```
```
Any implementation of this PRNG MUST fulfill three validation criteria: the one described in (for the TinyMT32 32-bit unsigned integer generator), and the two others detailed in (for the mapping to 4-bit and 8-bit intervals).
Because of the way the mapping functions work, it is unlikely that an implementation that fulfills the first criterion fails to fulfill the two others.

```
```
The coding coefficients, used during the encoding process, are generated at the RLC encoder by the generate_coding_coefficients()
function each time a new repair symbol needs to be produced.
The fraction of coefficients that are non zero (i.e., the density) is controlled by the DT (Density Threshold) parameter.
DT has values between 0 (the minimum value) and 15 (the maximum value), and the average probability of having a non zero coefficient equals (DT + 1) / 16.
In particular, when DT equals 15 the function guaranties that all coefficients are non zero (i.e., maximum density).
These considerations apply to both the RLC over GF(2) and RLC over GF(2^^8), the only difference being the value of the m parameter.
With the RLC over GF(2) FEC Scheme (), m is equal to 1.
With RLC over GF(2^^8) FEC Scheme (), m is equal to 8.
```
```

```
```
The two RLC FEC Schemes specified in this document reuse the Finite Fields defined in , section 8.1.
More specifically, the elements of the field GF(2^^m) are represented by polynomials with binary coefficients (i.e., over GF(2)) and degree lower or equal to m-1.
The addition between two elements is defined as the addition of binary polynomials in GF(2), which is equivalent to a bitwise XOR operation on the binary representation of these elements.
With GF(2^^8), multiplication between two elements is the multiplication modulo a given irreducible polynomial of degree 8.
The following irreducible polynomial MUST be used for GF(2^^8):
x^^8 + x^^4 + x^^3 + x^^2 + 1

With GF(2), multiplication corresponds to a logical AND operation.
The two RLC FEC Schemes require the computation of a linear combination of source symbols, using the coding coefficients produced by the generate_coding_coefficients() function and stored in the cc_tab[] array.
With the RLC over GF(2^^8) FEC Scheme, a linear combination of the ew_size source symbol present in the encoding window, say src_0 to src_ew_size_1, in order to generate a repair symbol, is computed as follows.
For each byte of position i in each source and the repair symbol, where i belongs to {0; E-1}, compute:
repair[i] = cc_tab[0] * src_0[i] XOR cc_tab[1] * src_1[i] XOR ... XOR cc_tab[ew_size - 1] * src_ew_size_1[i]

where * is the multiplication over GF(2^^8).
In practice various optimizations need to be used in order to make this computation efficient (see in particular ).
With the RLC over GF(2) FEC Scheme (binary case), a linear combination is computed as follows.
The repair symbol is the XOR sum of all the source symbols corresponding to a coding coefficient cc_tab[j] equal to 1 (i.e., the source symbols corresponding to zero coding coefficients are ignored).
The XOR sum of the byte of position i in each source is computed and stored in the corresponding byte of the repair symbol, where i belongs to {0; E-1}.
In practice, the XOR sums will be computed several bytes at a time (e.g., on 64 bit words, or on arrays of 16 or more bytes when using SIMD CPU extensions).
With both FEC Schemes, the details of how to optimize the computation of these linear combinations are of high practical importance but out of scope of this document.

```
```
This fully-specified FEC Scheme defines the Sliding Window Random Linear Codes (RLC) over GF(2^^8).
Following the guidelines of , section 5.6, this section provides
the FEC Framework Configuration Information (or FFCI).
This FCCI needs to be shared (e.g., using SDP) between the FECFRAME sender and receiver
instances in order to synchronize them.
It includes a FEC Encoding ID, mandatory for any FEC Scheme specification, plus scheme-specific elements.
FEC Encoding ID:
the value assigned to this fully specified FEC Scheme MUST be XXXX,
as assigned by IANA ().

When SDP is used to communicate the FFCI, this FEC Encoding ID is carried in
the 'encoding-id' parameter.
The FEC Scheme-Specific Information (FSSI) includes elements that are specific to the present FEC Scheme.
More precisely:
a non-negative integer that indicates the size of each encoding symbol in bytes;
a non-negative integer between 0 and 255 (both inclusive) used to initialize window sizes.
A value of 0 indicates this parameter is not considered (e.g., a fixed encoding window size may be chosen).
A value between 1 and 255 inclusive is required by certain of the parameter derivation techniques described in ;

This element is required both by the sender (RLC encoder) and the receiver(s) (RLC decoder).
When SDP is used to communicate the FFCI, this FEC Scheme-specific information is carried in
the 'fssi' parameter in textual representation as specified in .
For instance:
fssi=E:1400,WSR:191
In that case the name values "E" and "WSR" are used to convey the E and WSR parameters respectively.
If another mechanism requires the FSSI to be carried as an opaque octet string, the encoding format consists
of the following three octets, where the E field is carried in "big-endian" or "network order" format, that is,
most significant byte first:
Encoding symbol length (E): 16-bit field;
Window Size Ratio Parameter (WSR): 8-bit field.

These three octets can be communicated as such, or for instance, be subject to an additional Base64 encoding.
A FEC Source Packet MUST contain an Explicit Source FEC Payload ID that is appended to the
end of the packet as illustrated in .
More precisely, the Explicit Source FEC Payload ID is composed of the following field,
carried in "big-endian" or "network order" format, that is, most significant byte first
():
this unsigned integer identifies the first source symbol of the ADUI corresponding to this FEC Source Packet.
The ESI is incremented for each new source symbol, and after reaching the maximum value
(2^32-1), wrapping to zero occurs.

A FEC Repair Packet MAY contain one or more repair symbols.
When there are several repair symbols, all of them MUST have been generated from the same encoding window,
using Repair_Key values that are managed as explained below.
A receiver can easily deduce the number of repair symbols within a FEC Repair Packet by
comparing the received FEC Repair Packet size (equal to the UDP payload size when UDP is the underlying
transport protocol) and the symbol size, E, communicated in the FFCI.
A FEC Repair Packet MUST contain a Repair FEC Payload ID that is prepended to the
repair symbol as illustrated in .
More precisely, the Repair FEC Payload ID is composed of the following fields where all integer fields are carried
in "big-endian" or "network order" format, that is, most significant byte first ():
this unsigned integer is used as a seed by the coefficient generation function ()
in order to generate the desired number of coding coefficients.
This repair key may be a monotonically increasing integer value that loops back to 0 after reaching 65535
(see ).
When a FEC Repair Packet contains several repair symbols, this repair key value is that of the first repair symbol.
The remaining repair keys can be deduced by incrementing by 1 this value, up to a maximum value of 65535 after which it loops back to 0.
this unsigned integer carries the Density Threshold (DT) used by the coding coefficient generation function .
More precisely, it controls the probability of having a non zero coding coefficient, which equals (DT+1) / 16.
When a FEC Repair Packet contains several repair symbols, the DT value applies to all of them;
this unsigned integer indicates the number of source symbols in the encoding window when this repair symbol was generated.
When a FEC Repair Packet contains several repair symbols, this NSS value applies to all of them;
this unsigned integer indicates the ESI of the first source symbol in the encoding window when this repair symbol was generated.
When a FEC Repair Packet contains several repair symbols, this FSS_ESI value applies to all of them;

All the procedures of apply to this FEC Scheme.
This fully-specified FEC Scheme defines the Sliding Window Random Linear Codes (RLC) over GF(2) (binary case).
FEC Encoding ID:
the value assigned to this fully specified FEC Scheme MUST be YYYY,
as assigned by IANA ().

When SDP is used to communicate the FFCI, this FEC Encoding ID is carried in
the 'encoding-id' parameter.
All the considerations of apply here.
All the considerations of apply here.
All the considerations of apply here, with the only exception that the Repair_Key field
is useless if DT = 15 (indeed, in that case all the coefficients are necessarily equal to 1 and the coefficient generation function does not use any PRNG).
When DT = 15 the FECFRAME sender MUST set the Repair_Key field to zero on transmission and a receiver MUST ignore it on receipt.
All the procedures of apply to this FEC Scheme.
This section provides a high level description of a Sliding Window RLC encoder.
Whenever a new FEC Repair Packet is needed, the RLC encoder instance first gathers the ew_size source symbols currently in the sliding encoding window.
Then it chooses a repair key, which can be a monotonically increasing integer value, incremented for each repair symbol up to a maximum
value of 65535 (as it is carried within a 16-bit field) after which it loops back to 0.
This repair key is communicated to the coefficient generation function () in order to generate
ew_size coding coefficients.
Finally, the FECFRAME sender computes the repair symbol as a linear combination of the ew_size source symbols using the ew_size coding coefficients
().
When E is small and when there is an incentive to pack several repair symbols within the same FEC Repair Packet, the appropriate number of repair symbols
are computed.
In that case the repair key for each of them MUST be incremented by 1, keeping the same ew_size source symbols, since only the first repair key will
be carried in the Repair FEC Payload ID.
The FEC Repair Packet can then be passed to the transport layer for transmission.
The source versus repair FEC packet transmission order is out of scope of this document and several approaches exist that are implementation-specific.
Other solutions are possible to select a repair key value when a new FEC Repair Packet is needed, for instance, by choosing a random integer between 0 and 65535.
However, selecting the same repair key as before (which may happen in case of a random process) is only meaningful if the encoding window has changed,
otherwise the same FEC Repair Packet will be generated.
This section provides a high level description of a Sliding Window RLC decoder.
A FECFRAME receiver needs to maintain a linear system whose variables are the received and lost source symbols.
Upon receiving a FEC Repair Packet, a receiver first extracts all the repair symbols it contains (in case several repair symbols are packed together).
For each repair symbol, when at least one of the corresponding source symbols it protects has been lost, the receiver adds an equation to the linear system
(or no equation if this repair packet does not change the linear system rank).
This equation of course re-uses the ew_size coding coefficients that are computed by the same coefficient generation function
(Section ), using the repair key and encoding window descriptions carried in the Repair FEC Payload ID.
Whenever possible (i.e., when a sub-system covering one or more lost source symbols is of full rank), decoding is performed in order to recover
lost source symbols.
Gaussian elimination is one possible algorithm to solve this linear system.
Each time an ADUI can be totally recovered, padding is removed (thanks to the Length field, L, of the ADUI) and the ADU is assigned to the corresponding
application flow (thanks to the Flow ID field, F, of the ADUI).
This ADU is finally passed to the corresponding upper application.
Received FEC Source Packets, containing an ADU, MAY be passed to the application either immediately or after some time to guaranty an ordered delivery to
the application.
This document does not mandate any approach as this is an operational and management decision.
With real-time flows, a lost ADU that is decoded after the maximum latency or an ADU received after this delay has no value to the application.
This raises the question of deciding whether or not an ADU is late.
This decision MAY be taken within the FECFRAME receiver (e.g., using the decoding window, see )
or within the application (e.g., using RTP timestamps within the ADU).
Deciding which option to follow and whether or not to pass all ADUs, including those assumed late, to the application are operational decisions that depend
on the application and are therefore out of scope of this document.
Additionally, discusses a backward compatible optimization whereby late source symbols MAY still be used within
the FECFRAME receiver in order to improve transmission robustness.
Editor's notes: RFC Editor, please remove this section motivated by RFC 6982 before publishing the RFC. Thanks.
An implementation of the Sliding Window RLC FEC Scheme for FECFRAME exists:
Organisation: Inria
Description: This is an implementation of the Sliding Window RLC FEC Scheme limited to GF(2^^8).
It relies on a modified version of our OpenFEC (http://openfec.org) FEC code library.
It is integrated in our FECFRAME software (see ).
Maturity: prototype.
Coverage: this software complies with the Sliding Window RLC FEC Scheme.
Licensing: proprietary.
Contact: vincent.roca@inria.fr

The FEC Framework document provides a fairly comprehensive
analysis of security considerations applicable to FEC Schemes.
Therefore, the present section follows the security considerations section of
and only discusses specific topics.
The Sliding Window RLC FEC Scheme specified in this document does not change the
recommendations of .
To summarize, if confidentiality is a concern, it is RECOMMENDED that one of the
solutions mentioned in is used with special
considerations to the way this solution is applied (e.g., is encryption applied
before or after FEC protection, within the end-system or in a middlebox), to the operational
constraints (e.g., performing FEC decoding in a protected environment may be
complicated or even impossible) and to the threat model.
The Sliding Window RLC FEC Scheme specified in this document does not change the
recommendations of .
To summarize, it is RECOMMENDED that one of the solutions mentioned in
is used on both the FEC Source and Repair Packets.
The FEC Scheme specified in this document defines parameters that
can be the basis of attacks.
More specifically, the following parameters of the FFCI may be modified
by an attacker who targets receivers ():
FEC Encoding ID:
changing this parameter leads a receiver to consider a different FEC Scheme.
The consequences are severe, the format of the Explicit Source FEC Payload ID
and Repair FEC Payload ID of received packets will probably differ, leading to
various malfunctions.
Even if the original and modified FEC Schemes share the same format, FEC decoding
will either fail or lead to corrupted decoded symbols.
This will happen if an attacker turns value YYYY (i.e., RLC over GF(2)) to value XXXX (RLC over GF(2^^8)),
an additional consequence being a higher processing overhead at the receiver.
In any case, the attack results in a form of Denial of Service (DoS) or corrupted content.
Encoding symbol length (E):
setting this E parameter to a different value will confuse a receiver.
If the size of a received FEC Repair Packet is no longer multiple of the modified E value,
a receiver quickly detects a problem and SHOULD reject the packet.
If the new E value is a sub-multiple of the original E value (e.g., half the original value),
then receivers may not detect the problem immediately.
For instance, a receiver may think that a received FEC Repair Packet contains more repair symbols
(e.g., twice as many if E is reduced by half), leading to malfunctions whose nature depends on
implementation details.
Here also, the attack always results in a form of DoS or corrupted content.

It is therefore RECOMMENDED that security measures be taken to
guarantee the FFCI integrity, as specified in .
How to achieve this depends on the way the FFCI is communicated from the sender
to the receiver, which is not specified in this document.
Similarly, attacks are possible against the Explicit Source FEC Payload ID
and Repair FEC Payload ID.
More specifically, in case of a FEC Source Packet, the following value can be modified by an attacker who targets receivers:
Encoding Symbol ID (ESI):
changing the ESI leads a receiver to consider a wrong ADU, resulting in severe consequences, including
corrupted content passed to the receiving application;

And in case of a FEC Repair Packet:
Repair Key:
changing this value leads a receiver to generate a wrong coding coefficient sequence, and therefore
any source symbol decoded using the repair symbols contained in this packet will be corrupted;
DT:
changing this value also leads a receiver to generate a wrong coding coefficient sequence, and therefore
any source symbol decoded using the repair symbols contained in this packet will be corrupted.
In addition, if the DT value is significantly increased, it will generate a higher processing overhead at a receiver.
In case of very large encoding windows, this may impact the terminal performance;
NSS:
changing this value leads a receiver to consider a different set of source symbols, and therefore
any source symbol decoded using the repair symbols contained in this packet will be corrupted.
In addition, if the NSS value is significantly increased, it will generate a higher processing overhead at a receiver,
which may impact the terminal performance;
FSS_ESI:
changing this value also leads a receiver to consider a different set of source symbols and therefore
any source symbol decoded using the repair symbols contained in this packet will be corrupted.

It is therefore RECOMMENDED that security measures are taken to guarantee the
FEC Source and Repair Packets as stated in .
The Sliding Window RLC FEC Scheme specified in this document does not change the
recommendations of .
The Sliding Window RLC FEC Scheme specified in this document does not change the
recommendations of concerning the use of
the IPsec/ESP security protocol as a mandatory to implement (but not mandatory
to use) security scheme.
This is well suited to situations where the only insecure domain is the one
over which the FEC Framework operates.
In addition to the above security considerations, inherited from ,
the present document introduces several formulae, in particular in .
It is RECOMMENDED to check that the computed values stay within reasonable bounds since numerical overflows,
caused by an erroneous implementation or an erroneous input value, may lead to hazardous behaviours.
However, what "reasonable bounds" means is use-case and implementation dependent and is not detailed in this document.
also mentions the possibility of "using the
timestamp field of an RTP packet header" when applicable.
A malicious attacker may deliberately corrupt this header field in order to trigger hazardous behaviours at a FECFRAME receiver.
Protection against this type of content corruption can be addressed with the above recommendations on a baseline secure operation.
In addition, it is also RECOMMENDED to check that the timestamp value be within reasonable bounds.
The FEC Framework document provides a fairly comprehensive
analysis of operations and management considerations applicable to FEC Schemes.
Therefore, the present section only discusses specific topics.
The present document specifies two FEC Schemes that differ on the Finite Field used for the coding coefficients.
It is expected that the RLC over GF(2^^8) FEC Scheme will be mostly used since it warrants a higher packet loss protection.
In case of small encoding windows, the associated processing overhead is not an issue (e.g., we measured decoding speeds between
745 Mbps and 2.8 Gbps on an ARM Cortex-A15 embedded board in for an encoding window of size 18 or 23 symbols).
Of course the CPU overhead will increase with the encoding window size, because more operations in the GF(2^^8) finite field will
be needed.
The RLC over GF(2) FEC Scheme offers an alternative.
In that case operations symbols can be directly XOR-ed together which warrants high bitrate encoding and decoding operations, and
can be an advantage with large encoding windows.
However, packet loss protection is significantly reduced by using this FEC Scheme.
In addition to the choice of the Finite Field, the two FEC Schemes define a coding coefficient density threshold (DT) parameter.
This parameter enables a sender to control the code density, i.e., the proportion of coefficients that are non zero on average.
With RLC over GF(2^^8), it is usually appropriate that small encoding windows be associated to a density threshold equal to 15,
the maximum value, in order to warrant a high loss protection.
On the opposite, with larger encoding windows, it is usually appropriate that the density threshold be reduced.
With large encoding windows, an alternative can be to use RLC over GF(2) and a density threshold equal to 7 (i.e., an average density equal to 1/2) or smaller.
Note that using a density threshold equal to 15 with RLC over GF(2) is equivalent to using an XOR code that computes the XOR sum of all the source symbols in the encoding window.
In that case: (1) only a single repair symbol can be produced for any encoding window, and (2) the repair_key parameter becomes useless (the coding coefficients generation function does not rely on the PRNG).
This document registers two values in the "FEC Framework (FECFRAME)
FEC Encoding IDs" registry as follows:
YYYY refers to the Sliding Window Random Linear Codes (RLC) over GF(2) FEC Scheme for Arbitrary Packet Flows, as defined in of this document.
XXXX refers to the Sliding Window Random Linear Codes (RLC) over GF(2^^8) FEC Scheme for Arbitrary Packet Flows, as defined in of this document.

The authors would like to thank the three TSVWG chairs, Wesley Eddy, our shepherd, David Black and Gorry Fairhurst, as well as Spencer Dawkins, our responsible AD,
and all those who provided comments, namely (alphabetical order) Alan DeKok, Jonathan Detchart, Russ Housley, Emmanuel Lochin, Marie-Jose Montpetit, and Greg Skinner.
Last but not least, the authors are really grateful to the IESG members, in particular Benjamin Kaduk, Mirja Kuhlewind, Eric Rescorla, and Adam Roach for their highly valuable feedbacks that greatly contributed to improve this specification.

```
```
&rfc2119;
&rfc8174;
&rfc6363;
&rfc6364;
Forward Error Correction (FEC) Framework Extension to Sliding Window Codes
TinyMT32 Pseudo Random Number Generator (PRNG)
&rfc5170;
&rfc5510;
&rfc6726;
&rfc6816;
&rfc6865;
&rfc8406;
Block or Convolutional AL-FEC Codes? A Performance Comparison for Robust Low-Latency Communications
Less Latency and Better Protection with AL-FEC Sliding Window Codes: a Robust Multimedia CBR Broadcast Case Study
A Complete Treatment of Software Implementations of Finite Field Arithmetic for Erasure Coding Applications
PRNG determinism, for a given seed, is a requirement.
Consequently, in order to validate an implementation of the TinyMT32 PRNG, the following criteria MUST be met.
The first criterion focusses on the tinymt32_rand256(), where the 32-bit integer of the core TinyMT32 PRNG is scaled down to an 8-bit integer.
Using a seed value of 1, the first 50 values returned by: tinymt32_rand256() as 8-bit unsigned integers
MUST be equal to values provided in .
The second criterion focusses on the tinymt32_rand16(), where the 32-bit integer of the core TinyMT32 PRNG is scaled down to a 4-bit integer.
Using a seed value of 1, the first 50 values returned by: tinymt32_rand16() as 4-bit unsigned integers
MUST be equal to values provided in .
This annex discusses the adequacy of the TinyMT32 PRNG and the tinymt32_rand16() and tinymt32_rand256() functions, to the RLC FEC Schemes.
The goal is to assess the adequacy of these two functions in producing coding coefficients that are sufficiently different from one another, across various repair symbols with repair key values in sequence (we can expect this approach to be commonly used by implementers, see ).
This section is purely informational and does not claim to be a solid evaluation.
The two RLC FEC Schemes use the PRNG to produce pseudo-random coding coefficients (), each time a new repair symbol is needed.
A different repair key is used for each repair symbol, usually by incrementing the repair key value ().
For each repair symbol, a limited number of pseudo-random numbers is needed, depending on the DT and encoding window size (), using either tinymt32_rand16() or tinymt32_rand256().
Therefore we are more interested in the randomness of small sequences of random numbers mapped to 4-bit or 8-bit integers, than in the randomness of a very large sequence of random numbers which is not representative of the usage of the PRNG.
Evaluation of tinymt32_rand16():
We first generate a huge number (1,000,000,000) of small sequences (20 pseudo-random numbers per sequence), and perform statistics on the number of occurrences of each of the 16 possible values across all sequences.
The results () show that all possible values are almost equally represented, or said differently, that the tinymt32_rand16() output converges to a uniform distribution where each of the 16 possible values would appear exactly 1 / 16 * 100 = 6.25% of times.
Other types of biases may exist that may be visible with smaller tests, for instance to evaluate the convergence speed to a uniform distribution.
We therefore perform 200 tests, each of them consisting in producing 200 sequences, keeping only the first value of each sequence.
We use non overlapping repair keys for each sequence, starting with value 0 and increasing it after each use.
shows across all 200 tests, for each of the 16 possible pseudo-random number values, the minimum (resp. maximum) number of times it appeared in a test, as well as the average number of occurrences across the 200 tests.
Although the distribution is not perfect, there is no major bias.
On the opposite, in the same conditions, the Park Miller linear congruential PRNG of with a result scaled down to 4-bit values, using seeds in sequence starting from 1, returns systematically 0 as the first value during some time, then after a certain repair key value threshold, it systematically returns 1, etc.
Evaluation of tinymt32_rand256():
The same approach is used here.
Results (not shown) are similar: occurrences vary between 7,810,3368 (i.e., 0.3905%) and 7,814,7952 (i.e., 0.3907%).
Here also we see a convergence to the theoretical uniform distribution where each of the 256 possible values would appear exactly 1 / 256 * 100 = 0.390625% of times.
defines several parameters to control the encoder or decoder.
This annex proposes techniques to derive these parameters according to the target use-case.
This annex is informational, in the sense that using a different derivation technique will not prevent the encoder and decoder to interoperate: a decoder can still recover an erased source symbol without any error.
However, in case of a real-time flow, an inappropriate parameter derivation may lead to the decoding of erased source packets after their validity period, making them useless to the target application.
This annex proposes an approach to reduce this risk, among other things.
The FEC Schemes defined in this document can be used in various manners, depending on the target use-case:
the source ADU flow they protect may or may not have real-time constraints;
the source ADU flow may be a Constant Bitrate (CBR) or Variable BitRate (VBR) flow;
with a VBR source ADU flow, the flow's minimum and maximum bitrates may or may not be known;
and the communication path between encoder and decoder may be a CBR communication path (e.g., as with certain LTE-based broadcast channels) or not (general case, e.g., with Internet).

The parameter derivation technique should be suited to the use-case, as described in the following sections.
In the following, we consider a real-time flow with max_lat latency budget.
The encoding symbol size, E, is constant.
The code rate, cr, is also constant, its value depending on the expected communication loss model (this choice is out of scope of this document).
In a first configuration, the source ADU flow bitrate at the input of the FECFRAME sender is fixed and equal to br_in (in bits/s), and this value is known by the FECFRAME sender.
It follows that the transmission bitrate at the output of the FECFRAME sender will be higher, depending on the added repair flow overhead.
In order to comply with the maximum FEC-related latency budget, we have:
dw_max_size = (max_lat * br_in) / (8 * E)

assuming that the encoding and decoding times are negligible with respect to the target max_lat.
This is a reasonable assumption in many situations (e.g., see in case of small window sizes).
Otherwise the max_lat parameter should be adjusted in order to avoid the problem.
In any case, interoperability will never be compromized by choosing a too large value.
In a second configuration, the FECFRAME sender generates a fixed bitrate flow, equal to the CBR communication path bitrate equal to br_out (in bits/s), and this value is known by the FECFRAME sender, as in .
The maximum source flow bitrate needs to be such that, with the added repair flow overhead, the total transmission bitrate remains inferior or equal to br_out.
We have:
dw_max_size = (max_lat * br_out * cr) / (8 * E)

assuming here also that the encoding and decoding times are negligible with respect to the target max_lat.
For decoding to be possible within the latency budget, it is required that the encoding window maximum size be smaller than or at most equal to the decoding window maximum size.
The ew_max_size is the main parameter at a FECFRAME sender, but its exact value has no impact on the the FEC-related latency budget.
The ew_max_size parameter is computed as follows:
ew_max_size = dw_max_size * WSR / 255

In line with , WSR = 191 is considered as a reasonable value (the resulting encoding to decoding window size ratio is then close to 0.75), but other values between 1 and 255 inclusive are possible, depending on the use-case.
The dw_max_size is computed by a FECFRAME sender but not explicitly communicated to a FECFRAME receiver.
However, a FECFRAME receiver can easily evaluate the ew_max_size by observing the maximum Number of Source Symbols (NSS) value contained in the Repair FEC Payload ID of received FEC Repair Packets ().
A receiver can then easily compute dw_max_size:
dw_max_size = max_NSS_observed * 255 / WSR

A receiver can then chose an appropriate linear system maximum size:
ls_max_size ≥ dw_max_size

It is good practice to use a larger value for ls_max_size as explained in , which does not impact maximum latency nor interoperability.
In any case, for a given use-case (i.e., for target encoding and decoding devices and desired protection levels in front of communication impairments) and for the computed ew_max_size, dw_max_size and ls_max_size values, it is RECOMMENDED to check that the maximum encoding time and maximum memory requirements at a FECFRAME sender, and maximum decoding time and maximum memory requirements at a FECFRAME receiver, stay within reasonable bounds.
When assuming that the encoding and decoding times are negligible with respect to the target max_lat, this should be verified as well, otherwise the max_lat SHOULD be adjusted accordingly.
The particular case of session start needs to be managed appropriately since the ew_size, starting at zero, increases each time a new source ADU is received by the FECFRAME sender, until it reaches the ew_max_size value.
Therefore a FECFRAME receiver SHOULD continuously observe the received FEC Repair Packets, since the NSS value carried in the Repair FEC Payload ID will increase too, and adjust its ls_max_size accordingly if need be.
With a CBR flow, session start is expected to be the only moment when the encoding window size will increase.
Similarly, with a CBR real-time flow, the session end is expected to be the only moment when the encoding window size will progressively decrease.
No adjustment of the ls_max_size is required at the FECFRAME receiver in that case.
In the following, we consider a real-time source ADU flow with a max_lat latency budget and a variable bitrate (VBR) measured at the entry of the FECFRAME sender.
A first approach consists in considering the smallest instantaneous bitrate of the source ADU flow, when this parameter is known, and to reuse the derivation of .
Considering the smallest bitrate means that the encoding and decoding window maximum size estimations are pessimistic: these windows have the smallest size required to enable on-time decoding at a FECFRAME receiver.
If the instantaneous bitrate is higher than this smallest bitrate, this approach leads to an encoding window that is unnecessarily small, which reduces robustness in front of long erasure bursts.
Another approach consists in using ADU timing information (e.g., using the timestamp field of an RTP packet header, or registering the time upon receiving a new ADU).
From the global FEC-related latency budget, the FECFRAME sender can derive a practical maximum latency budget for encoding operations, max_lat_for_encoding.
For the FEC Schemes specified in this document, this latency budget SHOULD be computed with:
max_lat_for_encoding = max_lat * WSR / 255

It follows that any source symbols associated to an ADU that has timed-out with respect to max_lat_for_encoding SHOULD be removed from the encoding window.
With this approach there is no pre-determined ew_size value: this value fluctuates over the time according to the instantaneous source ADU flow bitrate.
For practical reasons, a FECFRAME sender may still require that ew_size does not increase beyond a maximum value ().
With both approaches, and no matter the choice of the FECFRAME sender, a FECFRAME receiver can still easily evaluate the ew_max_size by observing the maximum Number of Source Symbols (NSS) value contained in the Repair FEC Payload ID of received FEC Repair Packets.
A receiver can then compute dw_max_size and derive an appropriate ls_max_size as explained in .
When the observed NSS fluctuates significantly, a FECFRAME receiver may want to adapt its ls_max_size accordingly.
In particular when the NSS is significantly reduced, a FECFRAME receiver may want to reduce the ls_max_size too in order to limit computation complexity.
A balance must be found between using an ls_max_size "too large" (which increases computation complexity and memory requirements) and the opposite (which reduces recovery performance).
Finally there are configurations where a source ADU flow has no real-time constraints.
FECFRAME and the FEC Schemes defined in this document can still be used.
The choice of appropriate parameter values can be directed by practical considerations.
For instance, it can derive from an estimation of the maximum memory amount that could be dedicated to the linear system at a FECFRAME receiver, or the maximum computation complexity at a FECFRAME receiver, both of them depending on the ls_max_size parameter.
The same considerations also apply to the FECFRAME sender, where the maximum memory amount and computation complexity depend on the ew_max_size parameter.
Here also, the NSS value contained in FEC Repair Packets is used by a FECFRAME receiver to determine the current coding window size and ew_max_size by observing its maximum value over the time.
This annex introduces non normative considerations.
It is provided as suggestions, without any impact on interoperability.
For more information see .
With a real-time source ADU flow, it is possible to improve the decoding performance of sliding window codes without impacting maximum latency, at the cost of extra memory and CPU overhead.
The optimization consists, for a FECFRAME receiver, to extend the linear system beyond the decoding window maximum size, by keeping a certain number of old source symbols whereas their associated ADUs timed-out:
ls_max_size > dw_max_size

Usually the following choice is a good trade-off between decoding performance and extra CPU overhead:
ls_max_size = 2 * dw_max_size

When the dw_max_size is very small, it may be preferable to keep a minimum ls_max_size value (e.g., LS_MIN_SIZE_DEFAULT = 40 symbols).
Going below this threshold will not save a significant amount of memory nor CPU cycles.
Therefore:
ls_max_size = max(2 * dw_max_size, LS_MIN_SIZE_DEFAULT)

Finally, it is worth noting that a receiver that benefits from an FEC protection significantly higher than what is required to recover from packet losses, can choose to reduce the ls_max_size.
In that case lost ADUs will be recovered without relying on this optimization.
It means that source symbols, and therefore ADUs, may be decoded even if the added latency exceeds the maximum value permitted by the application (the "late source symbols" of ).
It follows that the corresponding ADUs will not be useful to the application.
However, decoding these "late symbols" significantly improves the global robustness in bad reception conditions and is therefore recommended for receivers experiencing bad communication conditions .
In any case whether or not to use this optimization and what exact value to use for the ls_max_size parameter are local decisions made by each receiver independently, without any impact on the other receivers nor on the source.

```
```