ADS-B Receiver Set, RTL2832U & R820T2 Tuner, MCX Input. Low-Cost Software Defined Radio Compatible with Many SDR Software Packages. R820T2 Tuner & ESD-Safe Antenna Input, Guaranteed

AISspotter - PIA Dual channel USB AIS Reciever + GPS

SeaAngel FT-Tec Electronics Seanexx RX-200 Dual Reciever
Seanexx RX-110 AIS + GPS
The RX110 models as an AIS receiver with integrated GNSS receiver with which you can also display your own GPS position.

AIS+GPS+BT and Wifi Adapters for Pilots - $


Homemade AIS Antennas

the Muck 162mhz Flower Pot

High Gain Collinear AIS (162MHz) Receiving Aerial
by Neal Arundale - M1CHS

[hsi88]

I bought 3 different dongles, plus all the antenna adapters and although I got them working with SD#, AISMON and OpenCPN, they were a PIA to work with. None of them received the AIS signals without readjustment and the hook up was pretty hokey. It was trial and error because I have NO idea what most of the settings in SD# actually do. I'm not a radio man!

I bought this AIS receiver instead. Plugs right into the laptop via USB mini cable. The other end needed a UHF adapter for my VHF mast antenna (SO239 to BNC), but other than that, it was pretty much plug and play with the downloaded driver.

Internal software handles both AIS signals and switches automatically between them, and converts the signals to NMEA sentences for you.

Easy peasy. $55.

https://www.tindie.com/products/astu...-ais-receiver/

[rgleason]

Thanks hsi88 for the suggestion!

[rgleason]

the Muck, tried your quick and easy 162 mhz antenna this afternoon. Used duct tape to keep the 8 coils together, then taped the antenna to thin wood garden rod and taped the antenna and coax to that rod.

I tested the antenna with my two garbage rtlsdr dongles and found that the original worked better for the weather stations, but that does not really inform much.

I have ordered a dAISy and hope to have some positive results with the antenna

[the_muck]

Thanks for testing, i have with this FPA antenna a range up to 35Nm. But you can try the easy Dipol antenna too.

When i use rtl_ais i have to enable the RTL chip AGC, without it not works. So is use AISRec demo for testing my Hardware and SDR# for the PPM correction.

The interest think is that rtl_ais and PNAIS, sometimes not find ships near me. When i Start AISRec i find much more Ships...

[seandepagnier]

Originally Posted by duichan

I have just returned from a week's trip across the English Channel, during which I was able to test rtl_ais exhaustively (with a Newsky R820T2 dongle on a RPi2 running standard Raspbian Wheezy and OpenCPN 4.0). I was very pleased with the results. Using only a rail-mounted rubber duck antenna the range was at least 16 miles (sometimes more). I had sufficient confidence to trust it on our return journey, crossing busy shipping lanes in reduced visibility (2 NM). Even in the Solent, with 100 active targets being processed, the CPU load was not such as to cause any concern. I can confirm that the auto calibration facility really does need a lot of AIS targets to prevent it from crashing. Interestingly, the error came out at 32ppm, which is exactly the same as the value I got by eyeballing BBC Radio 4 FM with SDR#.

Great.

Is there any possibility you can post pictures of the dongle and antenna? I ahve no idea what a rubber duck antenna is.

As for the auto calibration facility, which one specifically are you talking about?

I just updated the rtl_sdr plugin to hopefully include support for rtl_ais, but of course I still want to build it directly into the plugin if possible because people using windows will always have problems until this is done.

The issue I have is that rtl_ais is compiled using mingw correct? While I use mingw to compile opencpn (so there would be no problems) most people continue to use visual studio. I think there is a problem linking with the libusb libraries, but maybe not if they are c not c++?

[duichan]

Hi Sean

The SDR dongle is this one. It uses the R820T2 chip which is said to have slightly greater sensitivity and lower noise than its R820T predecessor. It also has an SMD crystal which has significantly better frequency stability. I think a very similar dongle is available from Nooelec in the States.

The antenna I used is here. It's just a small, basic VHF antenna that we normally keep as a back-up in case the mast head one fails.

The auto calibration facility I was referring to is your "Auto Calibrate" button under "Error Correction Calibration" in the preferences dialogue.

I can't comment on any Windows version as have been working exclusively with the RPi2.

Looking forward to your further developments

Regards

Andrew

[seandepagnier]

Andrew

My autocalibration is inferior because it restarts rtl_fm with different error corrections repeatedly until it can decode some messages then fine tunes it a little.

A really simple way (not least cpu though) would be to add a flag to rtl_ais for calibration. Then instead of using a passed error correction value, it would try lots of error corrections (say every 5 ppm) for the input stream. Once it successfully decodes a message, it could then fine tune the error by trying every ppm and taking the median of the ppms that successfully work.

In theory this method should work with only a single ais message received, and could use threads if a single rpiv2 can't handle it in real time, otherwise at some point it's going to have to drop or skip data which might require more than one ais message, but still should eventually work. What do you think?