Recent Posts

Pages: 1 2 [3] 4 5 ... 10
21
AIS Dispatcher / Does aisdispatcher on Linux work?
« Last post by beantaxi on 11 February, 2017, 08:09:18 am »
Hi all,

AISHub sounds great. I read about aisdispatcher, and tried to fire it up very simply on Linux:

./aisdispatcher -G -t -h data.aishub.net -p 1234 -H localhost:5000

aisdispatcher just tries to reconnect every 10 seconds with no activity
tcpdump shows no activity from data.aishub.net
socat shows no data coming in on port 5000

It looks like absolutely nothing is happening on data.aishub.net:1234. Should I be doing something differently?

Thanks!
22
AIS Data Sharing / Re: Experimental project for historical AIS data
« Last post by ilkertemir on 03 February, 2017, 09:44:31 pm »
Took one step further and developed an interface that allows downloading Historical AIS Data automatically: https://www.partmarine.com/historical_ais_data
23
Software development / Re: Message 5 encoding - where to split?
« Last post by coecles on 27 January, 2017, 05:08:23 pm »
Dave, 

Using any Linux device to capture the messages allows one to readily parse and timestamp the receipt of the messages.

The device clock should be synced with one the global time servers to avoid drift.  By parsing upon receipt, we can also discard any malformed messages early, including those with invalid checksums.

We extract the MMSI and message type, add a timestamp, message source, and archive the messages to S3 in compressed format. 

For our own receivers, the timestamp would be accurate to within milliseconds of the message receipt.  From AISHub, the messages are downsampled, so the timestamp should be accurate to within about a minute.

Flat files can be rapidly processed to select archived messages by date, vessel, and type and then reparse the AIVDM content.

Of course, more fields could be extracted, such as lat-lon, to allow simple message selection options without reparsing.

We modify the millisecond field by bumping by a millisecond as necessary to ensure separate message groups have unique ascending timestamps to preserve their sequence in analytics processing.

Example:

!AIVDM,2,1,6,A,5=WU1rD1n4pT<4hiuE<l58U@D<R2Q84@Tv0<vT0k2P5227;l=<S@DRE1@j;P,0*74
2016-12-05T17:44:27.093Z aishub   05 913916393 !AIVDM,2,2,6,A,kkKUSP00000,2*3F
2016-12-05T17:44:27.121Z aishub   01 367598950 !AIVDM,1,1,,A,15NTMIPP00I<WqPA1usWEgvn0>`<,0*76
2016-12-05T17:44:27.122Z aishub   01 316030917 !AIVDM,1,1,,B,14eHwi@009o8hB>LD=plvlr80>`<,0*02
2016-12-05T17:44:27.123Z aishub   21 993692027 !AIVDM,1,1,,A,ENkb9NqR@@@@@@@@@@@@@@@@@@@;Wquu:l9k800003vP000,2*36

For analytics with Spark and Kafka, we process these files, extract fields of interest, do some calculations, and create new files in CSV format:

id,timestamp,messageType,source,latitude,longitude,distance,duration,speed,speedOverGround,heading,courseOverGround,course,motion,duplicate
279202354,2016-12-18T04:24:22.103Z,02,aishub,44.71559833,20.99782667,0,0,0,0,0,155.1,0,U,N
211590050,2016-12-18T04:24:22.106Z,01,aishub,53.52658333,8.57730333,3.2,60.7,0.1,0,0,292.2,329.3,S,N
413823141,2016-12-18T04:24:22.111Z,18,aishub,31.36492167,121.49740667,200.2,101.4,3.8,3.4,166,166.0,170.8,I,N
413697380,2016-12-18T04:24:22.115Z,19,aishub,25.98303333,119.43211,9.3,198.6,0.1,0,8,8.5,347.6,S,N

Gary
24
Software development / Re: Message 5 encoding - where to split?
« Last post by malcolmh on 26 January, 2017, 02:06:09 pm »
Messages 4 & 11 both contain full UTC date & time.
25
AIS Data Sharing / Re: Data Sharing Formats
« Last post by ilkertemir on 21 January, 2017, 10:43:28 pm »
Hello, I checked your website, this looks like a cool project! I like your idea. While I can't comment on the AISHub policy, I can offer using it for an archival project I am running.

I run the AIS station 2404 also run a little AIS history project on my website PartMarine (https://www.partmarine.com/) to archive historical AIS data. I would be interested in leveraging your AIS data to include in that archive.

For instance here is an example for historical location for Greenpeace's Rainbow Warrior:
https://www.partmarine.com/vessel/mmsi/244163000.

This comes from AISHub colllecters. VentureFarther's dataset could enhance the archive.
26
AIS Data Sharing / can someone please help me ?!?
« Last post by ben on 10 January, 2017, 06:03:25 pm »
hello everyone,
can please someone share some live data, please ?
I just want to display one single ship on a map and talking maybe about 100 requests a month.
please message me if you can help
ben
27
Software development / Re: Message 5 encoding - where to split?
« Last post by Dave61 on 10 January, 2017, 09:37:02 am »
Thanks very much for the reply Malcolm. I got it working now, splitting the message just before the destination field. So far it seems to work well with 166 vessels successfully sending their Msg05 data.

I just wish AIS data had a full timestamp instead of only seconds as there is no real way to send or keep 'old data' in AIVDM format!  :)

Cheers,

Dave
28
Software development / Re: Message 5 encoding - where to split?
« Last post by coecles on 06 January, 2017, 05:10:09 am »
Dave -

Curious - what would be your goal in feeding the messages to AISHub?

Gary
29
Software development / Re: Message 5 encoding - where to split?
« Last post by malcolmh on 05 January, 2017, 10:51:22 am »
There is no rule for this in the standard, but the usual convention is to split the message on a six-bit boundary, such that the fill bit field is zero on all but the last sentence.
30
Software development / Message 5 encoding - where to split?
« Last post by Dave61 on 05 January, 2017, 06:39:39 am »
I am writing some software to encode vessel position data from a database & put it into AIS format so I can send it by UDP to OpenCPN and maybe feed it to other sources such as AISHub once I prove it is working.

I have Message 1,2,3 working fine so far - vessels are showing up on the map properly.
I am having trouble trying to build a message 5 (static data) as this kind of message is too long to fit in 1 message so it normally is split into 2 messages like this:

!AIVDM,2,1,3,B,55P5TL01VIaAL@7WKO@mBplU@<PDhh000000001S;AJ::4A80?4i@E53,0*3E
!AIVDM,2,2,3,B,1@0000000000000,2*55

I have my 424 bits ready to convert. But so far I can't figure out if there is a rule for WHERE I should split the binary data before converting it to NMEA and putting in 2 consecutive messages.
Does anyone know where the split should occur?
Pages: 1 2 [3] 4 5 ... 10