I received from my friend Nicola, and gladly publish it, his work on Russian-Ukrainian Datalink.
When
reading these notes, one should be aware that they have been pieced together
from numerous sources and then consolidated. Carefully controlled machine
translation from Russian and Ukrainian has been used. It should also be taken
into consideration that the workings of the protocols and the nature of the
application data they transport as well as the composition of the automated,
military systems they serve may not have been fully understood. Markings in
yellow indicate that translation is not available or that the given translation
is tentative or poorly understood. Russian and Ukrainian designations have mostly
been transliterated into Latin characters.Summing up: this is a work in
progress.
1.
Introduction and history
Russian-UkrainianDatalink
(RUSUKRDL) is a fictive designationwhich for the sake of simplicity is used in
these notes to designate a collection of data link and application layer
protocols used in air defense, fire controland automated battle field data networks
of the armed forces of the Russian Federation,Ukraine and probably also of other
former Soviet republics.
The description below
will concentrate on the Akkord family of protocols. Akkord seems to be a
fairly old family of data transmission protocols.In 1953, the Scientific
Research Institute of Electrical Devices (NII ETU) separated from the
Krasnaya Zarya plant, the main activity of which was the creation of technical
means of data transmission and photo-telegraph equipment. The customers were
the Air Defense Forces, the Navy, and the Space Forces. The data transmission protocols
and equipment created by the Research Institute of ETU, "Araks",
"Aragva", "Pogoda", "Akkord-SS-PD",
"Akkord-SS-PS" and a number of others are still in operation in the ministry
of defense of the Russian Federation.
The
operation of early Akkordprotocols are described in detail in a textbook
from 1975. Akkord-50 was one of the first protocols and used ITA-2 with
blocks of 12x5 data bits + 2 repetition flag bits + 13 checksum bits in fixed
data block mode transmitting at 50 or 100 Baud.
Exchange
of traffic works like this: Sending station (PD) sends signal “start of data” (SSSS) and the receiving station
acknowledges with ‘I’. Then the sending station sends ‘I’ and data block 1 with the two flag
bits set to ‘00’. If the sending station does not receive a receipt (‘I’) within 1.5 s after having
transmitted a data block, it retransmits the unacknowledged block, now with the
flag bits sets to ‘01’.
Akkord-1200
is a protocol designed to be used for data transmission at 600 or 1200 Baud.

In
the figure above, the upper diagram shows the structure of a block at 600 Baud,
while the lower diagram shows the structure of a 1200 Baud block. The former
has 4 block number bits + 112 data bits + 16 check bits (= 132 bits), while the
latter utilizes 4 + 240 + 16 bits (= 260 bits). The generator polynomial is x16+x12+x5+1,
CCITT-16, the same as used in laterAkkordprotocols described below. It
should be obvious from the above illustration, that the structure of Akkord-1200
is very similar to Akkord-SS-PD.
A
special phasing block initiates the transmission of data:
The
block consists of 4 flag bits + 60 bits with an optimal number of bit
transitions + 16 bits of a special phasing sequence, totaling 80 bits.

As
is obvious from the illustration above, Akkord-1200 transmits a data
block adding an ‘A’ to the first block, which is acknowledged by the receiving station.
Block B(2) arrives in error, and a request for retransmission is issued. Block
C (3) is transmitted. When the sending station receives the request, it
retransmits blocks B and C.
2.
Protocol stack
The
protocols discussed in this document are datalink layer protocols and the data
they transport. The physical level network, transport and application layers is
consequently not part of this discussion, but most of the intercepted radio links
have been characterized by a telegraph speed of 1200 bps FSK with a shift of
800 Hz in the frequency range 5-7 MHz, which suggests short to medium range
distances. Sources also mention the use of PSK, both absolute and differential.
DATA
|
APPLICATION
|
TRANSPORT
|
NETWORK
|
DATA LINK
|
PHYSICAL
|
2.1
Datalink layer protocols
Protocol name
|
Block length [(n, k] bits]
|
Speed
|
Used with
|
Remarks
|
Akkord-SS-PD
(“Akkord-165”)
|
(165,
144), (117, 96), (69, 48)
|
1200,
2400, 4800, 9600 bps
|
AI-010,
55Ts6, R-050
|
|
Akkord-SS-PS
|
(69,
48)
|
2400
bps
|
AI-011
|
|
Irtysh
|
(117,
96), (69, 48)
|
|
T-235-1L
|
|
Aragva
|
(126,
110)
|
1200
bps
|
5Ts55
|
|
Araks
|
?
|
?
|
?
|
Details
unknown
|
Pogoda
|
?
|
?
|
?
|
Details
unknown
|
Almaz
|
?
|
50
Baud
|
KSA 5D72, 88E8
|
|
?
|
(20,
15)
|
|
5Ts49
|
ITA-2
(MTK-2)
|
Pautina-1M
|
(32,
28)
|
60
|
?
|
2
channels
|
STA-2M
|
15
… 26 characters
|
|
STA-2M
|
ITA-2
without check bits
|
RB
|
?
|
?
|
?
|
Details
unknown
|
ASPD-U
|
(40,
36)
|
234,
468
|
|
|
2.1.2
Akkord-SS-PD
This protocol
appears to be the main protocol used. The packet structure of this protocol is
as follows:

The
4 m-bits are so called service bits which are used to signal various modes of
operation at link level. The C-bit – also a service bit - indicates the address
mode used, and the CRC is the cyclic redundancy check protecting the preceding
bits. The protocol is transparent and may transport various types of
information.
2.1.2.1
m-bits
These
bits indicate the various operation modes of the protocol.
Service bits
|
Service bit values
|
Type of packet
|
m
|
C
|
1st Bit
|
2nd Bit
|
3rd Bit
|
4th Bit
|
X
|
X
|
X
|
X
|
1
|
Addressed
|
1
|
0
|
X
|
X
|
0
|
Broadcast
|
1
|
1
|
X
|
X
|
0
|
Idle
|
Note - "X" is the arbitrary value of this bit (0
or 1)
|
m1 = Flag for erasure mode operation (in Russian DRS, see below)
m2= Flag for feedback (ARQ) operation (in Russian DRO, see below)
Only for duplex operation:
m3 = Receipt for received data block
m4 = Communication channel status indicator
m-bits
|
Remarks
|
1000
|
Observed
with (165, 144)and (117, 96) payload packets
|
1010
|
Observed
with (69, 48) payload packets
|
1100
|
Observed
with (117, 96) scrambled idle (‘silence’) packets, and (69, 48)scrambled packets
|
1101
|
Bit
sync packets
|
Other
sources indicate this for Akkord-165:
Digitvalues
|
Codagram types
|
m
|
C
|
0101
|
1
|
Phasing combination (F1 or
F2)
|
010Ñ…
|
0
|
Idle combination (Ñ… – arbitrary value)
|
zzz0
|
0
|
Information block, where the
value of zzz determine the retransmission delay code
|
m - service digits
C - decisive feedback flag
2.1.2.2
Data bits
Data
transmission is carried out in blocks of 48, 96 or 144 data bits structured in
a number of – 2, 4, or 6 - 24-bit data words. The data field contains the data
payload.
2.1.2.2
C-bit
The
C-bit indicates whether the data block isunaddressed (‘0’) or addressed (‘1’) to a single recipient or is a
broadcast. The address range is 1 to 15.
2.1.2.3
Checksum
The
m-bits, the data bits and the c-bit are protected by a 16-bit cyclic redundancy
check with a generator polynomial G(x) = x16+x12+x5+1, which is the standard CCITT CRC-16. This will detect
six errors and correct three.
2.1.2.3
Protocol operation
The protocol is able
to operate in three connection modes: Simplex, full duplex and half duplex. In
duplex two sub-modes are available: Data erasure mode, where packets received
in error are dropped, and feedback mode, where packets received in error are
requested to be retransmitted by the opposite party by transmitting block
repetition requests(RQ). If ‘erasure mode’ is disabled, packets in error are
forwarded to the recipient (DTE) accompanied with an error indication to warn
the end consumer. In simplex and half-duplex modes only erasure operation is
available.
SIMPLEX (SPX)
Format: Full Akkord
format
Initial sync: 2 x
block length reversals + sync seq (normal)
Cyclic sync: Each 21th
block is sync seq (normal)
No traffic: Idle
sequence
FULL DUPLEX
(FDX)
OK ODK-1
Format: Full Akkord
format
Initial sync: 2 x
block length reversals + sync seq (normal)
Cyclic sync: Each 96th
block is sync seq (normal)
No traffic: Idle
sequence
OK ODK-2
Format: All bits
except the 16-bit CRC are used for data.
Initial sync: 2x block
length reversals + sync seq (normal) + sync seq (inverted)
Cyclic sync: Each 96th
and 97th block is sync seq (normal) + sync seq (inverted)
No traffic: Idle
sequence
OK ODK-2-1
Like ODK 2, but
connection is closed down, when no traffic is received from the DTE.
HALF DUPLEX
(HDX) (Alternate Duplex mode)
PDK-1
Format: Full Akkord
format
Initial sync: 2 x
block length reversals + sync seq (normal)
Cyclic sync: Each 21th
block is sync seq (normal)
No traffic: Connection
closed down
Other sources use the
terms Simplex-1, Duplex-1, Duplex-2 and Duplex-3:
SIMPLEX-1,
DUPLEX-1
Sync seq every 95th
block.
DUPLEX-2
Resync is only
initiated if a certain number of Block Repeat Requests
is received. Resync is attempted until sync is achieved, i.e. when the exchange
of sync seqs with F1 and F2 flags takes place. This happens when the opposite
party responds with F2 when sync seqs with F1 are received.
DUPLEX-3
Sync every 20th
block. After initial sync synchronization is only resumed if a Phase Request
signal is received from the opposite party. If sync is not achieved after 96
blocks, then sync seqs are sent every 20th block.
2.1.2.3.1
Synchronization
2.1.2.3.2 Idle
The idle packet for
the (69, 48) and (165, 144) formats is structured like this:
m = 1101
(4 bits)
data =
‘01’ (reversals) (48 | 144 bits)
c-bit (1
bit)
crc (16
bits)
In the idle state the
(117, 96) format seems to transmit packets with their contents set to the bits
of a pseudo-random bit stream.
2.1.2.3.2.
Request for repetition
In duplex mode and with
feedback mode selected, the transmitting party stores the already transmitted
data blocks and waits for acknowledgement from the opposite party. If an
acknowledgement is not received within the time of four data blocks, then the
stored data block is retransmitted. This procedure continues until
acknowledgement is received.
3. Application
protocols
In this document, the
term application protocol is used to describe the data transported by the
underlying protocol(s).
It seems that the family
of Akkord protocols are used for a number of purposes including transmission of
defense radio monitoring data and maybe most importantly in automated air
defense systems deploying radar stations and altimeters, computation centers,
fire control for surface to air missiles (SAM) and control and command for interceptors.
3.1. (69,48)
3.1.2. Irtysh
(T-235-1L)
The transfer of
information from a radar station via the AI-011 DCE is made automatically using
three types of messages with the following priorities:
1. Coordinates of the radar station and information
about its operating modes.
2. Target coordinates and bearings for AShP setters.
3. Target coordinates and signs and a flagfor the radar image acquisition channel (automatic or PAS).
Messages of the
first priority are sent three times passing the DNA radar direction
relative to North.
Messages of the
second priority about bearings for AShP setters issued only
through the automatic pickup channel.
Messages of the
third priority are target information packages sent as they are created.