Beacon Ambiguity

Ideas, technical topics, help and discussion for ATV enthusiasts
Forum rules
This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.

Thank you
Post Reply
EA7KIR
Posts: 95
Joined: Tue Feb 09, 2021 3:38 pm

Beacon Ambiguity

Post by EA7KIR » Thu May 30, 2024 7:04 pm

A few weeks ago a user of my Q100 receiver noticed it was reporting the beacon audio codec to be MP3, but thought it should report MPA. I've been down a rabbit hole ever since...

According to the various screen shots found on the Internet, Ryde/Portsdown, Winterhill, OpenTuner and MiniTiouner all have their own opinion. They can't all be right, so I asked Thomas DG5NGI, because he designed and built the beacon encoder. Here's what he told me...

* The beacon is transmitting MPEG-1 Audio Layer 2 in MPEG-2 Transport Stream.

* The stream_type indicator being sent is "3", which is the correct value to indicate "MPEG-1 Audio".

* The correct spec is in ISO/IEC 13818-1, but as this is copyrighted and watermarked, he can't send a copy to me. However, the ATSC standards people did a great job copying this table into an Excel file and are sharing it online. Relevant info is under the "PMT Stream Type" tab. Download from here https://www.atsc.org/wp-content/uploads ... ev-66.xlsx

I find the document to be incomprehensible. So I'm asking which 3-Letter acronym would be correct. Should it be MP1, MP2, MP3, MPA, MPG or something else?
Michael EA7KIR

G4EWJ
Posts: 1456
Joined: Wed Feb 17, 2010 10:11 am

Re: Beacon Ambiguity

Post by G4EWJ » Fri May 31, 2024 6:18 pm

I'm still trying to work it out myself. It seems that MPA is a generic term for MPEG-1 or MPEG-2 audio on stream type 3, as indicated in the PMT (Program Map Table). To see exactly what it is, you have to look at the audio packets themselves. Maybe a 3 character abbreviation isn't possible if you want to be completely accurate.

Below is the Wireshark analysis of an audio packet from the QO-100 beacon, where it identifies it as MPEG-1, Layer2. The inset is from Wikipedia and shows the bit patterns that give the MPEG and Layer types. The 'C' in the highlighted 'FC' value at the bottom covers MPEG type ID, Layer and Protection.

Brian
.
mpeg-audio.png
mpeg-audio.png (216.15 KiB) Viewed 5051 times

EA7KIR
Posts: 95
Joined: Tue Feb 09, 2021 3:38 pm

Re: Beacon Ambiguity

Post by EA7KIR » Fri May 31, 2024 9:37 pm

Brian, this leads me to believe streaming Longmynd into Ffplay, with appropriate flags, will reveal which decompression is being used. I’m away from home for the next few days so will not be able to try it till I return.
Michael EA7KIR

G4EWJ
Posts: 1456
Joined: Wed Feb 17, 2010 10:11 am

Re: Beacon Ambiguity

Post by G4EWJ » Sat Jun 01, 2024 11:00 am

MPV reports the beacon audio as:

mp2(MP2(MPEG audio layer 2))

ffplay reports it as:

Stream #0:1[0x102]: Audio: mp2 (mp3float) ([3][0][0][0] / 0x0003), 48000 Hz, stereo, fltp, 64 kb/s

I don't know what mp3float indicates.

Brian

EA7KIR
Posts: 95
Joined: Tue Feb 09, 2021 3:38 pm

Re: Beacon Ambiguity

Post by EA7KIR » Mon Jun 03, 2024 8:41 pm

Here’s what we know so far…

THOMAS says: MPEG-1 Audio
WIRSHARK says: MPEG-1
MPV says: mp2(MP2(MPEG audio layer 2))
FFPLAY says: mp2 (mp3float)

So we may as well call it: SCRAMBLED EGGS
Michael EA7KIR

User avatar
ON1AVO
Posts: 41
Joined: Sun Nov 20, 2011 2:35 pm

Re: Beacon Ambiguity

Post by ON1AVO » Tue Jun 04, 2024 7:17 am

Michael,

Some additional info:
- Potplayer (sorry in Dutch)
- Dektec StreamXpert

73 de ON1AVO
Attachments
Potplayer
Potplayer
Schermafbeelding 2024-06-04 091253.png (14.38 KiB) Viewed 4629 times
Dektec
Dektec
Schermafbeelding 2024-06-04 090655.png (104.48 KiB) Viewed 4629 times
73 de ON1AVO
Freddy

EA7KIR
Posts: 95
Joined: Tue Feb 09, 2021 3:38 pm

Re: Beacon Ambiguity

Post by EA7KIR » Tue Jun 04, 2024 8:34 am

Thank you ON1AVO.

Here’s what we know so far…

THOMAS says: MPEG-1 Audio
WIRSHARK says: MPEG-1
MPV says: mp2(MP2(MPEG audio layer 2))
FFPLAY says: mp2 (mp3float)
POTPLAYER says: MPEG Audio (Ox50) - Ingebouwde FFmpeg Decoder(mp1float)
DEKTEC says: MPEG-1 Audio

However, DEKTEC reports Scrambled: No, but doesn't mention EGGS. DEKTEC also reports MP1-float, which conflicts with FFPLAY saying MP3-float.
Michael EA7KIR

G8GKQ
Site Admin
Posts: 2969
Joined: Sun Mar 22, 2009 1:21 pm

Re: Beacon Ambiguity

Post by G8GKQ » Tue Jun 04, 2024 7:29 pm

Hi Michael

I checked using VLC and TS Reader, both of which are programs which I trust.

VLC: MPEG Audio Layer 1/2 (mpga)
TS Reader: MPEG 1 Audio Layer 2

Given all the other responses (espcially Thomas's) I am confident that the beacon audio is MPEG 1 Audio Layer 2.

So the question is what is the correct abbreviation? Wikipedia suggests that it should be MP2. https://en.wikipedia.org/wiki/MPEG-1_Audio_Layer_II. This is in the same series as MP3 which is very widely used abbreviaton.

However, the decode indicator on the BATC receivers is based on the "Stream Type Indicator" which is 3. This is generic to MP1, MP2 and MP3, so hence I believe the use of MPA as the descriptor.

Dave

EA7KIR
Posts: 95
Joined: Tue Feb 09, 2021 3:38 pm

Re: Beacon Ambiguity

Post by EA7KIR » Wed Jun 05, 2024 9:16 am

Hi Dave,

I think your belief makes perfect sense. But if this is correct, there'll need to be changes to the codec map in Ryde, because MPA is currently defined as type 32.

Code: Select all

        codecmap = {
             2:rydeplayer.sources.common.CodecEnum.MP2,
             3:rydeplayer.sources.common.CodecEnum.MP3,
             4:rydeplayer.sources.common.CodecEnum.MP3,
            15:rydeplayer.sources.common.CodecEnum.AAC,
            16:rydeplayer.sources.common.CodecEnum.H263,
            27:rydeplayer.sources.common.CodecEnum.H264,
            32:rydeplayer.sources.common.CodecEnum.MPA,
            36:rydeplayer.sources.common.CodecEnum.H265,
            51:rydeplayer.sources.common.CodecEnum.H266,
            129:rydeplayer.sources.common.CodecEnum.AC3,
            }
Adding to the confusion while Thomas is saying MPEG-1 Audio (to which I think we all agree), I'm told Matthias DD1US is saying MPEG-4 ACC - probably a communication error, as I'm also told both were instrumental in building the beacon.
Michael EA7KIR

Post Reply

Return to “General ATV Discussion”