This is a purely informative rendering of an RFC that includes verified errata. This rendering may not be used as a reference.
The following 'Verified' errata have been incorporated in this document:
EID 1006
Network Working Group M. Watson
Request for Comments: 5052 M. Luby
Obsoletes: 3452 L. Vicisano
Category: Standards Track Digital Fountain
August 2007
Forward Error Correction (FEC) Building Block
Status of This Memo
This document specifies an Internet standards track protocol for the
Internet community, and requests discussion and suggestions for
improvements. Please refer to the current edition of the "Internet
Official Protocol Standards" (STD 1) for the standardization state
and status of this protocol. Distribution of this memo is unlimited.
Copyright Notice
Copyright (C) The IETF Trust (2007).
Abstract
This document describes how to use Forward Error Correction (FEC)
codes to efficiently provide and/or augment reliability for bulk data
transfer over IP multicast. This document defines a framework for
the definition of the information that needs to be communicated in
order to use an FEC code for bulk data transfer, in addition to the
encoded data itself, and for definition of formats and codes for
communication of that information. Both information communicated
with the encoded data itself and information that needs to be
communicated 'out-of-band' are considered. The procedures for
specifying new FEC codes, defining the information communication
requirements associated with those codes and registering them with
the Internet Assigned Numbers Authority (IANA) are also described.
The requirements on Content Delivery Protocols that wish to use FEC
codes defined within this framework are also defined. The companion
document titled "The Use of Forward Error Correction (FEC) in
Reliable Multicast" describes some applications of FEC codes for
delivering content. This document obsoletes RFC 3452.
Table of Contents
1. Introduction ....................................................3
2. Definitions and Abbreviations ...................................4
3. Requirements Notation ...........................................4
4. Rationale .......................................................5
5. Applicability Statement .........................................6
6. Functionality ...................................................6
6.1. FEC Schemes ................................................8
6.2. FEC Object Transmission Information .......................10
6.2.1. Transport of FEC Object Transmission Information ...11
6.2.2. Opacity of FEC Object Transmission Information .....12
6.2.3. Mandatory FEC Object Transmission
Information Elements ...............................12
6.2.4. Common FEC Object Transmission Information
Elements ...........................................12
6.2.5. Scheme-Specific FEC Object Transmission
Information Element ................................13
6.3. FEC Payload ID ............................................13
7. FEC Scheme Specifications ......................................14
8. CDP Specifications .............................................17
9. Common Algorithms ..............................................18
9.1. Block Partitioning Algorithm ..............................18
9.1.1. First Step .........................................18
9.1.2. Second step ........................................19
10. Requirements from Other Building Blocks .......................20
11. Security Considerations .......................................20
12. IANA Considerations ...........................................21
12.1. Explicit IANA Assignment Guidelines ......................21
13. Changes from RFC 3452 .........................................22
14. Acknowledgments ...............................................23
15. References ....................................................23
15.1. Normative References .....................................23
15.2. Informative References ...................................23
1. Introduction
This document describes how to use Forward Error Correction (FEC)
codes to provide support for reliable delivery of content within the
context of a Content Delivery Protocol (CDP). This document
describes a building block as defined in [10], specifically Section
4.2 of that document, and follows the general guidelines provided in
[5].
The purpose of this building block is to define a framework for
forward error correction such that:
1. CDPs can be designed to operate with a range of different FEC
codes/schemes, without needing to know details of the specific
FEC code/scheme that may be used.
2. FEC schemes can be designed to operate with a range of different
CDPs, without needing to know details of the specific CDPs.
Note that a 'CDP' in the context of this document may consist of
several distinct protocol mechanisms and may support any kind of
application requiring reliable transport -- for example, object
delivery and streaming applications.
This document also provides detailed guidelines on how to write an
RFC for an FEC scheme corresponding to a new FEC Encoding ID (for
both Fully-Specified and Under-Specified FEC Schemes -- see Section
4).
RFC 3452 [3], which is obsoleted by this document, contained a
previous version, which was published in the "Experimental" category.
RFC 3452 was published as an Experimental RFC in part due to the lack
at that time of specified congestion control strategies suitable for
use with Reliable Multicast protocols.
This Proposed Standard specification is thus based on RFC 3452 [3]
updated according to accumulated experience and growing protocol
maturity since the publication of RFC 3452 [3]. Said experience
applies both to this specification itself and to congestion control
strategies related to the use of this specification.
The differences between RFC 3452 [3] and this document are listed in
Section 13.
2. Definitions and Abbreviations
Object: An ordered sequence of octets to be transferred by the
transport protocol. For example, a file or stream.
Symbol: A unit of data processed by the Forward Error Correction
code. A symbol is always considered as a unit, i.e., it is either
completely received or completely lost.
Source symbol: A symbol containing information from the original
object.
Repair symbol: A symbol containing information generated by the FEC
code which can be used to recover lost source symbols.
Encoding symbol: A source symbol or a repair symbol.
Encoder: The FEC scheme specific functions required to transform a
object into FEC encoded data. That is, the functions that produce
repair symbols using source symbols.
Decoder: The FEC scheme-specific functions required to transform
received FEC-encoded data into a copy of the original object.
Receiver: A system supporting the receiving functions of a CDP and
FEC scheme according to this specification.
Sender: A system supporting the sending functions of a CDP and FEC
scheme according to this specification.
Source Block: A part of the object formed from a subset of the
object's source symbols.
CDP: Content Delivery Protocol
FEC: Forward Error Correction
3. Requirements Notation
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in [1].
4. Rationale
An FEC code, in the general sense, is a valuable basic component of
any CDP that is to provide reliable delivery of an object. Using FEC
codes is effective in the context of IP multicast and reliable
delivery because FEC encoding symbols can be useful to all receivers
for reconstructing an object even when the receivers have received
different encoding symbols. Furthermore, FEC codes can ameliorate or
even eliminate the need for feedback from receivers to senders to
request retransmission of lost packets.
Central to this document is the concept of an 'FEC Scheme', which we
distinguish from the concept of an 'FEC code' or 'FEC algorithm'. An
FEC scheme defines the ancillary information and procedures which,
combined with an FEC code or algorithm specification, fully define
how the FEC code can be used with CDPs. An FEC scheme may be
associated with a single standardized FEC code (A 'Fully-Specified'
FEC scheme) or may be applicable to many FEC codes (An 'Under-
Specified' FEC scheme).
This document describes a framework for the definition of FEC
schemes. Definition of actual FEC schemes is outside the scope of
this document. This document also defines requirements for reliable
CDPs that make use of FEC schemes. Any CDP that is compliant to the
requirements specified in this document can make use of any FEC
scheme that is defined within the framework described here. Note
that FEC schemes may place restrictions on the types of CDP they are
intended to be used with. For example, some FEC schemes may be
specific to particular types of application, such as file delivery or
streaming.
The goal of the FEC building block is to describe functionality
directly related to FEC codes that is common to all reliable CDPs and
to all FEC schemes, and to leave out any additional functionality
that is specific to particular CDPs or particular FEC schemes. The
primary functionality described in this document that is common to
all such CDPs that use FEC codes is the definition and transport of
three kinds of information from sender to receiver(s):
1) encoding symbols themselves,
2) ancillary information associated with encoding symbols (or
groups of such symbols, such as the group of symbols in a
single packet, or the group of symbols related to a single
source block), and
3) ancillary information associated with the whole object being
transferred.
It is important to note that this information is only required by the
receiver if one or more of the encoding symbols to which it relates
are received.
This document does not describe how receivers may request
transmission of particular encoding symbols for an object. This is
because although there are CDPs where requests for transmission are
of use, there are also CDPs that do not require such requests.
The companion document [4] should be consulted for a full explanation
of the benefits of using FEC codes for reliable content delivery
using IP multicast. FEC codes are also useful in the context of
unicast, and thus the scope and applicability of this document is not
limited to IP multicast.
5. Applicability Statement
The FEC building block does not provide any support for congestion
control. Any complete multicast CDP MUST provide congestion control
that conforms to [6], in particular, Section 3.2 of that document.
Thus, congestion control MUST be provided by another building block
when the FEC building block is used in a CDP.
A more complete description of the applicability of FEC codes can be
found in the companion document [4].
6. Functionality
This section describes FEC information that is to be sent either in
packets also containing FEC encoding symbols or 'out-of-band'. The
FEC information is associated with transmission of encoding symbols
related to a particular object. There are three classes of packets
that may contain FEC information: data packets, session-control
packets, and feedback packets. They generally contain different
kinds of FEC information. Note that some CDPs may not use session-
control or feedback packets.
Data packets may sometimes serve as session-control packets as well;
both data and session-control packets generally travel downstream
from the sender towards receivers and are sent to a multicast channel
or to a specific receiver using unicast. Session-control packets may
additionally travel upstream from receivers to senders.
As a general rule, feedback packets travel upstream from receivers to
the sender. Sometimes, however, they might be sent to a multicast
channel or to another receiver or to some intermediate node or
neighboring router that provides recovery services.
This document specifies both the FEC information that must be carried
in data packets and the FEC information that must be communicated
from sender to receiver(s) either out-of-band or in data packets.
Specification of protocol mechanisms for transporting this
information, for example, field and packet formats, is out of scope
of this document. Instead, this document specifies at a higher level
the information that must be communicated and provides detailed
requirements for FEC Scheme and Content Delivery Protocol
specifications, which are where the detailed field and packet formats
should be defined.
FEC information is classified as follows:
1. FEC information associated with an object
This is information that is essential for the FEC decoder to
decode a specific object. An example of this information is the
identity of the FEC scheme that is being used to encode the
object, in the form of the FEC Encoding ID. The FEC Encoding ID
is described further below. This information may also include
FEC scheme-specific parameters for the FEC decoder.
2. FEC information associated with specific encoding symbols for an
object
This is information that is associated with one or more encoding
symbols and is thus needed by the decoder whenever one or more of
those encoding symbols have been received. Depending on the FEC
scheme, information may be associated with individual symbols
and/or with groups of symbols. One common such grouping is the
group of symbols included within a single packet. Many FEC
schemes also segment the object being encoded into multiple
'source blocks', each of which is processed independently for FEC
purposes. Information about each source block is another type of
information associated with a group of encoding symbols -- in
this case, the group of symbols which are related to a given
source block.
Two 'containers' are provided for communicating the FEC information
described above, but there is not necessarily a one-to-one
correspondence between the class of FEC information and the mechanism
used. The two mechanisms are:
a. FEC Object Transmission Information
CDPs must provide a reliable mechanism for communicating certain
FEC information from sender to receiver(s). This information is
known as 'FEC Object Transmission Information' and its contents
depend on the particular FEC scheme. It includes all information
of the first class above and may include information of the
second class. The FEC Object Transmission Information can be
sent to a receiver within the data packet headers, within session
control packets, or by some other means.
b. FEC Payload ID
CDPs must provide a mechanism for communicating information which
identifies (for FEC purposes) the encoding symbols carried by a
packet. This information is known as the FEC Payload ID, and its
contents depend on the FEC scheme. It includes only information
of the second class above. A data packet that carries encoding
symbols MUST include an FEC Payload ID.
6.1. FEC Schemes
Two types of FEC scheme are defined by this document: 'Fully-
Specified' FEC schemes and 'Under-Specified' FEC schemes. An FEC
scheme is a Fully-Specified FEC scheme if the encoding scheme is
formally and Fully-Specified, in a way that independent implementors
can implement both encoder and decoder from a specification that is
an IETF RFC.
It is possible that an FEC scheme may not be a Fully-Specified FEC
scheme, because either a specification is simply not available or a
party exists that owns the encoding scheme and is not willing to
disclose the algorithm or specification. We refer to such an FEC
encoding scheme as an Under-Specified FEC scheme.
FEC schemes are identified by an FEC Encoding ID, which is an integer
identifier assigned by IANA. The FEC Encoding ID allows receivers to
select the appropriate FEC decoder. The value of the FEC Encoding ID
MUST be the same for all transmission of encoding symbols related to
a particular object, but MAY vary across different transmissions of
encoding symbols about different objects, even if transmitted to the
same set of multicast channels and/or using a single upper-layer
session.
The FEC Instance ID is an integer value that identifies a specific
instance of an Under-Specified FEC scheme. This value is not used
for Fully-Specified FEC schemes. The FEC Instance ID is scoped by
the FEC Encoding ID, and FEC Instance ID values are subject to IANA
registration.
The FEC Encoding ID for Fully-Specified FEC Schemes and both the FEC
Encoding ID and FEC Instance ID for Under-Specified FEC Schemes are
essential for the decoder to decode an object. Thus, they are part
of the FEC Object Transmission Information.
The following requirements apply to all FEC schemes, whether Fully-
Specified or Under-Specified:
o The type, semantics, and an encoding format for the FEC Payload ID
and the FEC Object Transmission Information MUST be defined.
o A value for the FEC Encoding ID MUST be reserved and associated
with the types, semantics, and encoding format of the FEC Payload
ID and the FEC Object Transmission Information.
The specification for an Under-Specified FEC Scheme MAY allocate a
sub-field within the Scheme-specific FEC Object Transmission
Information element which is for instance-specific information. Each
specific instance of the Under-Specified FEC Scheme may then use this
field in an instance-specific way. The FEC scheme should define the
scheme-specific FEC Object Transmission Information element in such a
way that receivers that do not support the received FEC Instance ID
can still parse and interpret the scheme-specific FEC Object
Transmission Information element with the exception of the instance-
specific field.
An already defined Under-Specified FEC Scheme (i.e., FEC Encoding ID
value) MUST be reused if the associated FEC Payload ID and FEC Object
Transmission Information have the required fields and encoding
formats for a new Under-Specified FEC scheme instance.
An instance of an Under-Specified FEC scheme is fully identified by
the tuple (FEC Encoding ID, FEC Instance ID). The tuple MUST
identify a single scheme instance that has at least one
implementation. The party that owns this tuple MUST be able to
provide information on how to obtain the Under-Specified FEC scheme
instance identified by the tuple, e.g., a pointer to a publicly
available reference-implementation or the name and contacts of a
company that sells it, either separately or embedded in another
product.
This specification reserves the range 0-127 for the values of FEC
Encoding IDs for Fully-Specified FEC schemes and the range 128-255
for the values of Under-Specified FEC schemes.
6.2. FEC Object Transmission Information
The FEC Object Transmission Information contains information which is
essential to the decoder in order to decode the encoded object. It
may also contain information which is required to decode certain
groups of encoding symbols, for example, individual Source Blocks
within the object. This information is communicated reliably by the
CDP to the receiver(s) as described in Section 8.
The FEC Object Transmission Information may consist of several
elements and each element may be one of three types, as follows:
Mandatory: These elements are defined in this specification and are
each mandatory for at least one of the two types of FEC Scheme.
Each FEC scheme specifies how the values of the Mandatory FEC
Object Transmission Information elements are determined and each
CDP specifies how this information is encoded and reliably
communicated to the receiver(s). The Mandatory FEC Object
Transmission Information includes the identification of the FEC
Scheme, which is needed by the receiver to determine whether it
supports the FEC Scheme.
Common: These elements are defined in this specification and are
optional to be used by an FEC scheme. Each FEC scheme specifies
which of the Common FEC Object Transmission Information elements
it uses and how the values of these elements are determined.
Scheme-specific: An FEC scheme may specify a single Scheme-specific
FEC Object Transmission Information element. The FEC scheme
specifies the type, semantics, and encoding format of the Scheme-
specific FEC Object Transmission Information element. The
resulting octet string is known as the "encoded Scheme-specific
FEC Object Transmission Information". Each CDP specifies how the
encoded Scheme-specific FEC Object Transmission is communicated
reliably to the receiver(s), i.e., exactly where it shall be
carried within packets of the CDP. Note that although from the
point of view of this specification and of CDPs, there is only a
single Scheme-specific FEC Object Transmission Information
element, the FEC scheme may specify this element to contain
multiple distinct pieces of information.
Each FEC scheme specifies an encoding format for the Common and
Scheme-specific FEC Object Transmission Information. Each CDP must
specify at least one of the following:
1. A means to reliably communicate the Common FEC Object
Transmission Information elements to the receiver(s) using the
encoding format defined by the FEC scheme.
2. An alternative, CDP-specific, encoding format for each of the
Common FEC Object Transmission Information elements.
The Mandatory and Common FEC Object Transmission Information elements
are defined in the sections below.
6.2.1. Transport of FEC Object Transmission Information
It is the responsibility of the CDP to reliably transport the FEC
Object Transmission Information to the receiver(s).
It is important to note that the encoding format of the Mandatory FEC
Object Transmission Information elements (the FEC Encoding ID) is
defined by the CDP. This is so that the receiver can identify the
FEC Scheme to be used for interpreting the remaining FEC Object
Transmission Information elements. All CDPs must define encoding
formats for the Mandatory FEC Object Transmission Information
element.
Common FEC Object Transmission Information elements can be
transported in two different ways: (a) the FEC Scheme defines an
encoding format for the Common FEC Object Transmission Information
elements that it uses, and the CDP transports this encoded data
block, or (b) the CDP defines an encoding format for each Common FEC
Object Transmission Information element and transports the
information in this format.
An FEC Scheme MUST define an encoding format for the Common FEC
Object Transmission Information elements that it uses. The resulting
octet string is known as the "encoded Common FEC Object Transmission
Information". A CDP MAY define individual encoding formats for each
of the Common FEC Object Transmission Information elements. The
choice of which way the Common FEC Object Transmission Information
elements shall be transported, (a) or (b), is made by the Content
Delivery Protocol, and a particular method SHOULD be defined in the
Content Delivery Protocol specification. Note that a CDP may provide
support for one or both options.
In the case that the CDP uses the encoding format specified by the
FEC scheme, it may simply concatenate the encoded Common FEC Object
Transmission Information and the encoded Scheme-specific FEC Object
Transmission Information, or it may carry each in a separate field or
wrapper within the CDP. In the former case, the concatenated octet
string is known as the encoded FEC Object Transmission Information.
The FEC scheme must define the encoding format for the Common FEC
Object Transmission Information elements that it uses in such a way
that the length of each element is either fixed or can be determined
from the encoded data itself.
The encoding format of the Scheme-specific FEC Object Transmission
Information element is defined by the FEC scheme. CDPs specify only
how the resulting octet sequence is communicated. As with the
encoding format for the Common FEC Object Transmission Information
elements, the length of the Scheme-specific FEC Object Transmission
Information must either be fixed or be possible to determine from the
encoded data itself.
6.2.2. Opacity of FEC Object Transmission Information
The Scheme-specific FEC Object Transmission Information element is
opaque to the CDP in the sense that inspecting the contents of this
element can only be done if FEC scheme-specific logic is included in
the CDP.
Any encoding formats defined by the FEC scheme for the Common FEC
Object Transmission Information elements are also opaque to the CDP
in the same sense.
Any encoding formats defined by the CDP for the Common FEC Object
Transmission Information elements are not opaque in this sense,
although it must be considered that different FEC Schemes may use
different combinations of the Common FEC Object Transmission
Information elements.
6.2.3. Mandatory FEC Object Transmission Information Elements
The Mandatory FEC Object Transmission Information element is:
FEC Encoding ID: an integer between 0 and 255 inclusive identifying
a specific FEC scheme (Fully-Specified or Under-Specified.)
6.2.4. Common FEC Object Transmission Information Elements
The Common FEC Object Transmission Information elements are described
below. Note that with the exception of the FEC Instance ID, this
specification does not provide complete definitions of these fields.
Instead, only aspects of the abstract type are defined. The precise
type and semantics are defined for each FEC scheme in the FEC scheme
specification.
FEC Instance ID: an integer between 0 and 65535 inclusive
identifying an instance of an Under-Specified FEC scheme
Transfer-Length: a non-negative integer indicating the length of the
object in octets
Encoding-Symbol-Length: a non-negative integer indicating the length
of each encoding symbol in octets
Maximum-Source-Block-Length: a non-negative integer indicating the
maximum number of source symbols in a source block
Max-Number-of-Encoding-Symbols: a non-negative integer indicating
the maximum number of encoding symbols (i.e., source plus repair
symbols in the case of a systematic code)
The FEC Instance ID MUST be used by all Under-Specified FEC schemes
and MUST NOT be used by Fully-Specified FEC Schemes.
FEC Schemes define the precise type of those of the above elements
that they use and in particular may restrict the value range of each
element. FEC Schemes also define an encoding format for the subset
of the above elements that they use. CDPs may also provide an
encoding format for each element; in which case, this encoding format
MUST be capable of representing values up to (2^^16)-1 in the case of
the FEC Instance ID, (2^^48)-1 in the case of the Transfer-Length,
and up to (2^^32)-1 for the other elements. CDPs may additionally or
alternatively provide a mechanism to transport the encoded Common FEC
Object Transmission information defined by the FEC scheme. For
example, FLUTE [8] specifies an XML-based encoding format for these
elements, but can also transport FEC scheme-specific encoding formats
within the EXT-FTI LCT header extension.
6.2.5. Scheme-Specific FEC Object Transmission Information Element
The Scheme-specific FEC Object Transmission Information element may
be used by an FEC Scheme to communicate information that is essential
to the decoder and that cannot adequately be represented within the
Mandatory or Common FEC Object Transmission Information elements.
From the point of view of a CDP, the Scheme-specific FEC Object
Transmission Information element is an opaque, variable length, octet
string. The FEC Scheme defines the structure of this octet string,
which may contain multiple distinct elements.
6.3. FEC Payload ID
The FEC Payload ID contains information that indicates to the FEC
decoder the relationships between the encoding symbols carried by a
particular packet and the FEC encoding transformation. For example,
if the packet carries source symbols, then the FEC Payload ID
indicates which source symbols of the object are carried by the
packet. If the packet carries repair symbols, then the FEC Payload
ID indicates how those repair symbols were constructed from the
object.
The FEC Payload ID may also contain information about larger groups
of encoding symbols of which those contained in the packet are part.
For example, the FEC Payload ID may contain information about the
source block the symbols are related to.
The FEC Payload ID for a given packet is essential to the decoder if
and only if the packet itself is received. Thus, it must be possible
to obtain the FEC Payload ID from the received packet. Usually, the
FEC Payload ID is simply carried explicitly as a separate field
within each packet. In this case, the size of the FEC Payload ID
field SHOULD be a small fraction of the packet size. Some FEC
schemes may specify means for deriving the relationship between the
carried encoding symbols and the object implicitly from other
information within the packet, such as protocol headers already
present. Such FEC schemes could obviously only be used with CDPs
which provided the appropriate information from which the FEC Payload
ID could be derived.
The encoding format of the FEC Payload ID, including its size, is
defined by the FEC Scheme. CDPs specify how the FEC Payload ID is
carried within data packets, i.e., the position of the FEC Payload ID
within the CDP packet format and the how it is associated with
encoding symbols.
FEC schemes for systematic FEC codes (that is, those codes in which
the original source data is included within the encoded data) MAY
specify two FEC Payload ID formats, one for packets carrying only
source symbols and another for packets carrying at least one repair
symbol. CDPs must include an indication of which of the two FEC
Payload ID formats is included in each packet if they wish to support
such FEC Schemes.
7. FEC Scheme Specifications
A specification for a new FEC scheme MUST include the following
things:
1. The FEC Encoding ID value that uniquely identifies the FEC
scheme. This value MUST be registered with IANA as described in
Section 12.
2. The type, semantics, and encoding format of one or two FEC
Payload IDs. Where two FEC Payload ID formats are specified,
then the FEC scheme MUST be a systematic FEC code and one FEC
Payload ID format MUST be designated for use with packets
carrying only source symbols, and the other FEC Payload ID format
MUST be designated for use with packets carrying at least one
repair symbol.
3. The type and semantics of the FEC Object Transmission
Information. The FEC Scheme MAY define additional restrictions
on the type (including value range) of the Common FEC Object
Transmission Information elements.
4. An encoding format for the Common FEC Object Transmission
Information elements used by the FEC Scheme.
Fully-Specified FEC schemes MUST further specify:
1. A full specification of the FEC code.
This specification MUST precisely define the valid FEC Object
Transmission Information values, the valid FEC Payload ID values,
and the valid packet payload sizes for any given object (where
packet payload refers to the space -- not necessarily contiguous
-- within a packet dedicated to carrying encoding symbol octets).
Furthermore, given an object, valid values for each of the FEC
Object Transmission Information elements used by the FEC Scheme,
a valid FEC Payload ID value, and a valid packet payload size,
the specification MUST uniquely define the values of the encoding
symbol octets to be included in the packet payload of a packet
with the given FEC Payload ID value.
A common and simple way to specify the FEC code to the required
level of detail is to provide a precise specification of an
encoding algorithm which, given an object, valid values for each
of the FEC Object Transmission Information elements used by the
FEC Scheme for the object, a valid FEC Payload ID, and packet
payload length as input produces the exact value of the encoding
symbol octets as output.
2. A description of practical encoding and decoding algorithms.
This description need not be to the same level of detail as for
(1) above; however, it must be sufficient to demonstrate that
encoding and decoding of the code is both possible and practical.
FEC scheme specifications MAY additionally define the following:
1. Type, semantics, and encoding format of a Scheme-specific FEC
Object Transmission Information element.
Note that if an FEC scheme does not define a Scheme-specific FEC
Object Transmission Information element, then such an element MUST
NOT be introduced in future versions of the FEC Scheme. This
requirement is included to ensure backwards-compatibility of CDPs
designed to support only FEC Schemes that do not use the Scheme-
specific FEC Object Transmission Information element.
Whenever an FEC scheme specification defines an 'encoding format' for
an element, this must be defined in terms of a sequence of octets
that can be embedded within a protocol. The length of the encoding
format MUST either be fixed, or it must be possible to derive the
length from examining the encoded octets themselves. For example,
the initial octets may include some kind of length indication.
FEC schemes SHOULD make use of the Common FEC Object Transmission
Information elements in preference to including information in a
Scheme-specific FEC Object Transmission Information element.
FEC scheme specifications SHOULD use the terminology defined in this
document and SHOULD follow the following format:
1. Introduction <define whether the scheme is Fully-Specified or
Under-Specified>
<describe the use-cases addressed by this FEC scheme>
2. Formats and Codes
2.1 FEC Payload ID(s) <define the type and format of one or two
FEC Payload IDs>
2.2 FEC Object Transmission Information
2.2.1 Mandatory <define the value of the FEC Encoding ID for
this FEC scheme>
2.2.2 Common <describe which Common FEC Object Transmission
Information elements are used by this FEC scheme, define
their value ranges, and define an encoding format for
them>
2.2.3 Scheme-Specific <define the Scheme-specific FEC Object
Transmission Information, including an encoding format, if
required>
3. Procedures <describe any procedures that are specific to this FEC
scheme, in particular derivation and interpretation of the fields
in the FEC Payload ID and FEC Object Transmission Information.>
4. FEC code specification (for Fully-Specified FEC schemes only)
<provide a complete specification of the FEC Code>
Specifications MAY include additional sections such as those
containing examples.
Each FEC scheme MUST be specified independently of all other FEC
schemes; for example, in a separate specification or a completely
independent section of a larger specification.
8. CDP Specifications
A specification for a CDP that uses this building block MUST include
the following things:
1. Definitions of an encoding format for the Mandatory FEC Object
Transmission Information element.
2. A means to reliably communicate the Mandatory FEC Object
Transmission Information element from sender to receiver(s) using
the encoding format defined in (1).
3. Means to reliably communicate the Common FEC Object Transmission
Information element from sender to receiver(s) using either or
both of (a) the encoding format defined by the FEC Scheme or (b)
encoding formats defined by the CDP
4. A means to reliably communicate the Scheme-specific FEC Object
Transmission Information element from sender to receiver(s) using
the encoding format of the Scheme-specific FEC Object
Transmission Information element defined by the FEC scheme.
5. A means to communicate the FEC Payload ID in association with a
data packet. Note that the encoding format of the FEC Payload ID
is defined by the FEC Scheme.
If option (b) of (3) above is used, then the CDP MUST specify an
encoding format for the Common FEC Object Transmission Information
elements.
CDPs MAY additionally specify the following things:
1. A means to indicate whether the FEC Payload ID within a packet is
encoded according to the format for packets including only source
symbols or according to the format for packets including at least
one repair symbol.
9. Common Algorithms
This section describes certain algorithms that are expected to be
commonly required by FEC schemes or by CDPs. FEC Schemes and CDPs
SHOULD use these algorithms in preference to scheme- or protocol-
specific algorithms, where appropriate.
9.1. Block Partitioning Algorithm
This algorithm computes a partitioning of an object into source
blocks so that all source blocks are as close to being equal length
as possible. A first number of source blocks are of the same larger
length, and the remaining second number of source blocks are of the
same smaller length.
This algorithm is described in two steps, the second of which may be
useful in itself as an independent algorithm in some cases. In the
first step, the number of source symbols (T) and the number of source
blocks (N) are derived from the Object transfer length (L), Maximum
Source Block Length (B), and Symbol Length (E).
In the second step, the partitioning of the object is derived from
the number of source symbols (T) and the number of source blocks (N).
The partitioning is defined in terms of a first number of source
blocks (I), a second number of source blocks (N-I), the length of
each of the first source blocks (A_large), and the length of each of
the second source blocks (A_small).
The following notation is used in the description below:
ceil[x] denotes x rounded up to the nearest integer.
floor[x] denotes x rounded down to the nearest integer.
9.1.1. First Step
Input:
B -- Maximum Source Block Length, i.e., the maximum number of source
symbols per source block
L -- Transfer Length in octets
E -- Encoding Symbol Length in octets
Output:
T -- the number of source symbols in the object.
N -- the number of source blocks into which the object shall be
partitioned.
Algorithm:
1. The number of source symbols in the transport object is computed
as T = ceil[L/E].
2. The transport object shall be partitioned into N = ceil[T/B]
source blocks.
9.1.2. Second step
EID 1006 (Verified) is as follows:
Section: 9.1.2
Original Text:
... the last source block is L-((L-1)/E) rounded down to the nearest
integer)*E octets in length.
Corrected Text:
... the last source block is L-floor((L-1)/E)*E octets in length.
Notes:
Apparently, this sections has been crafted to give a formulation of
the algorithm avoiding the distinction between various cases, and in
particular a separate formulation for the "regular" corner case of
an object that can be partitioned exactly into blocks of equal size.
The formulae given in Section 9.1.2 make use of the standard
functions 'ceil' and 'floor' (restated in Section 9.1), but the
final paragraph of the section, at the bottom of page 19, tries to
paraphrase the 'floor' function (see above).
BTW:
Many FEC schemes are only prepared to deal with encoding symbols of
equal size. To accommodate this, wouldn't it therefore have been
preferable to specify padding (to full size E) of the last symbol of
the last block, for the purpose of this common, default algorithm ?
--- VERIFIER NOTES ---
It is the typical case (not a non-standard case) that the object size is not
an even multiple of some nice encoding source block length, and thus
typically A_small not= A_large. Furthermore, it is the typical case that L
is not a multiple of E. Thus, what you characterize as the "regular" case
is actually quite atypical in the real-world.
Also, any application can pad out the last source symbol of a source block
if it wants if the FEC encoder/decoder can't handle it, the specification
does not mandate a particular implementation. On the other hand, it is
unnecessary, and usually wasteful, to actually send those padding bytes over
the wire, and this specification specifies what is sent on the wire and how.
This is why it is like it is.
Input:
T -- the number of source symbols in the object.
N -- the number of source blocks into which the object is
partitioned.
Output:
I -- the number of larger source blocks.
A_large -- the length of each of the larger source blocks in
symbols.
A_small -- the length of each of the smaller source blocks in
symbols.
Algorithm:
1. A_large = ceil[T/N]
2. A_small = floor[T/N]
3. I = T - A_small * N
Each of the first I source blocks then consists of A_large source
symbols; each source symbol is E octets in length. Each of the
remaining N-I source blocks consist of A_small source symbols; each
source symbol is E octets in length, except that the last source
symbol of the last source block is L-((L-1)/E) rounded down to the
nearest integer)*E octets in length.
10. Requirements from Other Building Blocks
The FEC building block does not provide any support for congestion
control. Any complete CDP MUST provide congestion control that
conforms to [6], and thus this MUST be provided by another building
block when the FEC building block is used in a CDP.
There are no other specific requirements from other building blocks
for the use of this FEC building block. However, any CDP that uses
the FEC building block may use other building blocks, for example, to
provide support for sending higher level session information within
data packets containing FEC encoding symbols.
11. Security Considerations
Data delivery can be subject to denial-of-service attacks by
attackers which send corrupted packets that are accepted as
legitimate by receivers. This is particularly a concern for
multicast delivery because a corrupted packet may be injected into
the session close to the root of the multicast tree, in which case,
the corrupted packet will arrive at many receivers. This is
particularly a concern for the FEC building block because the use of
even one corrupted packet containing encoding data may result in the
decoding of an object that is completely corrupted and unusable. It
is thus RECOMMENDED that source authentication and integrity checking
are applied to decoded objects before delivering objects to an
application. For example, a SHA-1 hash [7] of an object may be
appended before transmission, and the SHA-1 hash is computed and
checked after the object is decoded, but before it is delivered to an
application. Source authentication SHOULD be provided, for example,
by including a digital signature verifiable by the receiver and
computed on top of the hash value. It is also RECOMMENDED that a
packet authentication protocol such as Timed Efficient Stream Loss-
Tolerant Authentication (TESLA) [9] be used to detect and discard
corrupted packets upon arrival. Furthermore, it is RECOMMENDED that
Reverse Path Forwarding checks be enabled in all network routers and
switches along the path from the sender to receivers to limit the
possibility of a bad agent successfully injecting a corrupted packet
into the multicast tree data path.
Another security concern is that some FEC information may be obtained
by receivers out-of-band in a session description, and if the session
description is forged or corrupted, then the receivers will not use
the correct protocol for decoding content from received packets. To
avoid these problems, it is RECOMMENDED that measures be taken to
prevent receivers from accepting incorrect session descriptions,
e.g., by using source authentication to ensure that receivers only
accept legitimate session descriptions from authorized senders.
12. IANA Considerations
Values of FEC Encoding IDs and FEC Instance IDs are subject to IANA
registration. They are in the registry named "Reliable Multicast
Transport (RMT) FEC Encoding IDs and FEC Instance IDs" located at
time of publication at:
http://www.iana.org/assignments/rmt-fec-parameters
FEC Encoding IDs and FEC Instance IDs are hierarchical: FEC Encoding
IDs scope independent ranges of FEC Instance IDs. Only FEC Encoding
IDs that correspond to Under-Specified FEC schemes scope a
corresponding set of FEC Instance IDs.
The FEC Encoding ID and FEC Instance IDs are non-negative integers.
In this document, the range of values for FEC Encoding IDs is 0 to
255. Values from 0 to 127 are reserved for Fully-Specified FEC
schemes, and Values from 128 to 255 are reserved for Under-Specified
FEC schemes, as described in more detail in Section 6.1.
12.1. Explicit IANA Assignment Guidelines
This document defines a name-space for FEC Encoding IDs named:
ietf:rmt:fec:encoding
The values that can be assigned within the "ietf:rmt:fec:encoding"
name-space are numeric indexes in the range [0, 255], boundaries
included. Assignment requests are granted on a "IETF Consensus"
basis as defined in [2]. Section 7 defines explicit requirements
that documents defining new FEC Encoding IDs should meet.
This document also defines a name-space for FEC Instance IDs named:
ietf:rmt:fec:encoding:instance
The "ietf:rmt:fec:encoding:instance" name-space is a sub-name-space
associated with the "ietf:rmt:fec:encoding" name-space. Each value
of "ietf:rmt:fec:encoding" assigned in the range [128, 255] has a
separate "ietf:rmt:fec:encoding:instance" sub-name-space that it
scopes. Values of "ietf:rmt:fec:encoding" in the range [0, 127] do
not scope a "ietf:rmt:fec:encoding:instance" sub-name-space.
The values that can be assigned within each "ietf:rmt:fec:encoding:
instance" sub-name-space are non-negative integers less than 65536.
Assignment requests are granted on a "First Come First Served" basis
as defined in [2]. The same value of "ietf:rmt:fec:encoding:
instance" can be assigned within multiple distinct sub-name-spaces,
i.e., the same value of "ietf:rmt:fec:encoding:instance" can be used
for multiple values of "ietf:rmt:fec:encoding".
Requestors of "ietf:rmt:fec:encoding:instance" assignments MUST
provide the following information:
o The value of "ietf:rmt:fec:encoding" that scopes the "ietf:rmt:
fec:encoding:instance" sub-name-space. This must be in the range
[128, 255].
o Point of contact information
o A pointer to publicly accessible documentation describing the
Under-Specified FEC scheme, associated with the value of "ietf:
rmt:fec:encoding:instance" assigned, and a way to obtain it (e.g.,
a pointer to a publicly available reference-implementation or the
name and contacts of a company that sells it, either separately or
embedded in a product).
It is the responsibility of the requestor to keep all the above
information up to date.
13. Changes from RFC 3452
This section lists the changes between the Experimental version of
this specification, [3], and this version:
o The requirements for definition of a new FEC Scheme and the
requirements for specification of new Content Delivery Protocols
that use FEC Schemes are made more explicit to permit independent
definition of FEC Schemes and Content Delivery Protocols.
o The definitions of basic FEC Schemes have been removed with the
intention of publishing these separately.
o The FEC Object Transmission Information (OTI) is more explicitly
defined, and in particular, three classes of FEC OTI (Mandatory,
Common, and Scheme-specific) are introduced to permit reusable
definition of explicit fields in Content Delivery Protocols to
carry these elements.
o FEC Schemes are required to specify a complete encoding for the
FEC Object Transmission, which can be carried transparently by
Content Delivery protocols (instead of defining explicit
elements).
o The possibility for FEC Schemes to define two FEC Payload ID
formats for use with source and repair packets, respectively, in
the case of systematic FEC codes is introduced.
o The file blocking algorithm from FLUTE is included here as a
common algorithm that is recommended to be reused by FEC Schemes
when appropriate.
14. Acknowledgments
This document is largely based on RFC 3452 [3], and thus thanks are
due to the additional authors of that document: J. Gemmell, L. Rizzo,
M. Handley, and J. Crowcroft.
15. References
15.1. Normative References
[1] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", BCP 14, RFC 2119, March 1997.
[2] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA
Considerations Section in RFCs", BCP 26, RFC 2434, October
1998.
15.2. Informative References
[3] Luby, M., Vicisano, L., Gemmell, J., Rizzo, L., Handley, M.,
and J. Crowcroft, "Forward Error Correction (FEC) Building
Block", RFC 3452, December 2002.
[4] Luby, M., Vicisano, L., Gemmell, J., Rizzo, L., Handley, M.,
and J. Crowcroft, "The Use of Forward Error Correction (FEC) in
Reliable Multicast", RFC 3453, December 2002.
[5] Kermode, R. and L. Vicisano, "Author Guidelines for Reliable
Multicast Transport (RMT) Building Blocks and Protocol
Instantiation documents", RFC 3269, April 2002.
[6] Mankin, A., Romanov, A., Bradner, S., and V. Paxson, "IETF
Criteria for Evaluating Reliable Multicast Transport and
Application Protocols", RFC 2357, June 1998.
[7] Federal Information Processing Standards Publication (FIPS PUB)
180-1, Secure Hash Standard, 17 April 1995.
[8] Paila, T., Luby, M., Lehtonen, R., Roca, V., and R. Walsh,
"FLUTE - File Delivery over Unidirectional Transport", RFC
3926, October 2004.
[9] Perrig, A., Song, D., Canetti, R., Tygar, J., and B. Briscoe,
"Timed Efficient Stream Loss-Tolerant Authentication (TESLA):
Multicast Source Authentication Transform Introduction", RFC
4082, June 2005.
[10] Whetten, B., Vicisano, L., Kermode, R., Handley, M., Floyd, S.,
and M. Luby, "Reliable Multicast Transport Building Blocks for
One-to-Many Bulk-Data Transfer", RFC 3048, January 2001.
Authors' Addresses
Mark Watson
Digital Fountain
39141 Civic Center Drive
Suite 300
Fremont, CA 94538
U.S.A.
EMail: mark@digitalfountain.com
Michael Luby
Digital Fountain
39141 Civic Center Drive
Suite 300
Fremont, CA 94538
U.S.A.
EMail: luby@digitalfountain.com
Lorenzo Vicisano
Digital Fountain
39141 Civic Center Drive
Suite 300
Fremont, CA 94538
U.S.A.
EMail: lorenzo@digitalfountain.com
Full Copyright Statement
Copyright (C) The IETF Trust (2007).
This document is subject to the rights, licenses and restrictions
contained in BCP 78, and except as set forth therein, the authors
retain all their rights.
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Intellectual Property
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the procedures with respect to rights in RFC documents can be
found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at
ietf-ipr@ietf.org.
Acknowledgement
Funding for the RFC Editor function is currently provided by the
Internet Society.