M17_Project
/
M17_spec
mirror of https://github.com/M17-Project/M17_spec.git
1
0
Fork 0
Code Issues Releases Wiki Activity

Merge branch 'QA' of https://github.com/M17-Project/M17_spec into QA

pull/102/head
GitSync 2022-04-19 11:51:02 +00:00
parent 52125559ce 933e260c13
commit e19d10537f
29 changed files with 1948 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
.github/workflows/convert-to-pdf.yml vendored 100644
View File

@ -0,0 +1,29 @@
name: Docs to PDF
# This workflow is triggered on pushes to the repository.
on:
push:
branches:
- main
# Paths can be used to only trigger actions when you have edited certain files, such as a file within the /pages directory
paths:
- 'pages/**'
jobs:
converttopdf:
name: Build PDF
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- uses: baileyjm02/markdown-to-pdf@v1
with:
input_dir: pages
output_dir: pdfs
images_dir: pages
# for example <img src="./images/file-name.png">
# image_import: ./images
# Default is true, can set to false to only get PDF files
build_html: false
- uses: actions/upload-artifact@v1
with:
name: pages
path: pdfs

339
LICENSE 100644
View File

@ -0,0 +1,339 @@
GNU GENERAL PUBLIC LICENSE
Version 2, June 1991
Copyright (C) 1989, 1991 Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
Preamble
The licenses for most software are designed to take away your
freedom to share and change it. By contrast, the GNU General Public
License is intended to guarantee your freedom to share and change free
software--to make sure the software is free for all its users. This
General Public License applies to most of the Free Software
Foundation's software and to any other program whose authors commit to
using it. (Some other Free Software Foundation software is covered by
the GNU Lesser General Public License instead.) You can apply it to
your programs, too.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.
To protect your rights, we need to make restrictions that forbid
anyone to deny you these rights or to ask you to surrender the rights.
These restrictions translate to certain responsibilities for you if you
distribute copies of the software, or if you modify it.
For example, if you distribute copies of such a program, whether
gratis or for a fee, you must give the recipients all the rights that
you have. You must make sure that they, too, receive or can get the
source code. And you must show them these terms so they know their
rights.
We protect your rights with two steps: (1) copyright the software, and
(2) offer you this license which gives you legal permission to copy,
distribute and/or modify the software.
Also, for each author's protection and ours, we want to make certain
that everyone understands that there is no warranty for this free
software. If the software is modified by someone else and passed on, we
want its recipients to know that what they have is not the original, so
that any problems introduced by others will not reflect on the original
authors' reputations.
Finally, any free program is threatened constantly by software
patents. We wish to avoid the danger that redistributors of a free
program will individually obtain patent licenses, in effect making the
program proprietary. To prevent this, we have made it clear that any
patent must be licensed for everyone's free use or not licensed at all.
The precise terms and conditions for copying, distribution and
modification follow.
GNU GENERAL PUBLIC LICENSE
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
0. This License applies to any program or other work which contains
a notice placed by the copyright holder saying it may be distributed
under the terms of this General Public License. The "Program", below,
refers to any such program or work, and a "work based on the Program"
means either the Program or any derivative work under copyright law:
that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another
language. (Hereinafter, translation is included without limitation in
the term "modification".) Each licensee is addressed as "you".
Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope. The act of
running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the
Program (independent of having been made by running the Program).
Whether that is true depends on what the Program does.
1. You may copy and distribute verbatim copies of the Program's
source code as you receive it, in any medium, provided that you
conspicuously and appropriately publish on each copy an appropriate
copyright notice and disclaimer of warranty; keep intact all the
notices that refer to this License and to the absence of any warranty;
and give any other recipients of the Program a copy of this License
along with the Program.
You may charge a fee for the physical act of transferring a copy, and
you may at your option offer warranty protection in exchange for a fee.
2. You may modify your copy or copies of the Program or any portion
of it, thus forming a work based on the Program, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions:
a) You must cause the modified files to carry prominent notices
stating that you changed the files and the date of any change.
b) You must cause any work that you distribute or publish, that in
whole or in part contains or is derived from the Program or any
part thereof, to be licensed as a whole at no charge to all third
parties under the terms of this License.
c) If the modified program normally reads commands interactively
when run, you must cause it, when started running for such
interactive use in the most ordinary way, to print or display an
announcement including an appropriate copyright notice and a
notice that there is no warranty (or else, saying that you provide
a warranty) and that users may redistribute the program under
these conditions, and telling the user how to view a copy of this
License. (Exception: if the Program itself is interactive but
does not normally print such an announcement, your work based on
the Program is not required to print an announcement.)
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote it.
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Program.
In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable
source code, which must be distributed under the terms of Sections
1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three
years, to give any third party, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)
The source code for a work means the preferred form of the work for
making modifications to it. For an executable work, complete source
code means all the source code for all modules it contains, plus any
associated interface definition files, plus the scripts used to
control compilation and installation of the executable. However, as a
special exception, the source code distributed need not include
anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the
operating system on which the executable runs, unless that component
itself accompanies the executable.
If distribution of executable or object code is made by offering
access to copy from a designated place, then offering equivalent
access to copy the source code from the same place counts as
distribution of the source code, even though third parties are not
compelled to copy the source along with the object code.
4. You may not copy, modify, sublicense, or distribute the Program
except as expressly provided under this License. Any attempt
otherwise to copy, modify, sublicense or distribute the Program is
void, and will automatically terminate your rights under this License.
However, parties who have received copies, or rights, from you under
this License will not have their licenses terminated so long as such
parties remain in full compliance.
5. You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Program or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Program (or any work based on the
Program), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Program or works based on it.
6. Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License.
7. If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Program at all. For example, if a patent
license would not permit royalty-free redistribution of the Program by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Program.
If any portion of this section is held invalid or unenforceable under
any particular circumstance, the balance of the section is intended to
apply and the section as a whole is intended to apply in other
circumstances.
It is not the purpose of this section to induce you to infringe any
patents or other property right claims or to contest validity of any
such claims; this section has the sole purpose of protecting the
integrity of the free software distribution system, which is
implemented by public license practices. Many people have made
generous contributions to the wide range of software distributed
through that system in reliance on consistent application of that
system; it is up to the author/donor to decide if he or she is willing
to distribute software through any other system and a licensee cannot
impose that choice.
This section is intended to make thoroughly clear what is believed to
be a consequence of the rest of this License.
8. If the distribution and/or use of the Program is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Program under this License
may add an explicit geographical distribution limitation excluding
those countries, so that distribution is permitted only in or among
countries not thus excluded. In such case, this License incorporates
the limitation as if written in the body of this License.
9. The Free Software Foundation may publish revised and/or new versions
of the General Public License from time to time. Such new versions will
be similar in spirit to the present version, but may differ in detail to
address new problems or concerns.
Each version is given a distinguishing version number. If the Program
specifies a version number of this License which applies to it and "any
later version", you have the option of following the terms and conditions
either of that version or of any later version published by the Free
Software Foundation. If the Program does not specify a version number of
this License, you may choose any version ever published by the Free Software
Foundation.
10. If you wish to incorporate parts of the Program into other free
programs whose distribution conditions are different, write to the author
to ask for permission. For software which is copyrighted by the Free
Software Foundation, write to the Free Software Foundation; we sometimes
make exceptions for this. Our decision will be guided by the two goals
of preserving the free status of all derivatives of our free software and
of promoting the sharing and reuse of software generally.
NO WARRANTY
11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
REPAIR OR CORRECTION.
12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.
END OF TERMS AND CONDITIONS
How to Apply These Terms to Your New Programs
If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these terms.
To do so, attach the following notices to the program. It is safest
to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least
the "copyright" line and a pointer to where the full notice is found.
<one line to give the program's name and a brief idea of what it does.>
Copyright (C) <year> <name of author>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
Also add information on how to contact you by electronic and paper mail.
If the program is interactive, make it output a short notice like this
when it starts in an interactive mode:
Gnomovision version 69, Copyright (C) year name of author
Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
This is free software, and you are welcome to redistribute it
under certain conditions; type `show c' for details.
The hypothetical commands `show w' and `show c' should show the appropriate
parts of the General Public License. Of course, the commands you use may
be called something other than `show w' and `show c'; they could even be
mouse-clicks or menu items--whatever suits your program.
You should also get your employer (if you work as a programmer) or your
school, if any, to sign a "copyright disclaimer" for the program, if
necessary. Here is a sample; alter the names:
Yoyodyne, Inc., hereby disclaims all copyright interest in the program
`Gnomovision' (which makes passes at compilers) written by James Hacker.
<signature of Ty Coon>, 1 April 1989
Ty Coon, President of Vice
This General Public License does not permit incorporating your program into
proprietary programs. If your program is a subroutine library, you may
consider it more useful to permit linking proprietary applications with the
library. If this is what you want to do, use the GNU Lesser General
Public License instead of this License.

8
README.md 100644
View File

@ -0,0 +1,8 @@
# M17_spec
Protocol specification for the [M17 Project](http://m17project.org/).
M17 is a modern, digital radio protocol built by hams, for hams.
It is still a work in progress, and this is the living specification of the protocol.
The specification can be viewed [here](https://spec.m17project.org/).

43
pages/01.welcome/default.md 100644
View File

@ -0,0 +1,43 @@
---
title: Welcome
---
## Part I - Air and IP Interface
* [M17 RF Protocol: Summary](/part-1/m17-rf-protocol-summary)
* [Glossary](/part-1/glossary)
* [Physical Layer](/part-1/physical-layer)
* [4FSK generation](/part-1/physical-layer#4fsk-generation)
* [Preamble](/part-1/physical-layer#preamble)
* [Bit types](/part-1/physical-layer#bit-types)
* [Error correction coding schemes and bit type conversion](/part-1/physical-layer#error-correction-coding-schemes-and-bit-type-conversion)
* [Data Link Layer](/part-1/data-link-layer)
* [Stream Mode](/part-1/data-link-layer#stream-mode)
* [Packet Mode](/part-1/data-link-layer#packet-mode)
* [Application Layer](/part-1/application-layer)
* [Packet Superframes](/part-1/application-layer#packet-superframes)
* [Encryption Types](/part-1/application-layer#encryption-types)
## Part II - Codeplug
* [Recommendation for the codeplug structure](/part-2/codeplug#recommendation-for-the-codeplug-structure)
* [Introduction](/part-2/codeplug#introduction)
* [Codeplug file structure](/part-2/codeplug#codeplug-file-structure)
* [Example Codeplug](/part-2/codeplug#example-codeplug)
## Appendix
* [Address Encoding](/appendix/address-encoding)
* [Callsign Encoding: base40](/appendix/address-encoding#callsign-encoding-base40)
* [Callsign Formats](/appendix/address-encoding#callsign-formats)
* [Decorrelator sequence](/appendix/decorrelator-sequence)
* [Interleaving](/appendix/interleaving)
* [M17 Internet Protocol (IP) Networking](/appendix/ip-networking)
* [Standard IP Framing](/appendix/ip-networking#standard-ip-framing)
* [Control Packets](/appendix/ip-networking#control-packets)
* [KISS Protocol](/appendix/kiss-protocol)
* [References](/appendix/kiss-protocol#references)
* [Glossary](/appendix/kiss-protocol#glossary)
* [M17 Protocols](/appendix/kiss-protocol#m17-protocols)
* [KISS Basics](/appendix/kiss-protocol#kiss-basics)
* [Packet Protocols](/appendix/kiss-protocol#packet-protocols)
* [Stream Protocol](/appendix/kiss-protocol#stream-protocol)
* [Mixing Modes](/appendix/kiss-protocol#mixing-modes)
* [Implementation Details](/appendix/kiss-protocol#implementation-details)

36
pages/02.part-1/01.m17-rf-protocol-summary/docs.md 100644
View File

@ -0,0 +1,36 @@
---
title: 'M17 RF Protocol: Summary'
taxonomy:
category:
- docs
---
M17 is an RF protocol that is:
* Completely open: open specification, open source code, open source hardware, open algorithms. Anyone must be able to build an M17 radio and interoperate with other M17 radios without having to pay anyone else for the right to do so.
* Optimized for amateur radio use.
* Simple to understand and implement.
* Capable of doing the things hams expect their digital protocols to do:
* Voice (eg: DMR, D-Star, etc)
* Point to point data (eg: Packet, D-Star, etc)
* Broadcast telemetry (eg: APRS, etc)
* Extensible, so more capabilities can be added over time.
To do this, the M17 protocol is broken down into three protocol layers, like a network:
1. Physical Layer: How to encode 1s and 0s into RF. Specifies RF modulation, symbol rates, bits per symbol, etc.
2. Data Link Layer: How to packetize those 1s and 0s into usable data. Packet vs Stream modes, headers, addressing, etc.
3. Application Layer: Accomplishing activities. Voice and data streams, control packets, beacons, etc.
This document attempts to document these layers.
.. [TETRA] Dunlop, John; Girma, Demessie; Irvine, James "Digital
Mobile Communications and the TETRA System" Wiley 1999,
ISBN: 9780471987925
.. [DMR] ETSI TS 102 361-1 V2.2.1 (2013-02): "Electromagnetic
compatibility and Radio spectrum Matters (ERM); Digital
Mobile Radio (DMR) Systems; Part 1: DMR Air Interface (AI)
protocol"
https://www.etsi.org/deliver/etsi_ts/102300_102399/10236101/02.02.01_60/ts_10236101v020201p.pdf

27
pages/02.part-1/02.glossary/docs.md 100644
View File

@ -0,0 +1,27 @@
---
title: Glossary
taxonomy:
category:
- docs
---
### ECC
Error Correcting Code
### FEC
Forward Error Correction
### Frame
The individual components of a stream, each of which contains payload data interleaved with frame signalling.
### Link Information Frame
The first frame of any transmission. It contains full LICH data.
### LICH
Link Information Channel. The LICH contains all information needed to establish an M17 link. The first frame of a transmission contains full LICH data, and subsequent frames each contain one sixth of the LICH data so that late-joiners can obtain the LICH.
### Packet
A single burst of transmitted data containing 100s to 1000s of bytes, after which the physical layer stops sending data.
### Superframe
A set of six consecutive frames which collectively contain full LICH data are grouped into a superframe.

1
pages/02.part-1/03.physical-layer/convolutional.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 20 KiB

274
pages/02.part-1/03.physical-layer/docs.md 100644
View File

@ -0,0 +1,274 @@
---
title: 'Physical Layer'
taxonomy:
category:
- docs
simple-responsive-tables:
active: true
media_order: 'convolutional.svg,frame_encoding.svg,link_setup_frame_encoding.svg,packet_frame_encoding.svg'
---
### 4FSK generation
M17 standard uses 4FSK modulation running at 4800 symbols/s (9600
bits/s) with a deviation index h=0.33 for transmission in 9 kHz
channel bandwidth. Channel spacing is 12.5 kHz. The symbol stream is
converted (upsampled) to a series of impulses which pass through a
root-raised-cosine (alpha=0.5) shaping filter before frequency modulation
at the transmitter and again after frequency demodulation at the
receiver.
[mermaid]
graph LR
id1[Dibits Input] --> id2[Upsampler] --> id3[RRC Filter] --> id4[Frequency Modulation] --> id5[4FSK Output]
style id1 fill:#ffffffff,stroke:#ffffffff,stroke-width:0px
style id2 fill:#fff,stroke:#000,stroke-width:2px
style id3 fill:#fff,stroke:#000,stroke-width:2px
style id4 fill:#fff,stroke:#000,stroke-width:2px
style id5 fill:#ffffffff,stroke:#ffffffff,stroke-width:0px
[/mermaid]
The bit-to-symbol mapping is shown in the table below.
<table>
<caption><span>Table 1 </span><span>Dibit symbol mapping to 4FSK deviation</span></caption>
<thead>
<tr>
<th colspan="2">Information bits</th>
<th rowspan="2">Symbol</th>
<th rowspan="2">4FSK deviation</th>
</tr>
<tr>
<th>Bit 1</th>
<th>Bit 0</th>
</tr>
</thead>
<tbody>
<tr>
<td>0</td>
<td>1</td>
<td>+3</td>
<td>+2.4 kHz</td>
</tr>
<tr>
<td>0</td>
<td>0</td>
<td>+1</td>
<td>+0.8 kHz</td>
</tr>
<tr>
<td>1</td>
<td>0</td>
<td>-1</td>
<td>-0.8 kHz</td>
</tr>
<tr>
<td>1</td>
<td>1</td>
<td>-3</td>
<td>-2.4 kHz</td>
</tr>
</tbody>
</table>
The most significant bits are sent first, meaning that the byte 0xB4 (= 0b10'11'01'00) in type 4 bits (see [bit types](#bit-types)) would be sent as the symbols -1 -3 +3 +1. All data fields utilize big-endian order of bytes unless specified otherwise.
### Preamble
Every transmission starts with a preamble, which shall consist of at least 40 ms of alternating outer symbols. This is equivalent to 40 milliseconds of a 2400 Hz tone. The last symbol transmitted within the preamble shall be -3 for all modes except BERT. This is to avoid unnecessary long constant symbol runs and increase zero-crossing rate.
### Bit types
The bits at different stages of the error correction coding are referred to with bit types, given in the following table.
Type | Description
---- | -----------
Type 1 | Data link layer bits
Type 2 | Bits after appropriate encoding
Type 3 | Bits after puncturing (only for convolutionally coded data, for other ECC schemes type 3 bits are the same as type 2 bits)
Type 4 | Decorrelated and interleaved (re-ordered) type 3 bits
Type 4 bits are used for transmission over the RF. Incoming type 4 bits shall be decoded to type 1 bits, which are then used to extract all the frame fields.
### Error correction coding schemes and bit type conversion
Two distinct ECC/FEC schemes are used for different parts of the transmission.
#### Link setup frame (LSF)
![link_setup_frame_encoding](link_setup_frame_encoding.svg?classes=caption "ECC Link Setup Frame Encoding")
240 DST, SRC, TYPE, META and CRC type 1 bits are convolutionally coded using rate 1/2 coder with constraint K=5. 4 tail bits are used to flush the encoder’s state register, giving a total of 244 bits being encoded. Resulting 488 type 2 bits are retained for type 3 bits computation. Type 3 bits are computed by puncturing type 2 bits using a scheme shown in chapter 4.4. This results in 368 bits, which in conjunction with the synchronization burst gives 384 bits (384 bits / 9600bps = 40 ms).
Interleaving type 3 bits produce type 4 bits that are ready to be transmitted. Interleaving is used to combat error bursts.
#### Subsequent frames
![frame_encoding](frame_encoding.svg?classes=caption "ECC stages of subsequent frames")
A 40-bit (type 1) chunk of the LSF along with a 3-bit modulo 6 counter (LICH_CNT) and 5 reserved bits (see Table 7) is partitioned into 4 12-bit parts and encoded using Golay (24, 12) code. This produces 96 encoded bits of type 2. These bits are used in the Link Information Channel (LICH).
16-bit FN and 128 bits of payload (144 bits total) are convolutionally encoded in a manner analogous to that of the link setup frame. A total of 148 bits is being encoded resulting in 296 type 2 bits. These bits are punctured to generate 272 type 3 bits.
96 type 2 bits of LICH are concatenated with 272 type 3 bits and re-ordered to form type 4 bits for transmission. This, along with 16-bit sync in the beginning of frame, gives a total of 384 bits
The LICH chunks allow for late listening and indepedent decoding to check destination address. The goal is to require less complexity to decode just the LICH and check if the full message should be decoded.
#### Packet Frame Encoding
![packet_frame_encoding](packet_frame_encoding.svg?classes=caption "Packet Frame Encoding")
#### End of Transmission marker (EoT)
Every transmission ends with a distinct symbol stream, which shall consist of at least 40 ms of repeating +3, +3, +3, +3, +3, +3, -3, +3 pattern, or 0x555D in hexadecimal notation.
#### Extended Golay(24,12) code
The extended Golay(24,12) encoder uses generating polynomial g given below to generate the 11 check bits. The check bits and an additional parity bit are appended to the 12 bit data, resulting in a 24 bit codeword. The resulting code is systematic, meaning that the input data (message) is embedded in the codeword.
\(g(x) = x^{11} + x^{10} + x^6 + x^5 + x^4 + x^2 + 1\)
This is equivalent to 0xC75 in hexadecimal notation. Both the generating matrix \(G\) and parity check matrix \(H\) are shown below.
\(
\begin{align}
G = \begin{bmatrix} I_k | P \end{bmatrix} = & \begin{bmatrix}
1&0&0&0&0&0&0&0&0&0&0&0&1&1&0&0&0&1&1&1&0&1&0&1\\
0&1&0&0&0&0&0&0&0&0&0&0&0&1&1&0&0&0&1&1&1&0&1&1\\
0&0&1&0&0&0&0&0&0&0&0&0&1&1&1&1&0&1&1&0&1&0&0&0\\
0&0&0&1&0&0&0&0&0&0&0&0&0&1&1&1&1&0&1&1&0&1&0&0\\
0&0&0&0&1&0&0&0&0&0&0&0&0&0&1&1&1&1&0&1&1&0&1&0\\
0&0&0&0&0&1&0&0&0&0&0&0&1&1&0&1&1&0&0&1&1&0&0&1\\
0&0&0&0&0&0&1&0&0&0&0&0&0&1&1&0&1&1&0&0&1&1&0&1\\
0&0&0&0&0&0&0&1&0&0&0&0&0&0&1&1&0&1&1&0&0&1&1&1\\
0&0&0&0&0&0&0&0&1&0&0&0&1&1&0&1&1&1&0&0&0&1&1&0\\
0&0&0&0&0&0&0&0&0&1&0&0&1&0&1&0&1&0&0&1&0&1&1&1\\
0&0&0&0&0&0&0&0&0&0&1&0&1&0&0&1&0&0&1&1&1&1&1&0\\
0&0&0&0&0&0&0&0&0&0&0&1&1&0&0&0&1&1&1&0&1&0&1&1\\
\end{bmatrix}
\newline\newline
H = \begin{bmatrix} P^T | I_k \end{bmatrix} = & \begin{bmatrix}
1&0&1&0&0&1&0&0&1&1&1&1&1&0&0&0&0&0&0&0&0&0&0&0\\
1&1&1&1&0&1&1&0&1&0&0&0&0&1&0&0&0&0&0&0&0&0&0&0\\
0&1&1&1&1&0&1&1&0&1&0&0&0&0&1&0&0&0&0&0&0&0&0&0\\
0&0&1&1&1&1&0&1&1&0&1&0&0&0&0&1&0&0&0&0&0&0&0&0\\
0&0&0&1&1&1&1&0&1&1&0&1&0&0&0&0&1&0&0&0&0&0&0&0\\
1&0&1&0&1&0&1&1&1&0&0&1&0&0&0&0&0&1&0&0&0&0&0&0\\
1&1&1&1&0&0&0&1&0&0&1&1&0&0&0&0&0&0&1&0&0&0&0&0\\
1&1&0&1&1&1&0&0&0&1&1&0&0&0&0&0&0&0&0&1&0&0&0&0\\
0&1&1&0&1&1&1&0&0&0&1&1&0&0&0&0&0&0&0&0&1&0&0&0\\
1&0&0&1&0&0&1&1&1&1&1&0&0&0&0&0&0&0&0&0&0&1&0&0\\
0&1&0&0&1&0&0&1&1&1&1&1&0&0&0&0&0&0&0&0&0&0&1&0\\
1&1&0&0&0&1&1&1&0&1&0&1&0&0&0&0&0&0&0&0&0&0&0&1\\
\end{bmatrix}
\end{align}
\)
The output of the Golay encoder is shown in the table below.
Field | Data | Check bits | Parity
----- | ---- | ---------- | ------
Position | 23..12 | 11..1 | 0 (LSB)
Length | 12 | 11 | 1
Four of these 24-bit blocks are used to reconstruct the LSF.
Sample MATLAB/Octave code snippet for generating \(G\) and \(H\) matrices is shown below.
```
P = hex2poly('0xC75');
[H,G] = cyclgen(23, P);
G_P = G(1:12, 1:11);
I_K = eye(12);
G = [I_K G_P P.'];
H = [transpose([G_P P.']) I_K];
```
### Convolutional encoder
The convolutional code shall encode the input bit sequence after appending 4 tail bits at the end of the sequence. Rate of the coder is R=½ with constraint length K=5. The encoder diagram and generating polynomials are shown below.
\(
\begin{align}
G_1(D) =& 1 + D^3 + D^4 \\
G_2(D) =& 1+ D + D^2 + D^4
\end{align}
\)
The output from the encoder must be read alternately.
![convolutional](convolutional.svg?classes=caption "Convolutional coder diagram")
### Code puncturing
Removing some of the bits from the convolutional coder’s output is called code puncturing. The nominal coding rate of the encoder used in M17 is ½. This means the encoder outputs two bits for every bit of the input data stream. To get other (higher) coding rates, a puncturing scheme has to be used.
Two different puncturing schemes are used in M17 stream mode:
1. \(P_1\) leaving 46 from 61 encoded bits
2. \(P_2\) leaving 11 from 12 encoded bits
Scheme \(P_1\) is used for the *link setup frame*, taking 488 bits of encoded data and selecting 368 bits. The \(gcd(368, 488)\) is 8 which, when used to divide, leaves 46 and 61 bits. However, a full puncture pattern requires the puncturing matrix entries count to be divisible by the number of encoding polynomials. For this case a partial puncture matrix is used. It has 61 entries with 46 of them being ones and shall be used 8 times, repeatedly. The construction of the partial puncturing pattern \(P_1\) is as follows:
\(
\begin{align}
M = & \begin{bmatrix}
1 & 0 & 1 & 1
\end{bmatrix} \\
P_{1} = & \begin{bmatrix}
1 & M_{1} & \cdots & M_{15}
\end{bmatrix}
\end{align}
\)
In which \(M\) is a standard 2/3 rate puncture matrix and is used 15 times, along with a leading \(1\) to form \(P_1\), an array of length 61.
The first pass of the partial puncturer discards \(G_1\) bits only, second pass discards \(G_2\), third - \(G_1\) again, and so on. This ensures that both bits are punctured out evenly.
Scheme \(P_2\) is for frames (excluding LICH chunks, which are coded differently). This takes 296 encoded bits and selects 272 of them. Every 12th bit is being punctured out, leaving 272 bits. The full matrix shall have 12 entries with 11 being ones.
The puncturing scheme \(P_2\) is defined by its partial puncturing matrix:
\(
\begin{align}
P_2 = & \begin{bmatrix}
1 & 1 & 1 & 1 & 1 & 1 \\
1 & 1 & 1 & 1 & 1 & 0
\end{bmatrix}
\end{align}
\)
The linearized representations are:
```
P1 = [1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1,
1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1,
0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1]
P2 = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0]
```
One additional puncturing scheme \(P_3\) is used in the packet mode. The puncturing scheme is defined by its puncturing matrix:
\(
\begin{align}
P_3 = & \begin{bmatrix}
1 & 1 & 1 & 1 \\
1 & 1 & 1 & 0
\end{bmatrix}
\end{align}
\)
The linearized representation is:
```
P3 = [1, 1, 1, 1, 1, 1, 1, 0]
```
### Interleaving
For interleaving a Quadratic Permutation Polynomial (QPP) is used. The polynomial \(\pi(x)=(45x+92x^2)\mod 368\) is used for a 368 bit interleaving pattern QPP. See appendix sec-interleaver for pattern.
To avoid transmitting long sequences of constant symbols (e.g. 010101…), a simple algorithm is used. All 46 bytes of type 4 bits shall be XORed with a pseudorandom, predefined stream. The same algorithm has to be used for incoming bits at the receiver to get the original data stream. See decorr-seq for sequence.

1
pages/02.part-1/03.physical-layer/frame_encoding.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 16 KiB

1
pages/02.part-1/03.physical-layer/link_setup_frame_encoding.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 14 KiB

1
pages/02.part-1/03.physical-layer/packet_frame_encoding.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 12 KiB

BIN
pages/02.part-1/04.data-link-layer/M17_stream.png 100644
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 59 KiB

257
pages/02.part-1/04.data-link-layer/docs.md 100644
View File

@ -0,0 +1,257 @@
---
title: 'Data Link Layer'
taxonomy:
category:
- docs
media_order: M17_stream.png
---
The Data Link layer is split into three modes:
* Packet mode
Data are sent in small bursts, on the order of 100s to 1000s of bytes at a time, after which the physical layer stops sending data. e.g. messages, beacons, etc.
* Stream mode
Data are sent in a continuous stream for an indefinite amount of time, with no break in physical layer output, until the stream ends. e.g. voice data, bulk data transfers, etc.
* BERT mode
PRBS9 is used to fill frames with a deterministic bit sequence. Frames are sent in a continuous sequence.
When the physical layer is idle (no RF being transmitted or received), the data link defaults to packet mode.
As is the convention with other networking protocols, all values are
encoded in big endian byte order.
### Stream Mode
In Stream Mode, an *indefinite* amount of payload data is sent continuously without breaks in the physical layer. The *stream* is broken up into parts, called *frames* to not confuse them with *packets* sent in packet mode. Frames contain payload data interleaved with frame signalling (similar to packets). Frame signalling is contained within the **Link Information Channel (LICH)**.
#### Sync Burst
All frames are preceded by a 16-bit *synchronization burst*.
* Link setup frames shall be preceded with 0x55F7.
* Stream frames shall be preceeded with 0xFF5D.
* Packet frames shall be preceeded with 0x75FF.
* BERT frames shall be preceeded with 0xDF55.
All syncwords are type 4 bits.
These sync words are based on [Barker codes](https://en.wikipedia.org/wiki/Barker_code).
#### Link setup frame
First frame of the transmission contains full LSF data. It’s called the **Link Setup Frame (LSF)**, and is not part of any superframes.
##### LSF Fields
Field | Length | Description
----- | ------ | -----------
DST | 48 bits | Destination address - Encoded callsign or a special number (eg. a group)
SRC | 48 bits | Source address - Encoded callsign of the originator or a special number (eg. a group)
TYPE | 16 bits | Information about the incoming data stream
META | 112 bits | Metadata field, suitable for cryptographic metadata like IVs or single-use numbers, or non-crypto metadata like the sender’s GNSS position.
CRC | 16 bits | CRC for the link setup data
TAIL | 4 bits | Flushing bits for the convolutional encoder that do not carry any information. Only included for RF frames, not included for IP purposes.
##### Bitfields of Type Field
Bits | Meaning
---- | -------
0 | Packet/stream indicator, 0=packet, 1=stream
1..2 | Data type indicator, $01_2$ =data (D), $10_2$ =voice (V), $11_2$ =V+D, $00_2$ =reserved
3..4 | Encryption type, $00_2$ =none, $01_2$ =AES, $10_2$ =scrambling, $11_2$ =other/reserved
5..6 | Encryption subtype (meaning of values depends on encryption type)
7..10 | Channel Access Number (CAN)
11..15 | Reserved (don’t care)
The fields in Table 3 (except TAIL) form initial LSF. It contains all information needed to establish M17 link. Later in the transmission, the initial LSF is divided into 6 “chunks” and transmitted beside the payload data. This allows late-joiners to reconstruct the LICH after collecting all the pieces, and start decoding the stream even though they missed the beginning of the transmission. The process of collecting full LSF takes 6 frames or 6\*40 ms = 240 ms. Four TAIL bits are needed for the convolutional coder to go back to state 0, so the ending trellis position is also known.
Voice coder rate is inferred from TYPE field, bits 1 and 2.
##### Voice Coder Rates
Data Type Indicator | Voice Coder Rate
------------------- | ----------------
$00_2$ | none / reserved
$01_2$ | no voice
$10_2$ | 3200 bps
$11_2$ | 1600 bps
#### Subsequent Frames
##### Fields for Frames other than LSF
Field | Length | Description
----- | ------ | -----------
LICH | 48 bits | LSF chunk, one of 6
FN | 16 bits | Frame number, starts from 0 and increments every frame to a max of 0x7fff where it will then wrap back to 0. High bit set indicates this frame is the last of the stream
PAYLOAD | 128 bits | Payload/data, can contain arbitrary data
TAIL | 4 bits | Flushing bits for the convolutional encoder that don’t carry any information
The most significant bit in the FN counter is used for transmission end signalling. When transmitting the last frame, it shall be set to 1 (one), and 0 (zero) in all other frames.
The payload is used so that earlier data in the voice stream is sent first. For mixed voice and data payloads, the voice data is stored first, then the data.
##### LSF Chunk Structure
Bits | Content
---- | -------
0..39 | 40 bits of full LSF
40..42 | A modulo 6 counter (LICH_CNT) for LSF re-assembly
43..47 | Reserved
##### Payload Example 1
`Codec2 encoded frame t + 0 | Codec2 encoded frame t + 1`
##### Payload Example 2
`Codec2 encoded frame t + 0 | Mixed data t + 0`
#### Superframes
Each frame contains a chunk of the LSF frame that was used to establish the stream. Frames are grouped into superframes, which is the group of 6 frames that contain everything needed to rebuild the original LSF packet, so that the user who starts listening in the middle of a stream (late-joiner) is eventually able to reconstruct the LSF message and understand how to receive the in-progress stream.
![M17_stream](M17_stream.png?classes=caption "Stream consisting of one superframe")
[mermaid]
graph TD
c0["conv. coder"]
p0["P_1 puncturer"]
i0["interleaver"]
w0["decorrelator"]
s0["prepend LSF_SYNC"]
l0["LICH combiner"]
chunker_40["chunk 40 bits"]
golay_24_12["Golay (24, 12)"]
c1["conv. coder"]
p1["P_2 puncturer"]
i1["interleaver"]
w1["decorrelator"]
s1["prepend FRAME_SYNC"]
fn["add FN"]
chunker_128["chunk 128 bits"]
framecomb["Frame Combiner"]
supercomb["Superframe Combiner"]
counter --> l0
LSF --> c0 --> p0 --> i0 --> w0 --> s0 --> supercomb
LSF --> chunker_40 --> l0 --> golay_24_12 --> framecomb
data --> chunker_128 --> fn --> c1 --> p1 --> framecomb
framecomb --> i1 --> w1 --> s1 --> supercomb
preamble --> supercomb
[/mermaid]
<center>An overview of the forward dataflow</center>
#### CRC
M17 uses a non-standard version of 16-bit CRC with polynomial $x^{16} + x^{14} + x^{12} + x^{11} + x^8 + x^5 + x^4 + x^2 + 1$ or 0x5935 and initial value of 0xFFFF. This polynomial allows for detecting all errors up to hamming distance of 5 with payloads up to 241 bits, which is less than the amount of data in each frame.
As M17’s native bit order is most significant bit first, neither the input nor the output of the CRC algorithm gets reflected.
The input to the CRC algorithm consists of DST, SRC (each 48 bits), 16 bits of TYPE field and 112 bits META, and then depending on whether the CRC is being computed or verified either 16 zero bits or the received CRC.
The test vectors in the following table are calculated by feeding the given message and then 16 zero bits to the CRC algorithm.
Message | CRC Output
------- | ----------
(empty string) | 0xFFFF
ASCII string "A" | 0x206E
ASCII string "123456789" | 0x772B
Bytes 0x00 to 0xFF | 0x1c31
### Packet Mode
In *packet mode*, a finite amount of payload data (for example – text messages or application layer data) is wrapped with a packet, sent over the physical layer, and is completed when done. ~~Any acknowledgement or retransmission is done at the application layer.~~
#### Link Setup Frame
Packet mode uses the same link setup frame that has been defined for stream mode above. The packet/stream indicator is set to 0 in the type field.
##### Bitfields of Type Field
Bits | Meaning
---- | -------
0 | Packet/stream indicator, 0=packet, 1=stream
1..2 | Data type indicator, $01_2$ =data (D), $10_2$ =voice (V), $11_2$ =V+D, $00_2$ =reserved
3..4 | Encryption type, $00_2$ =none, $01_2$ =AES, $10_2$ =scrambling, $11_2$ =other/reserved
5..6 | Encryption subtype (meaning of values depends on encryption type)
7..10 | Channel Access Number (CAN)
11..15 | Reserved (don’t care)
Raw packet frames have no packet type metadata associated with them. Encapsulated packet format is discussed in Packet Superframes in the Application Layer section. This provides data type information and is the preferred format for use on M17.
When encryption type is $00_2$, meaning no encryption, the encryption subtype bits are used to indicate the contents of the META field in the LSF. Since that space would otherwise go be unused, we can store small bits of data in that field such as free text or the sender’s GNSS position.
Encryption type and subtype bits, including the plaintext data formats when not using encryption, are described in more detail in the Application Layer section of this document.
Currently the contents of the source and destination fields are arbitrary as no behavior is defined which depends on the content of these fields. The only requirement is that the content is base-40 encoded.
#### Packet Format
M17 packet mode can transmit up to 798 bytes of payload data within one transmission. It acheives a base throughput of 5 kbps, and a net throughput of about 4.7 kbps for the largest data payload, and over 3 kbps for 100-byte payloads. [^1]
[^1]: Net throughput takes into account preamble and link setup overhead.
The packet superframe consists of 798 payload data bytes and a 2-byte CCITT CRC-16 checksum.
##### Byte Fields of Packet Superframe
Bytes | Meaning
----- | -------
1..798 | Packet payload
2 | CRC-16
The CRC used here is the same as described in [Chapter 2.4](https://spec.m17project.org/part-1/data-link-layer#crc).
Packet data is split into frames of 368 type 4 bits preceded by a packet-specific 16-bit sync word (0xFF5D). This is the same size frame used by stream mode.
The packet frame starts with a 210 bit frame of type 1 data. It is noteworthy that it does not terminate on a byte boundary.
The frame has 200 bits (25 bytes) of payload data, 6 bits of frame metadata, and 4 bits to flush the convolutional coder.
##### Bit Fields of Packet Frame
Bits | Meaning
---- | -------
0..199 | Packet payload
1 | EOF indicator
5 | Frame / Byte count
4 | Flush bits for convolutional coder
The metadata field contains a 1 bit **end of frame (EOF)** indicator, and a 5-bit frame/byte counter.
The **EOF** bit is 1 only on the last frame. The **counter** field is used to indicate the frame number when **EOF** is 0, and the number of bytes in the last frame when **EOF** is 1. This encodes the exact packet size, up to 800 bytes, in a 6-bit field.
##### Metadata Field with EOF = 0
Bits | Meaning
---- | -------
0 | Set to 0, Not end of frame
1..5 | Frame number, 0..31
##### Metadata Field with EOF = 1
Bits | Meaning
---- | -------
0 | Set to 1, End of frame
1..5 | Number of bytes in frame, 1..25
Note that it is non-conforming to send a last frame with a length of 0 bytes. The number of bytes **includes** 2-byte CRC.
#### Convolutional Coding
The entire frame is convolutionally coded, giving 420 bits of type 2 data. It is then punctured using a 7/8 puncture matrix (1,1,1,1,1,1,1,0) to give 368 type 3 bits. These are then interleaved and decorrelated to give 368 type 4 bits.
##### Packet Frame
Bits | Meaning
---- | -------
16 bits | Sync word 0xFF5D
368 bits | Payload
#### Carrier-sense Multiple Access
When sending packets, the sender is reponsible for ensuring the channel is clear before transmitting. CSMA is used to minimize collisions on a shared network. Specifically, P-persistent access is used. Each time slot is 40ms (one packet length) and the probability SHOULD default to 25%. In terms of the values used by the KISS protocol, these equate to a slot time of 4 and a P-persistence value of 63.
The benefit of this method is that it imposes no penalty on uncontested networks.

1
pages/02.part-1/05.application-layer/LFSR_16.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 22 KiB

1
pages/02.part-1/05.application-layer/LFSR_24.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 20 KiB

1
pages/02.part-1/05.application-layer/LFSR_8.svg 100644
View File

File diff suppressed because one or more lines are too long
Side by Side

After

Width:  |  Height:  |  Size: 18 KiB

6
pages/02.part-1/05.application-layer/application-layer.bib 100644
View File

@ -0,0 +1,6 @@
@BOOK{McGrew2002,
title = {Counter mode security: Analysis and recommendations},
author = {McGrew, David A.},
publisher = {Cisco Systems},
edition = {4.},
year = {2002}}

112
pages/02.part-1/05.application-layer/docs.md 100644
View File

@ -0,0 +1,112 @@
---
title: 'Application Layer'
taxonomy:
category:
- docs
media_order: 'LFSR_8.svg,LFSR_16.svg,LFSR_24.svg'
---
### Packet Superframes
Packet superframes are composed of a 1..n byte data type specifier, 0..797 bytes of payload data. The data type specifier is encoded in the same way as UTF-8. It provides efficient coding of common data types. And it can be extended to include a very large number of distinct packet data type codes.
The data type specifier can also be used as a protocol specifier. For example, the following protocol identifers are reserved in the M17 packet spec:
##### Reserved Protocols
Identifier | Protocol
---------- | --------
0x00 | RAW
0x01 | AX.25
0x02 | APRS
0x03 | 6LoWPAN
0x04 | IPv4
0x05 | SMS
0x06 | Winlink
The data type specifier is used to compute the CRC, along with the payload.
### Encryption Types
Encryption is **optional**. The use of it may be restricted within some radio services and countries, and should only be used if legally permissible.
#### Null Encryption
Encryption type = $00_2$
The “Encryption SubType” bits in the Stream Type field then indicate what data is stored in the 112 bits of the LSF META field.
Encryption SubType bits | LSF META data contents
----------------------- | ----------------------
$00_2$ | UTF-8 Text
$01_2$ | GNSS Position Data
$10_2$ | Reserved
$11_2$ | Reserved
All LSF META data must be stored in big endian byte order, as throughout the rest of this specification.
GNSS Position Data stores the 112 bit META field as follows:
Size in bits | Format | Contents
------------ | ------ | --------
8 | 8-bit signed integer | Latitude - degrees, integer part (-90..+90, positive values for northern hemisphere)
16 | 16-bit unsigned integer | Latitude - degrees, fractional part (eg. 0.5 -> 32,768)
16 | 16-bit unsigned integer | Longitude - degrees, integer part (-180..+180, positive values for eastern hemisphere)
16 | 16-bit unsigned integer | Longitude Longitude - degrees, fractional part (eg. 0.5 -> 32,768)
16 | unsigned integer | Altitude, in feet MSL. Stored +1500, so a stored value of 0 represents -1500 MSL. Subtract 1500 feet when parsing.
10 | unsigned integer | Course in degrees true North
10 | unsigned integer | Speed in miles per hour
20 | reserved values | Transmitter/Object description field
#### Scrambler
Encryption type = $01_2$
Scrambling is an encryption by bit inversion using a bitwise exclusive-or (XOR) operation between bit sequence of data and pseudorandom bit sequence.
Encrypting bitstream is generated using a Fibonacci-topology Linear-Feedback Shift Register (LFSR). Three different LFSR sizes are available: 8, 16 and 24-bit. Each shift register has an associated polynomial. The polynomials are listed in Table 7. The LFSR is initialised with a seed value of the same length as the shift register. Seed value acts as an encryption key for the scrambler algorithm. Figures 5 to 8 show block diagrams of the algorithm
Encryption subtype | LFSR polynomial | Seed length | Sequence period
------------------ | --------------- | ----------- | ---------------
$00_2$ | $x^8 + x^6 + x^5 + x^4 + 1$ | 8 bits | 255
$01_2$ | $x^{16} + x^{15} + x^{13} + x^4 + 1$ | 16 bits | 65,535
$10_2$ | $x^{24} + x^{23} + x^{22} + x^{17} + 1$ | 24 bits | 16,777,215
![LSFR_8](LFSR_8.svg?classes=caption "8-bit LSFR taps")
---
![LSFR_16](LFSR_16.svg?classes=caption "16-bit LSFR taps")
---
![LSFR_24](LFSR_24.svg?classes=caption "24-bit LSFR taps")
---
#### Advanced Encryption Standard (AES)
Encryption type = $10_2$
This method uses AES block cipher in counter (CTR) mode, with a 96-bit nonce that should never be used for more than one separate stream and a 32 bit CTR.
The 96-bit AES nonce value is extracted from the 96 most significant bits of the META field, and the remaining 16 bits of the META field form the highest 16 bits of the 32 bit counter. The FN (Frame Number) field value is then used to fill out the lower 16 bits of the counter, and always starts from 0 (zero) in a new voice stream.
The 16 bit frame number and 40 ms frames can provide for over 20 minutes of streaming without rolling over the counter.
> The effective capacity of the counter is 15 bits, as the MSB is used for transmission end signalling. At 40ms per frame, or 25 frames per second, and 2\*\*15 frames, we get 2\*\*15 frames / 25 frames per second = 1310 seconds, or 21 minutes and some change.
The random part of the nonce value should be generated with a hardware random number generator or any other method of generating non-repeating values.
To combat replay attacks, a 32-bit timestamp shall be embedded into the cryptographic nonce field. The field structure of the 96 bit nonce is shown in Table 9. Timestamp is 32 LSB portion of the number of seconds that elapsed since the beginning of 1970-01-01, 00:00:00 UTC, minus leap seconds (a.k.a. “unix time”).
##### 96 bit nonce field structure
| Timestamp | Random Data | CTR_HIGH |
| --------- | ----------- | -------- |
| 32 | 64 | 16 |
**CTR_HIGH** field initializes the highest 16 bits of the CTR, with the rest of the counter being equal to the FN counter.
!! In CTR mode, AES encryption is malleable. That is, an attacker can change the contents of the encrypted message without decrypting it. This means that recipients of AES-encrypted data must not trust that the data is authentic. Users who require that received messages are proven to be exactly as-sent by the sender should add application-layer authentication, such as HMAC. In the future, use of a different mode, such as Galois/Counter Mode, could alleviate this issue.

9
pages/02.part-1/chapter.md 100644
View File

@ -0,0 +1,9 @@
---
title: 'Part I - Air and IP Interface'
taxonomy:
category: docs
---
### Part I
# Air and IP Interface

9
pages/03.part-2/chapter.md 100644
View File

@ -0,0 +1,9 @@
---
title: 'Part II - Codeplug'
taxonomy:
category: docs
---
### Part II
# Codeplug

100
pages/03.part-2/codeplug/docs.md 100644
View File

@ -0,0 +1,100 @@
---
title: Codeplug
taxonomy:
category:
- docs
---
# Recommendation for the codeplug structure
## Introduction
Codeplugs are ordinary text files with .m17 extension. They provide an information on:
* channel banks
* channel frequencies
* destination IDs
* transmission mode
* payload type
* encryption mode
Codeplugs should be human-readable and easily editable with common text editors.
## Codeplug file structure
We recommend using YAML for the codeplug files.
### Keywords
```
codeplug:
author:
String - Codeplug author, max 16 characters
version:
Date and time in YYYY-MM-DDTHH:MM:SS format
bank:
name:
String - Channel bank name, 16 characters maximum
channel:
name:
String - Channel name, 16 characters maximum
descr:
String - Channel Description, 16 characters maximum
freq_rx:
Integer - Channel RX Frequency in Hz
freq_tx:
Integer - Channel TX Frequency in Hz
mode:
Integer - Channel mode. Valid modes are: 0 - Analog,
1 - Digital Voice, 2 - Digital Data, 3 - Digital Voice and Data
encr:
Integer - Is encryption enabled? 0 for no encryption,
1 - AES256, 2 - scrambler etc. (refer to M17_spec for details)
nonce:
String - 14-byte hex value without leading 0x. nonce for
ciphers or initial LFSR value for scrambler
gps:
Boolean - If true, and mode value enables digital data,
gps data will be transferred along with payload
```
## Example Codeplug
```yaml
codeplug:
author: SP5WWP
version: 2020-28-09T13:20:49
- bank:
name: M17
- channel:
name: M17_DMO
descr:
freq_rx: 439575000
freq_tx: 439575000
mode: 2
encr: 0
nonce: 0
gps: false
- channel:
name: M17_DMO_2
descr:
freq_rx: 439975000
freq_tx: 439975000
mode: 2
encr: 0
nonce: 0
gps: false
- bank:
name: Repeaters
- channel:
name: SR5MS
descr:
freq_rx: 439425000
freq_tx: 431825000
mode: 2
encr: 0
nonce: 0
gps: false
#codeplug end
```

142
pages/04.appendix/01.address-encoding/docs.md 100644
View File

@ -0,0 +1,142 @@
---
title: 'Address Encoding'
taxonomy:
category:
- docs
---
M17 uses 48 bits (6 bytes) long addresses. Callsigns (and other addresses) are encoded into these 6 bytes in the following ways:
* An address of 0 is invalid.
* Address values between 1 and 262143999999999 (which is 409−1), up to 9 characters of text are encoded using base40, described below.
* Address values between 262144000000000 (409) and 281474976710654 (248−2) are invalid
* An address of 0xFFFFFFFFFFFF is a broadcast. All stations should receive and listen to this message.
##### Address Scheme
Address Range | Category | Number of Addresses | Remarks
------------- | -------- | ------------------- | -------
0x000000000000 | RESERVED | 1 | For future use
0x000000000001 - 0xee6b27ffffff | Unit ID | 262143999999999 |
0xee6b28000000 - 0xfffffffffffe | RESERVED | 19330976710655 | For future use
0xffffffffffff | Broadcast | 1 | Valid only for destination
## Callsign Encoding: base40
9 characters from an alphabet of 40 possible characters can be encoded into 48 bits, 6 bytes. The base40 alphabet is:
* 0: A space. Invalid characters will be replaced with this.
* 1-26: “A” through “Z”
* 27-36: “0” through “9”
* 37: “-” (hyphen)
* 38: “/” (slash)
* 39: “.” (dot)
Encoding is little endian. That is, the right most characters in the encoded string are the most significant bits in the resulting encoding.
#### Example code: encode_base40()
```c
uint64_t encode_callsign_base40(const char *callsign) {
uint64_t encoded = 0;
for (const char *p = (callsign + strlen(callsign) - 1); p >= callsign; p-- ) {
encoded *= 40;
// If speed is more important than code space,
// you can replace this with a lookup into a 256 byte array.
if (*p >= 'A' && *p <= 'Z') // 1-26
encoded += *p - 'A' + 1;
else if (*p >= '0' && *p <= '9') // 27-36
encoded += *p - '0' + 27;
else if (*p == '-') // 37
encoded += 37;
// These are just place holders. If other characters make more sense,
// change these. Be sure to change them in the decode array below too.
else if (*p == '/') // 38
encoded += 38;
else if (*p == '.') // 39
encoded += 39;
else
// Invalid character or space, represented by 0, decoded as a space.
//encoded += 0;
}
return encoded;
}
```
#### Example code: decode_base40()
```c
char *decode_callsign_base40(uint64_t encoded, char *callsign) {
if (encoded >= 262144000000000) { // 40^9
*callsign = 0;
return callsign;
}
char *p = callsign;
for (; encoded > 0; p++) {
*p = " ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-/."[encoded % 40];
encoded /= 40;
}
*p = 0;
return callsign;
}
```
#### Why base40?
The longest commonly assigned callsign from the FCC is 6 characters. The minimum alphabet of A-Z, 0-9, and a “done” character mean the most compact encoding of an American callsign could be: $log2(37^6)=31.26$ bits, or 4 bytes.
Some countries use longer callsigns, and the US sometimes issues longer special event callsigns. Also, we want to extend our callsigns (see below). So we want more than 6 characters. How many bits do we need to represent more characters:
##### Bits per Characters
Characters | Bits | Bytes
---------- | ---- | -----
7 | $log2(37^7)=36.47$ | 5
8 | $log2(37^8)=41.67$ | 6
9 | $log2(37^9)=46.89$ | 6
10 | $log2(37^{10})=52.09$ | 7
Of these, 9 characters into 6 bytes seems the sweet spot. Given 9 characters, how large can we make the alphabet without using more than 6 bytes?
##### Alphabet Size vs. Bytes
Alphabet Size | Bits | Bytes
------------- | ---- | -----
37 | $log2(37^9)=46.89$ | 6
38 | $log2(38^9)=47.23$ | 6
39 | $log2(39^9)=47.57$ | 6
40 | $log2(40^9)=47.90$ | 6
41 | $log2(41^9)=48.22$ | 7
Given this, 9 characters from an alphabet of 40 possible characters, makes maximal use of 6 bytes.
## Callsign Formats
Government issued callsigns should be able to encode directly with no changes.
### Multiple Stations
To allow for multiple stations by the same operator, we borrow the use of the ‘-’ character from AX.25 and the SSID field. A callsign such as “AB1CD-1” is considered a different station than “AB1CD-2” or even “AB1CD”, but it is understood that these all belong to the same operator, “AB1CD”
### Temporary Modifiers
Similarly, suffixes are often added to callsign to indicate temporary changes of status, such as “AB1CD/M” for a mobile station, or “AB1CD/AE” to signify that I have Amateur Extra operating privileges even though the FCC database may not yet be updated. So the ‘/’ is included in the base40 alphabet. The difference between ‘-’ and ‘/’ is that ‘-’ are considered different stations, but ‘/’ are NOT. They are considered to be a temporary modification to the same station.
### Interoperability
It may be desirable to bridge information between M17 and other networks. The 9 character base40 encoding allows for this:
#### DMR
DMR unfortunately doesn’t have a guaranteed single name space. Individual IDs are reasonably well recognized to be managed by [RadioID.net](https://www.radioid.net/database/search#!) but Talk Groups are much less well managed. Talk Group XYZ on Brandmeister may be (and often is) different than Talk Group XYZ on a private cBridge system.
* DMR IDs are encoded as: D<number> eg: D3106728 for KR6ZY
* DMR Talk Groups are encoded by their network. Currently, the following networks are defined:
* Brandmeister: BM<number> eg: BM31075
* DMRPlus: DP<number> eg: DP262
* More networks to be defined here.
#### D-Star
D-Star reflectors have well defined names: REFxxxY which are encoded directly into base40.

32
pages/04.appendix/02.decorrelator-sequence/docs.md 100644
View File

@ -0,0 +1,32 @@
---
title: 'Decorrelator Sequence'
taxonomy:
category:
- docs
---
Seq. number | Value | Seq. number | Value
----------- | ----- | ----------- | -----
00 | 0xD6 | 23 | 0x6E
01 | 0xB5 | 24 | 0x68
02 | 0xE2 | 25 | 0x2F
03 | 0x30 | 26 | 0x35
04 | 0x82 | 27 | 0xDA
05 | 0xFF | 28 | 0x14
06 | 0x84 | 29 | 0xEA
07 | 0x62 | 30 | 0xCD
08 | 0xBA | 31 | 0x76
09 | 0x4E | 32 | 0x19
10 | 0x96 | 33 | 0x8D
11 | 0x90 | 34 | 0xD5
12 | 0xD8 | 35 | 0x80
13 | 0x98 | 36 | 0xD1
14 | 0xDD | 37 | 0x33
15 | 0x5D | 38 | 0x87
16 | 0x0C | 39 | 0x13
17 | 0xC8 | 40 | 0x57
18 | 0x52 | 41 | 0x18
19 | 0x43 | 42 | 0x2D
20 | 0x91 | 43 | 0x29
21 | 0x1D | 44 | 0x78
22 | 0xF8 | 45 | 0xC3

101
pages/04.appendix/03.interleaving/docs.md 100644
View File

@ -0,0 +1,101 @@
---
title: Interleaving
taxonomy:
category:
- docs
---
input index | output index | input index | output index | input index | output index | input index | output index
----------- | ------------ | ----------- | ------------ | ----------- | ------------ | ----------- | ------------
0 | 0 | 92 | 92 | 184 | 184 | 276 | 276
1 | 137 | 93 | 229 | 185 | 321 | 277 | 45
2 | 90 | 94 | 182 | 186 | 274 | 278 | 366
3 | 227 | 95 | 319 | 187 | 43 | 279 | 135
4 | 180 | 96 | 272 | 188 | 364 | 280 | 88
5 | 317 | 97 | 41 | 189 | 133 | 281 | 225
6 | 270 | 98 | 362 | 190 | 86 | 282 | 178
7 | 39 | 99 | 131 | 191 | 223 | 283 | 315
8 | 360 | 100 | 84 | 192 | 176 | 284 | 268
9 | 129 | 101 | 221 | 193 | 313 | 285 | 37
10 | 82 | 102 | 174 | 194 | 266 | 286 | 358
11 | 219 | 103 | 311 | 195 | 35 | 287 | 127
12 | 172 | 104 | 264 | 196 | 356 | 288 | 80
13 | 309 | 105 | 33 | 197 | 125 | 289 | 217
14 | 262 | 106 | 354 | 198 | 78 | 290 | 170
15 | 31 | 107 | 123 | 199 | 215 | 291 | 307
16 | 352 | 108 | 76 | 200 | 168 | 292 | 260
17 | 121 | 109 | 213 | 201 | 305 | 293 | 29
18 | 74 | 110 | 166 | 202 | 258 | 294 | 350
19 | 211 | 111 | 303 | 203 | 27 | 295 | 119
20 | 164 | 112 | 256 | 204 | 348 | 296 | 72
21 | 301 | 113 | 25 | 205 | 117 | 297 | 209
22 | 254 | 114 | 346 | 206 | 70 | 298 | 162
23 | 23 | 115 | 115 | 207 | 207 | 299 | 299
24 | 344 | 116 | 68 | 208 | 160 | 300 | 252
25 | 113 | 117 | 205 | 209 | 297 | 301 | 21
26 | 66 | 118 | 158 | 210 | 250 | 302 | 342
27 | 203 | 119 | 295 | 211 | 19 | 303 | 111
28 | 156 | 120 | 248 | 212 | 340 | 304 | 64
29 | 293 | 121 | 17 | 213 | 109 | 305 | 201
30 | 246 | 122 | 338 | 214 | 62 | 306 | 154
31 | 15 | 123 | 107 | 215 | 199 | 307 | 291
32 | 336 | 124 | 60 | 216 | 152 | 308 | 244
33 | 105 | 125 | 197 | 217 | 289 | 309 | 13
34 | 58 | 126 | 150 | 218 | 242 | 310 | 334
35 | 195 | 127 | 287 | 219 | 11 | 311 | 103
36 | 148 | 128 | 240 | 220 | 332 | 312 | 56
37 | 285 | 129 | 9 | 221 | 101 | 313 | 193
38 | 238 | 130 | 330 | 222 | 54 | 314 | 146
39 | 7 | 131 | 99 | 223 | 191 | 315 | 283
40 | 328 | 132 | 52 | 224 | 144 | 316 | 236
41 | 97 | 133 | 189 | 225 | 281 | 317 | 5
42 | 50 | 134 | 142 | 226 | 234 | 318 | 326
43 | 187 | 135 | 279 | 227 | 3 | 319 | 95
44 | 140 | 136 | 232 | 228 | 324 | 320 | 48
45 | 277 | 137 | 1 | 229 | 93 | 321 | 185
46 | 230 | 138 | 322 | 230 | 46 | 322 | 138
47 | 367 | 139 | 91 | 231 | 183 | 323 | 275
48 | 320 | 140 | 44 | 232 | 136 | 324 | 228
49 | 89 | 141 | 181 | 233 | 273 | 325 | 365
50 | 42 | 142 | 134 | 234 | 226 | 326 | 318
51 | 179 | 143 | 271 | 235 | 363 | 327 | 87
52 | 132 | 144 | 224 | 236 | 316 | 328 | 40
53 | 269 | 145 | 361 | 237 | 85 | 329 | 177
54 | 222 | 146 | 314 | 238 | 38 | 330 | 130
55 | 359 | 147 | 83 | 239 | 175 | 331 | 267
56 | 312 | 148 | 36 | 240 | 128 | 332 | 220
57 | 81 | 149 | 173 | 241 | 265 | 333 | 357
58 | 34 | 150 | 126 | 242 | 218 | 334 | 310
59 | 171 | 151 | 263 | 243 | 355 | 335 | 79
60 | 124 | 152 | 216 | 244 | 308 | 336 | 32
61 | 261 | 153 | 353 | 245 | 77 | 337 | 169
62 | 214 | 154 | 306 | 246 | 30 | 338 | 122
63 | 351 | 155 | 75 | 247 | 167 | 339 | 259
64 | 304 | 156 | 28 | 248 | 120 | 340 | 212
65 | 73 | 157 | 165 | 249 | 257 | 341 | 349
66 | 26 | 158 | 118 | 250 | 210 | 342 | 302
67 | 163 | 159 | 255 | 251 | 347 | 343 | 71
68 | 116 | 160 | 208 | 252 | 300 | 344 | 24
69 | 253 | 161 | 345 | 253 | 69 | 345 | 161
70 | 206 | 162 | 298 | 254 | 22 | 346 | 114
71 | 343 | 163 | 67 | 255 | 159 | 347 | 251
72 | 296 | 164 | 20 | 256 | 112 | 348 | 204
73 | 65 | 165 | 157 | 257 | 249 | 349 | 341
74 | 18 | 166 | 110 | 258 | 202 | 350 | 294
75 | 155 | 167 | 247 | 259 | 339 | 351 | 63
76 | 108 | 168 | 200 | 260 | 292 | 352 | 16
77 | 245 | 169 | 337 | 261 | 61 | 353 | 153
78 | 198 | 170 | 290 | 262 | 14 | 354 | 106
79 | 335 | 171 | 59 | 263 | 151 | 355 | 243
80 | 288 | 172 | 12 | 264 | 104 | 356 | 196
81 | 57 | 173 | 149 | 265 | 241 | 357 | 333
82 | 10 | 174 | 102 | 266 | 194 | 358 | 286
83 | 147 | 175 | 239 | 267 | 331 | 359 | 55
84 | 100 | 176 | 192 | 268 | 284 | 360 | 8
85 | 237 | 177 | 329 | 269 | 53 | 361 | 145
86 | 190 | 178 | 282 | 270 | 6 | 362 | 98
87 | 327 | 179 | 51 | 271 | 143 | 363 | 235
88 | 280 | 180 | 4 | 272 | 96 | 364 | 188
89 | 49 | 181 | 141 | 273 | 233 | 365 | 325
90 | 2 | 182 | 94 | 274 | 186 | 366 | 278
91 | 139 | 183 | 231 | 275 | 323 | 367 | 47

99
pages/04.appendix/04.ip-networking/docs.md 100644
View File

@ -0,0 +1,99 @@
---
title: 'IP Networking'
taxonomy:
category:
- docs
---
Digital modes are commonly networked together through linked repeaters using IP networking.
For commercial protocols like DMR, this is meant for linking metropolitan and state networks together and allows for easy interoperability between radio users. Amateur Radio uses this capability for creating global communications networks for all imaginable purposes, and makes ‘working the world’ with an HT possible.
M17 is designed with this use in mind, and has native IP framing to support it.
In competing radio protocols, a repeater or some other RF to IP bridge is required for linking, leading to the use of hotspots (tiny simplex RF bridges).
The TR-9 and other M17 radios may support IP networking directly, such as through the ubiquitous ESP8266 chip or similar. This allows them to skip the RF link that current hotspot systems require, finally bringing to fruition the “Amateur digital radio is just VoIP” dystopian future we were all warned about.
## Standard IP Framing
M17 over IP is big endian, consistent with other IP protocols. We have standardized on UDP port 17000, this port is recommended but not required. Later specifications may require this port.
##### Internet Frame Fields
Field | Size | Description
----- | ---- | -----------
MAGIC | 32 bits | Magic bytes 0x4d313720 (“M17 “)
StreamID (SID) | 16 bits | Random bits, changed for each PTT or stream, but consistent from frame to frame within a stream
LICH | 224 bits | The meaningful contents of a LICH frame (dst, src, streamtype, META field) as defined earlier.
FN | 16 bits | Frame number (exactly as would be transmitted as an RF stream frame, including the last frame indicator at (FN & 0x8000)
Payload | 128 bits | Payload (exactly as would be transmitted in an RF stream frame)
CRC16 | 16 bits | CRC for the entire packet, as defined earlier [CRC definition](https://spec.m17project.org/part-1/data-link-layer#crc)
The CRC checksum must be recomputed after modification or re-assembly of the packet, such as when translating from RF to IP framing.
## Control Packets
Reflectors use a few different types of control frames, identified by their magic:
* CONN - Connect to a reflector
* ACKN - acknowledge connection
* NACK - deny connection
* PING - keepalive for the connection from the reflector to the client
* PONG - keepalive response from the client to the reflector
* DISC - Disconnect (client->reflector or reflector->client)
#### CONN
##### Bytes of CONN Packet
Bytes | Purpose
----- | -------
0..3 | Magic - ASCII "CONN"
4..9 | 6-byte ‘From’ callsign including module in last character (e.g. “A1BCD D”) encoded as per Address Encoding
10 | Module to connect to - single ASCII byte A-Z
A client sends this to a reflector to initiate a connection. The reflector replies with ACKN on successful linking, or NACK on failure.
#### ACKN
##### Bytes of ACKN Packet
Bytes | Purpose
----- | -------
0..3 | Magic - ASCII "ACKN"
#### NACK
##### Bytes of NACK Packet
Bytes | Purpose
----- | -------
0..3 | Magic - ASCII "NACK"
#### PING
##### Bytes of PING Packet
Bytes | Purpose
----- | -------
0..3 | Magic - ASCII "PING"
4..9 | 6-byte ‘From’ callsign including module in last character (e.g. “A1BCD D”) encoded as per Address Encoding
#### PONG
##### Bytes of PONG Packet
Bytes | Purpose
----- | -------
0..3 | Magic - ASCII "PONG"
4..9 | 6-byte ‘From’ callsign including module in last character (e.g. “A1BCD D”) encoded as per Address Encoding
#### DISC
##### Bytes of DISC Packet
Bytes | Purpose
----- | -------
0..3 | Magic - ASCII "DISC"
4..9 | 6-byte ‘From’ callsign including module in last character (e.g. “A1BCD D”) encoded as per Address Encoding

215
pages/04.appendix/05.kiss-protocol/docs.md 100644
View File

@ -0,0 +1,215 @@
---
title: 'KISS Protocol'
taxonomy:
category:
- docs
---
The purpose of this appendix is to document conventions for adapting KISS TNCs to M17 packet and streaming modes. M17 is a more complex protocol, both at the baseband level and at the data link layer than is typical for HDLC-based protocols commonly used on KISS TNCs. However, it is well suited for modern packet data links, and can even be used to stream digital audio between a host and a radio.
This appendix assumes the reader is familiar with the streaming and packet modes defined in the M17 spec, and with KISS TNCs and the KISS protocol.
In all cases, the TNC expects to get the data payload to be sent and is responsible for frame construction, FEC encoding, puncturing, interleaving and decorrelation. It is also responsible for baseband modulation.
For streaming modes, all voice encoding (Codec2) is done on the host and not on the TNC. The host is also responsible for constructing the LICH.
### References
* [http://www.ax25.net/kiss.aspx](http://www.ax25.net/kiss.aspx)
* [https://packet-radio.net/wp-content/uploads/2017/04/multi-kiss.pdf](https://packet-radio.net/wp-content/uploads/2017/04/multi-kiss.pdf)
* [https://en.wikipedia.org/wiki/OSI_model](https://en.wikipedia.org/wiki/OSI_model)
### Glossary
#### TNC
Terminal node controller -- a baseband network interface device to allow host computers to send data over a radio network, similar to a modem. It connects a computer to a radio and handles the baseband portion of the physical layer and the data link layer of network protocol stack.
#### KISS
Short for "Keep it simple, stupid". A simplified TNC protocol designed to move everything except for the physical layer and the data link layer out of the TNC. Early TNCs could include everything up through the application layer of the OSI network model.
#### SLIP
[Serial Line Internet Protocol](https://en.wikipedia.org/wiki/Serial_Line_Internet_Protocol) -- the base protocol used by the KISS protocol, extended by adding a single **type indicator** byte at the start of a frame.
#### type indicator
A one byte code at the beginning of a KISS frame which indicates the TNC **port** and KISS **command**.
#### port
A logical port on a TNC. This allowed a single TNC to connect to multiple radios. Its specific use is loosely defined in the KISS spec. The high nibble of the KISS **type indicator**. Port 0xF is reserved.
#### command
A KISS command. This tells the TNC or host how to interpret the KISS frame contents. The low nibble of the KISS **type indicator**. Command 0xF is reserved.
#### CSMA
[Carrier-sense multiple access](https://en.wikipedia.org/wiki/Carrier-sense_multiple_access) -- a protocol used by network devices to minimize collisions on a shared communications channel.
#### HDLC
[High-Level Data Link Control](https://en.wikipedia.org/wiki/High-Level_Data_Link_Control) -- a data link layer framing protocol used in many AX.25 packet radio networks. Many existing protocol documents, including KISS, reference HDLC because of its ubiquity when the protocols were invented. However, HDLC is not a requirement for higher level protocols like KISS which are agnostic to the framing used at the data link layer.
#### EOS
End of stream -- an indicator bit in the frame number field of a stream data frame.
#### LICH
Link information channel -- a secondary data channel in the stream data frame containing supplemental information, including a copy of the link setup frame.
### M17 Protocols
This specification defines KISS TNC modes for M17 packet and streaming modes, allowing the KISS protocol to be used to send and receive M17 packet and voice data. Both are bidirectional. There are two packet modes defined. This is done to provide complete access to the M17 protocol while maintaining the greatest degree of backwards compatibility with existing packet applications.
These protocols map to specific KISS port. The host tells the TNC what type of data to transmit based on the port used in host to TNC transfers. And the TNC tells the host what data it has received by the port set on TNC to host transfers.
This document outlines first the two packet protocols, followed by the streaming protocol.
### KISS Basics
#### TX Delay
If a **KISS TX** delay $T_d$ greater than 0 is specified, the transmitter is keyed for $T_d ∗ 10ms$ with only a DC signal present.
The $T_d$ value should be adjusted to the minimum required by the transmitter in order to transmit the full preamble reliably.
Only a single 40ms preamble frame is ever sent.
!! A TX delay may be necessary because many radios require some time between when PTT is engaged and the transmitter can begin transmitting a modulated signal.
### Packet Protocols
In order to provide backward compatibility with the widest range of existing ham radio software, and to make use of features in the the M17 protocol itself, we will define two distinct packet interfaces BASIC and FULL.
The KISS protocol allows us to target specific modems using the port identifier in the control byte.
We first define basic packet mode as this is initially likely to be the most commonly used mode over KISS.
#### M17 Basic Packet Mode
Basic packet mode uses only the standard KISS protocol on TNC port 0. This is the default port for all TNCs. Packets are sent using command 0. Again, this is normal behavior for KISS client applications.
##### Sending Data
In basic mode, the TNC only expects to receive packets from the host, as it would for any other mode supported AFSK, G3RUH, etc.
If the TNC is configured for half-duplex, the TNC will do P-persistence CSMA using a 40ms slot time and obey the P value set via the KISS interface. CSMA is disabled in full-duplex mode.
The **TX Tail** value is deprecated and is ignored.
The TNC sends the preamble burst.
The TNC is responsible for constructing the link setup frame, identifying the content as a raw mode packet. The source field is an encoded TNC identifier, similar to the APRS TOCALL, but it can be an arbitrary text string up to 9 characters in length. The destination is set to the broadcast address.
In basic packet mode, it is expected that the sender callsign is embedded within the packet payload.
The TNC sends the link setup frame.
The TNC then computes the CRC for the full packet, splits the packet into data frames encode and modulate each frame back-to-back until the packet is completely transmitted.
If there is another packet to be sent, the preamble can be skipped and the TNC will construct the next link setup frame (it can re-use the same link setup frame as it does not change) and send the next set of packet frames.
##### Limitations
The KISS specification defines no limitation to the packet size allowed. Nor does it specify any means of returning error conditions back to the host. M17 packet protocol limits the raw packet payload size to 798 bytes. The TNC must drop any packets larger than this.
##### Receiving Data
When receiving M17 data, the TNC must receive and parse the link setup frame and verify that the following frames contain raw packet data.
The TNC is responsible for decoding each packet, assembling the packet from the sequence of frames received, and verifying the packet checksum. If the checksum is valid, the TNC transfers the packet, excluding the CRC to the host using **KISS port** 0.
#### M17 Full Packet Mode
The purpose of full packet mode is to provide access to the entire M17 packet protocol to the host. This allows the host to set the source and destination fields, filter received packets based on the content these fields, enable encryption, and send and receive type-coded frames.
Use M17 full packet mode by sending to **KISS port** 1. In this mode the host is responsible for sending both the link setup frame and the packet data. It does this by prepending the 30-byte link setup frame to the packet data, sending this to the TNC in a single KISS frame. The TNC uses the first 30 bytes as the link setup frame verbatim, then splits the remaining data into M17 packet frames.
As with basic mode, the TNC uses the **Duplex** setting to enable/disable CSMA, and uses the **P value** for CSMA, with a fixes slot time of “4” (40 ms).
##### Receiving Data
For TNC to host transfers, the same occurs. The TNC combines the link setup frame with the packet frame and sends both in one KISS frame to the host using **KISS port** 1.
### Stream Protocol
The streaming protocol is fairly trivial to describe. It is used by sending first a link setup frame followed by a stream of 26-byte data frames to KISS port 2.
#### Stream Format
##### M17 Kiss Stream Protocol
Frame Size | Contents
---------- | --------
30 | Link Setup Frame
26 | LICH + Payload
26 | LICH + Payload
... | ...
26 | LICH + Payload with EOS bit set
The host must not send any frame to any other KISS port while a stream is active (a frame with the EOS bit has not been sent).
It is a protocol violation to send anything other than a link setup frame with the stream mode bit set in the first field as the first frame in a stream transfer to KISS port 2. Any such frame is ignored.
It is a protocol violation to send anything to any other KISS port while a stream is active. If that happens the stream is terminated and the packet that caused the protocol violation is dropped.
#### Data Frames
The data frames contain a 6-byte (48-bit) LICH segment followed by a 20 byte payload segment consisting of frame number, 16-byte data payload and CRC. The TNC is responsible for parsing the frame number and detecting the end-of-stream bit to stop transmitting.
##### KISS Stream Data Frame
Frame Size | Contents
---------- | --------
6 | LICH (48 bits)
2 | Frame Number and EOS Flag
16 | Payload
2 | M17 CRC of frame number and payload
The TNC is responsible for FEC-encoding both the LICH the payload, as well as interleaving, decorrelation, and baseband modulation.
#### Timing Constraints
Streaming mode provides additional timing constraints on both host to TNC transfers and on TNC to host transfers. Payload frames must arrive every 40ms and must have a jitter below 40ms. In general, it is expected that the TNC has up to 2 frames buffered (buffering occurs while sending the preamble and link setup frames), it should be able to keep the transmit buffers filled with packet jitter of 40ms.
The TNC must stop transmitting if the transmit buffers are empty. The TNC communicates that it has stopped transmitting early (before seeing a frame with the end of stream indicator set) by sending an empty data frame to the host.
### TNC to Host Transfers
TNC to host transfers are similar in that the TNC first sends the 30-byte link setup frame received to the host, followed by a stream of 26-byte data frames as described above. These are sent using **KISS port** 2.
The TNC must send the link setup frame first. This means that tne TNC must be able to decode LICH segments and assemble a valid link setup frame before it sends the first data frame. The TNC will only send a link setup frame with a valid CRC to the host. After the link setup frame is sent, the TNC ignores the CRC and sends all valid frames (those received after a valid sync word) to the host. If the stream is lost before seeing an end-of-stream flag, the TNC sends a 0-byte data frame to indicate loss of signal.
The TNC must then re-acquire the signal by decoding a valid link setup frame from the LICH in order to resume sending to the host.
### Busy Channel Lockout
The TNC implements **busy channel lockout** by enabling half-duplex mode on the TNC, and disables **busy channel lockout** by enabling full-duplex mode. When busy channel lockout occurs, the TNC keeps the link setup frame and discards all data frames until the channel is available. It then sends the preamble, link setup frame, and starts sending the data frames as they are received.
Note: BCL will be apparent to a receiver as the first frame received after the link setup frame will not start with frame number 0.
#### Limitations
Information is lost by having the TNC decode the LICH. It is not possible to communicate to the host that the LICH bytes are known to be invalid.
Should we have the TNC signal the host by dropping known invalid LICH segments? The host can tell that the LICH is missing by looking at the frame size.
### Mixing Modes
An M17 KISS TNC need not keep track of state across distinct TNC ports. Packet transfers are sent one packet at a time. It is OK to send to port 0 and port 1 in subsequent transfers. It is also OK to send a packet followed immediately by a voice streams. As mentioned earlier, it is a protocol violation to sent a KISS frame to any other port while a stream is active. However, a packet can be sent immediately following a voice stream (after EOS is sent).
#### Back-to-back Transfers
The TNC is expected to detect back-to-back transfers from the host, even across different KISS ports, and suppress the generation of the preamble.
For example, a packet containing APRS data sent immediately on PTT key-up should be sent immediately after the EOS frame.
Back-to-back transfers are common for packet communication where the window size determines the number of unacknowledged frames which may be outstanding (unacknowledged). Packet applications will frequently send back-to-back packets (up to window size packets) before waiting for the remote end to send ACKs for each of the packets.
### Implementation Details
#### Polarity
One of the issues that must be addressed by the TNC designer, and one which the KISS protocol offers no ready solution for, is the issue of polarity.
A TNC must interface with a RF transceiver for a complete M17 physical layer implementation. RF transceivers may have different polarity for their TX and RX paths.
M17 defines that the +3 symbol is transmitted with a +2.4 kHz deviation (2.4 kHz above the carrier). **Normal polarity** in a transceiver results in a positive voltage driving the frequency higher and a lower voltage driving the frequency lower. **Reverse polarity** is the opposite. A higher voltage drives the frequency lower.
On the receive side the same issue exists. **Normal polarity** results in a positive voltage output when the received signal is above the carrier frequency. **Reverse polarity** results in a positive voltage when the frequency is below the carrier.
Just as with transmitter deviation levels and received signal levels, the polarity of the transmit and receive path must be adjustable on a 4-FSK modem. The way these adjustments are made to the TNC are not addressed by the KISS specification.

95
pages/04.appendix/06.file-formats/docs.md 100644
View File

@ -0,0 +1,95 @@
---
title: 'File Formats'
taxonomy:
category:
- docs
---
This appendix documents the file formats used for testing various M17 layers.
### Glossary
#### Bit numbering, Bit order, Most significant bit (MSB), Least significant bit (LSB)
[Bit numbering](https://en.wikipedia.org/wiki/Bit_numbering) is how bit positions are identified in a binary number. The least significant bit (LSB) is the bit position respresenting a value of 1. The most significant bit (MSB) is the bit position representing the highest value position. Bit order refers to the order in which bits are extracted from a binary number. This is important especially when sending binary values one bit at a time, or when constructing multiple-bit symbols. LSB first means the extraction happens from the least significant position first. MSB first means extraction happens from the most significant position first.
#### Deviation, Frequency Deviation
In this context, deviation how far from the center frequency a carrier is shifted. This can be positive or negative. For M17, the frequency deviation of the four symbols are shown in [Physical Layer](https://spec.m17project.org/part-1/physical-layer) Table 1.
#### Deviation Function (Transmit)
A function used to convert symbol values to frequency deviation in RF hardware. This can be used to set hardware registers, create voltages, etc. depending on the hardware used.
#### Deviation Function (Receive)
A function used to convert frequency deviation in RF hardware to symbol values. This can be used when reading hardware registers, sampling voltages, etc. depending on the hardware used.
#### Dibit
Two bits used to represent a symbol, as shown in [Physical Layer](https://spec.m17project.org/part-1/physical-layer) Table 1.
#### Endianness, Byte order, Big-endian (BE), Little-endian (LE)
[Endianness](https://en.wikipedia.org/wiki/Endianness) is the order of the bytes in a word of digital data. In this document, we will refer to big-endian (BE) and little-endian (LE).
BE means that the most significant byte of a word is at the lowest memory location, while LE means that the least significant byte is at the lowest memory location.
#### RF Sample Rate
The rate at which deviation values are updated. This will vary depending on the hardware. M17 test software commonly uses 48000 samples per second.
#### Root-raised-cosine (RRC) Filter
A filter used to in digital communications to help reduce intersymbol interference. The M17 [Physical Layer](https://spec.m17project.org/part-1/physical-layer) specifies a root-raised-cosine (RRC) filter with alpha = 0.5 [Root Raised Cosine](https://en.wikipedia.org/wiki/Root-raised-cosine_filter)
#### Symbol
An M17 [Physical Layer](https://spec.m17project.org/part-1/physical-layer) symbol of +3, +1, -1, and -3.
#### Symbol Rate
The rate at which new symbols are generated. For M17, this is 4800 symbols per second.
### File Extensions
Multiple files are used when testing the different elements of the M17 protocol. File extensions (the three characters after a period in a complete file name) are defined to standardize formats and usage.
Extension | Description | Data Format | Data Rate
--------- | ----------- | ----------- | ---------
aud | Mono audio | Signed 16-bit LE | 8000 samples per second
sym | M17 symbols | Signed 8-bit | 4800 symbols per second
bin | Packed M17 Dibits | MSB first, Unsigned 8-bit | 4800 symbols per second (1200 bytes per second)
rrc | RRC filtered and Scaled M17 symbols | Signed 16-bit LE | 48000 samples per second
dev | Deviation values | Varies | Varies
#### aud
Mono audio of signed 16-bit LE at a rate of 8000 samples per second. This is often referred to as a "raw" audio file and contains no embedded header information.
#### sym
M17 symbols (+3, +1, -1, -3) encoded as signed 8-bit values at rate of 4800 symbols per second.
#### bin
M17 symbols packed 2 bits per symbol (dibits), 4 symbols per byte (+3 = 01, +1 = 00, -1 = 10, -3 = 11) with the MSB first. These are unsigned 8-bit values at 4800 symbols per second, which is 4 symbols per byte at 1200 bytes per second.
#### rrc
RRC filtered and scaled M17 symbols. In order to generate a reasonable RRC waveform, the symbol rate (4800 symbols per second) is upsampled by a factor of 10 to an RRC sample rate of 48000 samples per second. Then the upsampled symbols are passed through the RRC filter. The output samples of the RRC filter are multiplied by 7168 to fit within a signed 16-bit LE representation (e.g. a +3 value would be +21504).
#### dev
Hardware specific deviation values. These would be obtained by passing RRC filtered values through a deviation function. Since these are device specific, it is recommended to use an underscore plus device type as part of the filename. For example, the Semtech SX1276 uses a deviation step size of 61 Hz per bit. An M17 1600 Hz frequency step is equivalent to an SX1276 deviation value change of 26. Since the SX1276 only accepts positive deviation steps, the deviation function for the SX1276 would be (rrc value + 3.0) x 13. The .dev file specific for the SX1276 would contain those values, and could have a name such as m17test_sx1276.dev
### Example file flows
These show the file types in order of processing for transmit and receive flows. Each "->" symbolizes processing required to move from one file type to the next.
#### Transmit
aud -> sym -> rrc -> dev
aud -> bin -> rrc -> dev
#### Receive
dev -> rrc -> sym -> aud
dev -> rrc -> bin -> aud
### To-Do
File formats for packet and voice + data streams.
### References
Bit numbering https://en.wikipedia.org/wiki/Bit_numbering
Endianness https://en.wikipedia.org/wiki/Endianness
M17 Physical Layer https://spec.m17project.org/part-1/physical-layer
Root Raised Cosine https://en.wikipedia.org/wiki/Root-raised-cosine_filter

7
pages/04.appendix/chapter.md 100644
View File

@ -0,0 +1,7 @@
---
title: Appendix
taxonomy:
category: docs
---
### Appendix

1
pdfs/.blank 100644
View File

@ -0,0 +1 @@