| lpwan Working Group | A. Minaburo |
| Internet-Draft | Acklio |
| Intended status: Informational | L. Toutain |
| Expires: June 20, 2018 | IMT-Atlantique |
| C. Gomez | |
| Universitat Politècnica de Catalunya | |
| December 17, 2017 |
LPWAN Static Context Header Compression (SCHC) and fragmentation for IPv6 and UDP
draft-ietf-lpwan-ipv6-static-context-hc-08
This document describes a header compression scheme and fragmentation functionality for very low bandwidth networks. These techniques are specially tailored for Low Power Wide Area Network (LPWAN).
The Static Context Header Compression (SCHC) offers a great level of flexibility when processing the header fields. SCHC compression is based on a common static context stored in a LPWAN device and in the network. Static context means that the stored information does not change during packet transmission. The context describes the field values and keeps information that will not be transmitted through the constrained network.
SCHC must be used for LPWAN networks because it avoids complex resynchronization mechanisms, which are incompatible with LPWAN characteristics. And also, because with SCHC, in most cases IPv6/UDP headers can be reduced to a small identifier called Rule ID. Even though, sometimes, a SCHC compressed packet will not fit in one L2 PDU, and the SCHC fragmentation protocol defined in this document may be used.
This document describes the SCHC compression/decompression framework and applies it to IPv6/UDP headers. This document also specifies a fragmentation and reassembly mechanism that is used to support the IPv6 MTU requirement over LPWAN technologies. Fragmentation is mandatory for IPv6 datagrams that, after SCHC compression or when it has not been possible to apply such compression, still exceed the L2 maximum payload size. Similar solutions for other protocols such as CoAP will be described in separate documents.
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."
This Internet-Draft will expire on June 20, 2018.
Copyright (c) 2017 IETF Trust and the persons identified as the document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
Header compression is mandatory to efficiently bring Internet connectivity to the node within a LPWAN network. Some LPWAN networks properties can be exploited to get an efficient header compression:
The Static Context Header Compression (SCHC) is defined for this environment. SCHC uses a context where header information is kept in the header format order. This context is static (the values of the header fields do not change over time) avoiding complex resynchronization mechanisms, incompatible with LPWAN characteristics. In most of the cases, IPv6/UDP headers are reduced to a small context identifier.
The SCHC header compression mechanism is independent of the specific LPWAN technology over which it will be used.
LPWAN technologies are also characterized, among others, by a very reduced data unit and/or payload size [I-D.ietf-lpwan-overview]. However, some of these technologies do not support layer two fragmentation, therefore the only option for them to support the IPv6 MTU requirement of 1280 bytes [RFC2460] is the use of a fragmentation protocol at the adaptation layer below IPv6. This draft defines also a fragmentation functionality to support the IPv6 MTU requirement over LPWAN technologies. Such functionality has been designed under the assumption that data unit reordering will not happen between the entity performing fragmentation and the entity performing reassembly.
LPWAN technologies have similar architectures but different terminology. We can identify different types of entities in a typical LPWAN network, see Figure 1:
o Devices (Dev) are the end-devices or hosts (e.g. sensors, actuators, etc.). There can be a high density of devices per radio gateway.
o The Radio Gateway (RGW), which is the end point of the constrained link.
o The Network Gateway (NGW) is the interconnection node between the Radio Gateway and the Internet.
o LPWAN-AAA Server, which controls the user authentication and the applications.
o Application Server (App)
+------+
() () () | |LPWAN-|
() () () () / \ +---------+ | AAA |
() () () () () () / \=====| ^ |===|Server| +-----------+
() () () | | <--|--> | +------+ |APPLICATION|
() () () () / \==========| v |=============| (App) |
() () () / \ +---------+ +-----------+
Dev Radio Gateways NGW
Figure 1: LPWAN Architecture
This section defines the terminology and acronyms used in this document.
Static Context Header Compression (SCHC) avoids context synchronization, which is the most bandwidth-consuming operation in other header compression mechanisms such as RoHC [RFC5795]. Based on the fact that the nature of data flows is highly predictable in LPWAN networks, some static contexts may be stored on the Device (Dev). The contexts must be stored in both ends, and it can either be learned by a provisioning protocol or by out of band means or it can be pre-provisioned, etc. The way the context is learned on both sides are out of the scope of this document.
Dev App
+--------------+ +--------------+
|APP1 APP2 APP3| |APP1 APP2 APP3|
| | | |
| UDP | | UDP |
| IPv6 | | IPv6 |
| | | |
| SCHC C/D | | |
| (context) | | |
+-------+------+ +-------+------+
| +--+ +----+ +---------+ .
+~~ |RG| === |NGW | === |SCHC C/D |... Internet ..
+--+ +----+ |(context)|
+---------+
Figure 2: Architecture
Figure 2 represents the architecture for compression/decompression, it is based on [I-D.ietf-lpwan-overview] terminology. The Device is sending applications flows using IPv6 or IPv6/UDP protocols. These flows are compressed by a Static Context Header Compression Compressor/Decompressor (SCHC C/D) to reduce headers size. The resulting information is sent to a layer two (L2) frame to a LPWAN Radio Network (RG) which forwards the frame to a Network Gateway (NGW). The NGW sends the data to an SCHC C/D for decompression which shares the same rules with the Dev. The SCHC C/D can be located on the Network Gateway (NGW) or in another place as long as a tunnel is established between the NGW and the SCHC C/D. The SCHC C/D in both sides must share the same set of Rules. After decompression, the packet can be sent on the Internet to one or several LPWAN Application Servers (App).
The SCHC C/D process is bidirectional, so the same principles can be applied in the other direction.
The main idea of the SCHC compression scheme is to send the Rule id to the other end instead of sending known field values. This Rule id identifies a rule that matches as much as possible the original packet values. When a value is known by both ends, it is not necessary to send it through the LPWAN network.
The context contains a list of rules (cf. Figure 3). Each Rule contains itself a list of fields descriptions composed of a field identifier (FID), a field length (FL), a field position (FP), a direction indicator (DI), a target value (TV), a matching operator (MO) and a Compression/Decompression Action (CDA).
/-----------------------------------------------------------------\ | Rule N | /-----------------------------------------------------------------\| | Rule i || /-----------------------------------------------------------------\|| | (FID) Rule 1 ||| |+-------+--+--+--+------------+-----------------+---------------+||| ||Field 1|FL|FP|DI|Target Value|Matching Operator|Comp/Decomp Act|||| |+-------+--+--+--+------------+-----------------+---------------+||| ||Field 2|FL|FP|DI|Target Value|Matching Operator|Comp/Decomp Act|||| |+-------+--+--+--+------------+-----------------+---------------+||| ||... |..|..|..| ... | ... | ... |||| |+-------+--+--+--+------------+-----------------+---------------+||/ ||Field N|FL|FP|DI|Target Value|Matching Operator|Comp/Decomp Act||| |+-------+--+--+--+------------+-----------------+---------------+|/ | | \-----------------------------------------------------------------/
Figure 3: Compression/Decompression Context
The Rule does not describe the original packet format which must be known from the compressor/decompressor. The rule just describes the compression/decompression behavior for the header fields. In the rule, the description of the header field should be performed in the format packet order.
The Rule also describes the compressed header fields which are transmitted regarding their position in the rule which is used for data serialization on the compressor side and data deserialization on the decompressor side.
The Context describes the header fields and its values with the following entries:
Rule IDs are sent between both compression/decompression elements. The size of the Rule ID is not specified in this document, it is implementation-specific and can vary regarding the LPWAN technology, the number of flows, among others.
Some values in the Rule ID space are reserved for other functionalities than header compression as fragmentation. (See Section 5).
Rule IDs are specific to a Dev. Two Devs may use the same Rule ID for different header compression. To identify the correct Rule ID, the SCHC C/D needs to combine the Rule ID with the Dev L2 identifier to find the appropriate Rule.
The compression/decompression process follows several steps:
+--- ... --+-------------- ... --------------+-----------+--...--+ | Rule ID |Compressed Hdr Fields information| payload |padding| +--- ... --+-------------- ... --------------+-----------+--...--+
Figure 4: LPWAN Compressed Format Packet
Matching Operators (MOs) are functions used by both SCHC C/D endpoints involved in the header compression/decompression. They are not typed and can be applied indifferently to integer, string or any other data type. The result of the operation can either be True or False. MOs are defined as follows:
The Compression Decompression Action (CDA) describes the actions taken during the compression of headers fields, and inversely, the action taken by the decompressor to restore the original value.
/--------------------+-------------+----------------------------\ | Action | Compression | Decompression | | | | | +--------------------+-------------+----------------------------+ |not-sent |elided |use value stored in ctxt | |value-sent |send |build from received value | |mapping-sent |send index |value from index on a table | |LSB(length) |send LSB |TV OR received value | |compute-length |elided |compute length | |compute-checksum |elided |compute UDP checksum | |Deviid |elided |build IID from L2 Dev addr | |Appiid |elided |build IID from L2 App addr | \--------------------+-------------+----------------------------/
Figure 5: Compression and Decompression Functions
Figure 5 summarizes the basics functions defined to compress and decompress a field. The first column gives the action’s name. The second and third columns outline the compression/decompression behavior.
Compression is done in the rule order and compressed values are sent in that order in the compressed message. The receiver must be able to find the size of each compressed field which can be given by the rule or may be sent with the compressed header.
If the field is identified as being variable, then its size must be sent first using the following coding:
The not-sent function is generally used when the field value is specified in the rule and therefore known by the both Compressor and Decompressor. This action is generally used with the “equal” MO. If MO is “ignore”, there is a risk to have a decompressed field value different from the compressed field.
The compressor does not send any value in the compressed header for the field on which compression is applied.
The decompressor restores the field value with the target value stored in the matched rule.
The value-sent action is generally used when the field value is not known by both Compressor and Decompressor. The value is sent in the compressed message header. Both Compressor and Decompressor must know the size of the field, either implicitly (the size is known by both sides) or explicitly in the compressed header field by indicating the length. This function is generally used with the “ignore” MO.
The mapping-sent is used to send a smaller index associated with the list of values in the Target Value. This function is used together with the “match-mapping” MO.
The compressor looks on the TV to find the field value and send the corresponding index. The decompressor uses this index to restore the field value.
The number of bits sent is the minimal size for coding all the possible indexes.
LSB action is used to avoid sending the known part of the packet field header to the other end. This action is used together with the “MSB” MO. A length can be specified in the rule to indicate how many bits have to be sent. If the length is not specified, the number of bits sent is the field length minus the bits’ length specified in the MSB MO.
The compressor sends the “length” Least Significant Bits. The decompressor combines the value received with the Target Value.
If this action is made on a variable length field, the remaining size in byte has to be sent before.
These functions are used to process respectively the Dev and the App Interface Identifiers (Deviid and Appiid) of the IPv6 addresses. Appiid CDA is less common since current LPWAN technologies frames contain a single address.
The IID value may be computed from the Device ID present in the Layer 2 header. The computation is specific for each LPWAN technology and may depend on the Device ID size.
In the downstream direction, these CDA may be used to determine the L2 addresses used by the LPWAN.
These classes of functions are used by the decompressor to compute the compressed field value based on received information. Compressed fields are elided during compression and reconstructed during decompression.
In LPWAN technologies, the L2 data unit size typically varies from tens to hundreds of bytes. If the entire IPv6 datagram after applying SCHC header compression or when SCHC header compression is not possible, fits within a single L2 data unit, the fragmentation mechanism is not used and the packet is sent. Otherwise, the datagram SHALL be broken into fragments.
LPWAN technologies impose some strict limitations on traffic, devices are sleeping most of the time and may receive data during a short period of time after transmission to preserve battery. To adapt the SCHC fragmentation to the capabilities of LPWAN technologies, it is desirable to enable optional fragment retransmission and to allow a gradation of fragment delivery reliability. This document does not make any decision with regard to which fragment delivery reliability option(s) will be used over a specific LPWAN technology.
An important consideration is that LPWAN networks typically follow the star topology, and therefore data unit reordering is not expected in such networks. This specification assumes that reordering will not happen between the entity performing fragmentation and the entity performing reassembly. This assumption allows to reduce complexity and overhead of the fragmentation mechanism.
This subsection describes the different fields in the fragmentation header frames (see the fragmentation frames format in Section 5.4) that are used to enable the fragmentation functionalities defined in this document, and the different reliability options supported.
This specification defines the following three fragment delivery reliability options:
The same reliability option MUST be used for all fragments of a packet. It is up to implementers and/or representatives of the underlying LPWAN technology to decide which reliability option to use and whether the same reliability option applies to all IPv6 packets or not. Note that the reliability option to be used is not necessarily tied to the particular characteristics of the underlying L2 LPWAN technology (e.g. the No ACK reliability option may be used on top of an L2 LPWAN technology with symmetric characteristics for uplink and downlink).
This document does not make any decision as to which fragment delivery reliability option(s) are supported by a specific LPWAN technology.
Examples of the different reliability options described are provided in Appendix B.
This section defines the fragment format, the All-0 and All-1 frames format, the ACK format and the Abort frames format.
A fragment comprises a fragmentation header, a fragment payload, and Padding bits (if any). A fragment conforms to the format shown in Figure 6. The fragment payload carries a subset of either a SCHC header or an IPv6 header or the original IPv6 packet data payload. A fragment is the payload in the L2 protocol data unit (PDU).
+-----------------+-----------------------+---------+
| Fragment Header | Fragment payload | padding |
+-----------------+-----------------------+---------+
Figure 6: Fragment format.
In the No ACK option, fragments except the last one SHALL contain the format as defined in Figure 7. The total size of the fragmentation header is R bits.
<------------ R ---------->
<--T--> <--N-->
+-- ... --+- ... -+- ... -+---...---+-+
| Rule ID | DTag | FCN | payload |P|
+-- ... --+- ... -+- ... -+---...---+-+
Figure 7: Fragmentation Format for Fragments except the Last One, No ACK option
In any of the Window mode options, fragments except the last one SHALL contain the fragmentation format as defined in Figure 8. The total size of this fragmentation header is R bits.
<------------ R ---------->
<--T--> 1 <--N-->
+-- ... --+- ... -+-+- ... -+---...---+-+
| Rule ID | DTag |W| FCN | payload |P|
+-- ... --+- ... -+-+- ... -+---...---+-+
Figure 8: Fragmentation Format for Fragments except the Last One, Window mode
The format of an ACK that acknowledges a window that is not the last one (denoted as ALL-0 window) is shown in Figure 9.
<-------- R ------->
<- T -> 1
+---- ... --+-... -+-+----- ... ---+
| Rule ID | DTag |W| bitmap | (no payload)
+---- ... --+-... -+-+----- ... ---+
Figure 9: ACK format for All-0 windows
To acknowledge the last window of a packet (denoted as All-1 window), a C bit (i.e. MIC checked) following the W bit is set to 1 to indicate that the MIC check computed by the receiver matches the MIC present in the ALL-1 fragment. If the MIC check fails, the C bit is set to 0 and the Bitmap for the All-1 window follows.
<-------- R -------> <- byte boundary ->
<- T -> 1 1
+---- ... --+-... -+-+-+
| Rule ID | DTag |W|1| (MIC correct)
+---- ... --+-... -+-+-+
+---- ... --+-... -+-+-+------- ... -------+
| Rule ID | DTag |W|0| bitmap | (MIC Incorrect)
+---- ... --+-... -+-+-+------- ... -------+
C
Figure 10: Format of an ACK for All-1 windows
The All-0 format is used for the last fragment of a window that is not the last window of the packet.
<------------ R ------------>
<- T -> 1 <- N ->
+-- ... --+- ... -+-+- ... -+--- ... ---+
| Rule ID | DTag |W| 0..0 | payload |
+-- ... --+- ... -+-+- ... -+--- ... ---+
Figure 11: All-0 fragment format
The All-0 empty fragment format is used by a sender to request an ACK in ACK-Always mode
<------------ R ------------>
<- T -> 1 <- N ->
+-- ... --+- ... -+-+- ... -+
| Rule ID | DTag |W| 0..0 | (no payload)
+-- ... --+- ... -+-+- ... -+
Figure 12: All-0 empty fragment format
In the No ACK option, the last fragment of an IPv6 datagram SHALL contain a fragmentation header that conforms to the format shown in Figure 13. The total size of this fragmentation header is R+M bits.
<------------- R ---------->
<- T -> <-N-><----- M ----->
+---- ... ---+- ... -+-----+---- ... ----+---...---+
| Rule ID | DTag | 1 | MIC | payload |
+---- ... ---+- ... -+-----+---- ... ----+---...---+
Figure 13: All-1 Fragmentation Format for the Last Fragment, No ACK option
In any of the Window modes, the last fragment of an IPv6 datagram SHALL contain a fragmentation header that conforms to the format shown in Figure 14. The total size of the fragmentation header in this format is R+M bits. It is used for request a retransmission
<------------ R ------------>
<- T -> 1 <- N -> <---- M ----->
+-- ... --+- ... -+-+- ... -+---- ... ----+---...---+
| Rule ID | DTag |W| 11..1 | MIC | payload |
+-- ... --+- ... -+-+- ... -+---- ... ----+---...---+
(FCN)
Figure 14: All-1 Fragmentation Format for the Last Fragment, Window mode
In either ACK-Always or ACK-on-error, in order to request a retransmission of the ACK for the All-1 window, the fragment sender uses the format shown in Figure 15. The total size of the fragmentation header in this format is R+M bits.
<------------ R ------------>
<- T -> 1 <- N -> <---- M ----->
+-- ... --+- ... -+-+- ... -+---- ... ----+
| Rule ID | DTag |W| 1..1 | MIC | (no payload)
+-- ... --+- ... -+-+- ... -+---- ... ----+
Figure 15: All-1 for Retries format fragment also called All-1 empty
The values for R, N, T and M are not specified in this document, and have to be determined in other documents (e.g. technology-specific profile documents).
The All-1 Abort format and the ACK abort have the following formats.
<------ byte boundary ------><--- 1 byte ---> +--- ... ---+- ... -+-+-...-+-+-+-+-+-+-+-+-+ | Rule ID | DTag |W| FCN | FF | (no MIC & no payload) +--- ... ---+- ... -+-+-...-+-+-+-+-+-+-+-+-+
Figure 16: All-1 Abort format
<------ byte boundary -----><--- 1 byte ---> +---- ... --+-... -+-+-+-+-+-+-+-+-+-+-+-+-+ | Rule ID | DTag |W| 1..1| FF | +---- ... --+-... -+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 17: ACK Abort format
The fragment receiver needs to identify all the fragments that belong to a given IPv6 datagram. To this end, the receiver SHALL use:
Then, the fragment receiver may determine the fragment delivery reliability option that is used for this fragment based on the Rule ID field in that fragment.
Upon receipt of a link fragment, the receiver starts constructing the original unfragmented packet. It uses the FCN and the order of arrival of each fragment to determine the location of the individual fragments within the original unfragmented packet. A fragment payload may carry bytes from a SCHC compressed IPv6 header, an uncompressed IPv6 header or an IPv6 datagram data payload. An unfragmented packet could be a SCHC compressed or an uncompressed IPv6 packet (header and data). For example, the receiver may place the fragment payload within a payload datagram reassembly buffer at the location determined from: the FCN, the arrival order of the fragments, and the fragment payload sizes. In Window mode, the fragment receiver also uses the W bit in the received fragments. Note that the size of the original, unfragmented packet cannot be determined from fragmentation headers.
Fragmentation functionality uses the FCN value, which has a length of N bits. The All-1 and All-0 FCN values are used to control the fragmentation transmission. The FCN will be assigned sequentially in a decreasing order starting from 2^N-2, i.e. the highest possible FCN value depending on the FCN number of bits, but excluding the All-1 value. In all modes, the last fragment of a packet must contains a MIC which is used to check if there are errors or missing fragments, and must use the corresponding All-1 fragment format. Also note that, a fragment with an All-0 format is considered the last fragment of the current window.
If the recipient receives the last fragment of a datagram (All-1), it checks for the integrity of the reassembled datagram, based on the MIC received. In No ACK, if the integrity check indicates that the reassembled datagram does not match the original datagram (prior to fragmentation), the reassembled datagram MUST be discarded. In Window mode, a MIC check is also performed by the fragment receiver after reception of each subsequent fragment retransmitted after the first MIC check.
In the No ACK mode there is no feedback communication from the fragment receiver. The sender will send the fragments of a packet until the last one without any possibility to know if errors or a losses have occurred. As in this mode there is not a need to identify specific fragments a one-bit FCN is used, therefore FCN All-0 will be used in all fragments except the last one. The latter will carry an All-1 FCN and will also carry the MIC. The receiver will wait for fragments and will set the Inactivity timer. The No ACK mode will use the MIC contained in the last fragment to check error. When the Inactivity Timer expires or when the MIC check indicates that the reassembled packet does not match the originall one, the receiver will release all resources allocated to reassembly of the packet. The initial value of the Inactivity Timer will be determined based on the characteristics of the underlying LPWAN technology and will be defined in other documents (e.g. technology-specific profile documents).
In Window modes, a jumping window protocol is using two windows alternatively, 0 and 1. An FCN set to All-0 indicates that the window is over (i.e. the fragment is the last one of the window) and allows to switch from one window to another. The All-1 FCN in a fragment indicates that it is the last fragment of the packet and therefore there will not be another window for the packet.
The Window mode offers two different reliability options modes: ACK-on-error and ACK-always.
The sender starts sending fragments using the two windows procedure. A delay between each fragment can be added to respect regulation rules or constraints imposed by the applications. Each time a fragment is sent the FCN is decreased by one and the sending information is set locally. When the FCN reaches value 0 and there are more fragments to be sent, an All-0 fragment is sent and the retransmission timer is set. The sender waits for an ACK to know if there were some transmission errors. If there are some errors the receiver sends an ACK with the corresponding errors in the Bitmap, otherwise, an ACK without Bitmap will be sent and a new window should be sent. When the last fragment is sent, and All-1 fragment with MIC is sent. The sender sets the retransmission timer to wait for the ACK corresponding to the last window. During this period, the sender starts listening to the radio and starts an Inactivity timer, which is dimensioned based on the received window available for the LPWAN technology in use. If the Inactivity timer expires an empty All-0 (or All-1 if the last fragment has been sent) fragment is sent to ask the receiver to resend its ACK. The window number is not changed.
When the sender receives an ACK, it checks the window value. The ACK fragments carrying an unexpected W bit are discarded. If the window number of the received ACK is correct, the sender compares the sending information with the received Bitmap. If the sending information is equal to the received Bitmap all the fragments sent during the window have been well received. If at least one fragment needs to be sent, the sender moves its sending window to the next value and sends the last fragment. If no more fragments have to be sent, then the fragmented packet transmission is finished.
If some fragments are missing (not set in the Bitmap) then the sender resends the missing fragments. When the retransmission is finished, it starts the retransmission timer (even if an All-0 or an All-1 has not been sent during the retransmission) and waits for ACK.
If the sending information is different from the received Bitmap the counter Attempts is increased and the sender resends the missing fragments again when a MAX_ACK_REQUESTS is reached, the sender sends an Abort and drops the fragments. The sender Aborts the transmission when a corrupt MIC has been received or when MAX_ACK_REQUESTS has reached.
At the beginning, the receiver side expects to receive window 0. Any fragment not belonging to the current window is discarded. All Fragments belonging to the correct window are accepted, the fragment number is computed based on the FCN value. The receiver updates the Bitmap with the correct received fragments.
When All-0 fragment is received, it indicates that all the fragments have been sent in the current window. Since the sender is not obliged to send a full window, some fragment number not set in the memory may not correspond to losses. It sends the corresponding ACK and the next window can start.
When All-1 fragment is received, it indicates that the transmission is finished. Since the last window is not full, the MIC will be used to detect if all the fragments have been received. A correct MIC indicates the end of the transmission but the receiver must stay alive an Inactivity timer period to answer to empty All-1 fragment the sender may send if the ACK is lost.
If All-1 fragment has not been received, the receiver expects a new window. It waits for the next fragment. If the window value has not changed, the received fragments are part of a retransmission. A receiver that has already received a fragment should discard it, otherwise, it updates the Bitmap. If all the bits of the Bitmap are set to one, the receiver may send an ACK without waiting for an All-0 fragment.
If the window value is set to the next value, this means that the sender has received a correct bitmap, which means that all the fragments have been received. The receiver changes the value of the expected window.
If the receiver receives an All-0 fragment, the sender may send one or more fragments per window. Otherwise, some fragments in the window have been lost.
If the receiver receives an All-1 fragment this means that the transmission should be finished. If the MIC is incorrect some fragments have been lost. It sends the ACK. In case of an incorrect MIC, the receivers wait for fragments belonging to the same window. After MAX_ACK_REQUESTS the receiver will Abort the transmission. It can also Abort when the Inactivity timer has expired.
The ACK-on-error is similar to the ACK-Always procedure, the difference is that in ACK-on-error the ACK is not sent at the end of each window but only when there is an error. In Ack-on-error mode, the retransmission timer expiration will be considered as a positive acknowledgment, it is set when receiving an All-0 or an All-1 fragment. If there are no more fragments then the fragmentation is finished.
When the All-1 last fragment is sent and the correct MIC have been received an ACK is sent to confirms the end of the correct transmission. If the retransmission timer expires an All-1 empty request the last ACK that MUST be sent to complete the fragmentation transmission.
If the sender receives an ACK, it checks the window value. ACKs with the non-expected window number are discarded. If the window number on the received Bitmap is correct, the sender verifies if the receiver has all the fragments. When all the fragments have been received the transmission of a new window should continue. Otherwise, when there is an error the counter Attempts is increased and the sender resends the missing fragments again. When a MAX_ACK_REQUESTS is reached, the sender sends an Abort. When the retransmission is finished, it starts listening to the ACK (even if an All-0 or an All-1 has not been sent during the retransmission) and set the retransmission timer. If the retransmission timer expires the transmission is aborted.
Unlike the sender, the receiver for ACK-on-error has some differences. First, we are not sending ACK unless there is an error or an unexpected behavior. The receiver starts with the expected window and maintains the information indicating which fragments it has received (All-0 and All-1 occupy the same position). After receiving a fragment an Inactivity timer is set, if nothing has been received and the Inactivity timer expires the transmission is aborted.
Any fragment not belonging to the current window is discarded. The Fragment Number is computed based on the FCN value. When an All-0 fragment is received and the Bitmap is full, the receiver changes the window value.
An All-0 fragment and not a full bitmap indicate that all the fragments have been sent in the current window. Since the sender is not obligated to send a full window, some fragment number not used may not correspond to losses. As the receiver does not know if the missing fragments are lost or normal missing fragments, it sends an ACK.
An All-1 fragment indicates that the transmission is finished. Since the last window is not full, the MIC will be used to detect if all the fragments have been received. A correct MIC indicates the end of the transmission.
If All-1 fragment has not been received, the receiver expects a new window. It waits for the next fragment. If the window value has not changed, the received fragments are part of a retransmission. A receiver that has already received a fragment should discard it. If all the bits of the Bitmap are set to one, the receiver waits for the next window without waiting for an All-0 fragment and it does not send an ACK either. While the receiver waits for next window and if the window value is set to the next value, and if an All-1 fragment with the next value window arrived the receiver goes to error and abort the transmission, it drops the fragments.
If the receiver receives an All-1 fragment this means that the transmission should be finished. If the MIC is incorrect some fragments have been lost. It sends an ACK.
In case of an incorrect MIC, the receivers wait for fragment belonging to the same window or the expiration of the Inactivity timer which will Abort the transmission.
The Fragmentation Bitmap is the optmization of what has been received. It is transmitted in the ACK format fragment when there are some missing fragments. An ACK message may introduce padding at the end to align transmitted data to a byte boundary. The first byte boundary includes one or more complete bytes, depending on the size of Rule ID and DTag.
The receiver generates the Bitmap which may have the size of a single downlink frame of the LPWAN technology used. To avoid this problem the FCN size should be set to the corresponding downlink size minus 1 bit for C bit. The C bit will be sent only in the ACK for the last frame of the packet (All-1).
<---- bitmap fragments ---->
| Rule ID | DTag |W|C|0|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|0|
|--- byte boundary ----| 1 byte next | 1 byte next |
Figure 18: Bitmap
Bitmap transmitted MUST be optimized in size to reduce frame size. The right-most bytes with all Bitmap bit set to 1 MUST be removed from the transmission. As the receiver knows the Bitmap size, it can reconstruct the value. In the example Figure 19 the last 2 bytes of the bitmap are set to 1, therefore, they are not sent.
In the last window, when checked bit C value is one, means that the MIC is corrected and the Bitmap is not sent. Otherwise, the Bitmap needs to be sent after the C bit. Note that the introduction of a C bit may force to reduce the number of fragments to allow the bitmap to fit in a frame.
<------- R ------->
<- T -> 1
+---- ... --+-... -+-+-+-+
| Rule ID | DTag |W|1|0|
+---- ... --+-... -+-+-+-+
|---- byte boundary -----|
Figure 19: Bitmap transmitted fragment format
Figure 20 shows an example of an ACK (N=3), where the Bitmap indicates that the second and the fifth fragments have not been correctly received.
<------ R ------>6 5 4 3 2 1 0 (*)
<- T -> 1
| Rule ID | DTag |W|1|0|1|1|0|1|all-0|padding| Bitmap
|--- byte boundary ----| 1 byte next |
(*)=(FCN values indicating the order)
+---- ... --+-... -+-+-+-+-+-+-+-+-+-+
| Rule ID | DTag |W|1|0|1|1|0|1|1|P| transmitted Bitmap
+---- ... --+-... -+-+-+-+-+-+-+-+-+-+
|--- byte boundary ----| 1 byte next |
Figure 20: Example of the bitmap in Window mode, in any window except the last one, for N=3)
Figure 21 shows an example of an ACK (N=3), where the bitmap indicates that the MIC check has failed but there is no missing fragments.
<------- R -------> 6 5 4 3 2 1 7 (*)
<- T -> 1 1
| Rule ID | DTag |W|0|1|1|1|1|1|1|1|padding| Bitmap
|---- byte boundary ----| 1 byte next | 1 byte next |
C
+---- ... --+-... -+-+-+-+
| Rule ID | DTag |W|0|1| transmitted Bitmap
+---- ... --+-... -+-+-+-+
|---- byte boundary -----|
(*) = (FCN values indicating the order)
Figure 21: Example of the Bitmap in Window mode for the last window, for N=3)
For ACK-Always or ACK-on-error, implementers may opt to support a single window size or multiple window sizes. The latter, when feasible, may provide performance optimizations. For example, a large window size may be used for packets that need to be carried by a large number of fragments. However, when the number of fragments required to carry a packet is low, a smaller window size, and thus a shorter bitmap, may be sufficient to provide feedback on all fragments. If multiple window sizes are supported, the Rule ID may be used to signal the window size in use for a specific packet transmission.
Note that the same window size MUST be used for the transmission of all fragments that belong to a packet.
In some LPWAN technologies, as part of energy-saving techniques, downlink transmission is only possible immediately after an uplink transmission. In order to avoid potentially high delay for fragmented datagram transmission in the downlink, the fragment receiver MAY perform an uplink transmission as soon as possible after reception of a fragment that is not the last one. Such uplink transmission may be triggered by the L2 (e.g. an L2 ACK sent in response to a fragment encapsulated in a L2 frame that requires an L2 ACK) or it may be triggered from an upper layer.
For fragmented packet transmission in the downlink, and when ACK Always is used, the fragment receiver MAY support timer-based ACK retransmission. In this mechanism, the fragment receiver initializes and starts a timer (the Inactivity Timer is used) after the transmission of an ACK, except when the ACK is sent in response to the last fragment of a packet (All-1 fragment). In the latter case, the fragment receiver does not start a timer after transmission of the ACK.
If, after transmission of an ACK that is not an All-1 fragment, and before expiration of the corresponding Inactivity timer, the fragment receiver receives a fragment that belongs to the current window (e.g. a missing fragment from the current window) or to the next window, the Inactivity timer for the ACK is stopped. However, if the Inactivity timer expires, the ACK is resent and the Inactivity timer is reinitialized and restarted.
The default initial value for the Inactivity timer, as well as the maximum number of retries for a specific ACK, denoted MAX_ACK_RETRIES, are not defined in this document, and need to be defined in other documents (e.g. technology-specific profiles). The initial value of the Inactivity timer is expected to be greater than that of the Retransmission timer, in order to make sure that a (buffered) fragment to be retransmitted can find an opportunity for that transmission.
When the fragment sender transmits the All-1 fragment, it initializes and starts its retransmission timer to a long value (e.g. several times the initial Inactivity timer). If an ACK is received before expiration of this timer, the fragment sender retransmits any lost fragments reported by the ACK, or if the ACK confirms successful reception of all fragments of the last window, transmission of the fragmented packet ends. If the timer expires, and no ACK has been received since the start of the timer, the fragment sender assumes that the all-1 fragment has been successfully received (and possibly, the last ACK has been lost: this mechanism assumes that the retransmission timer for the all-1 fragment is long enough to allow several ACK retries if the all-1 fragment has not been received by the fragment receiver, and it also assumes that it is unlikely that several ACKs become all lost).
SCHC header, either for compression, fragmentation or acknowledgment does not preserve byte alignment. Since most of the LPWAN network technologies payload is expressed in an integer number of bytes; the sender will introduce at the end some padding bits while the receiver must be able to eliminate them.
The algorithm for padding bit elimination for compressed or fragmented frames is simple. Based on the following principle: * The SCHC header is not aligned on a byte boundary, but its size in bits is given by the rule.
In that case, a receiver after decoding the SCHC header, must take the maximum multiple of 8 bits as data. The remaining bits are padding bits.
This section lists the different IPv6 and UDP header fields and how they can be compressed.
This field always holds the same value. Therefore, the TV is 6, the MO is “equal” and the “CDA “not-sent””.
If the DiffServ field identified by the rest of the rule do not vary and is known by both sides, the TV should contain this well-known value, the MO should be “equal” and the CDA must be “not-sent.
If the DiffServ field identified by the rest of the rule varies over time or is not known by both sides, then there are two possibilities depending on the variability of the value, the first one is to do not compressed the field and sends the original value, or the second where the values can be computed by sending only the LSB bits:
If the Flow Label field identified by the rest of the rule does not vary and is known by both sides, the TV should contain this well-known value, the MO should be “equal” and the CDA should be “not-sent”.
If the Flow Label field identified by the rest of the rule varies during time or is not known by both sides, there are two possibilities depending on the variability of the value, the first one is without compression and then the value is sent and the second where only part of the value is sent and the decompressor needs to compute the original value:
If the LPWAN technology does not add padding, this field can be elided for the transmission on the LPWAN network. The SCHC C/D recomputes the original payload length value. The TV is not set, the MO is set to “ignore” and the CDA is “compute-IPv6-length”.
If the payload length needs to be sent and does not need to be coded in 16 bits, the TV can be set to 0x0000, the MO set to “MSB (16-s)” and the CDA to “LSB”. The ‘s’ parameter depends on the expected maximum packet length.
On other cases, the payload length field must be sent and the CDA is replaced by “value-sent”.
If the Next Header field identified by the rest of the rule does not vary and is known by both sides, the TV should contain this Next Header value, the MO should be “equal” and the CDA should be “not-sent”.
If the Next header field identified by the rest of the rule varies during time or is not known by both sides, then TV is not set, MO is set to “ignore” and CDA is set to “value-sent”. A matching-list may also be used.
The End System is generally a device and does not forward packets. Therefore, the Hop Limit value is constant. So, the TV is set with a default value, the MO is set to “equal” and the CDA is set to “not-sent”.
Otherwise the value is sent on the LPWAN: TV is not set, MO is set to ignore and CDA is set to “value-sent”.
Note that the field behavior differs in upstream and downstream. In upstream, since there is no IP forwarding between the Dev and the SCHC C/D, the value is relatively constant. On the other hand, the downstream value depends of Internet routing and may change more frequently. One solution could be to use the Direction Indicator (DI) to distinguish both directions to elide the field in the upstream direction and send the value in the downstream direction.
As in 6LoWPAN [RFC4944], IPv6 addresses are split into two 64-bit long fields; one for the prefix and one for the Interface Identifier (IID). These fields should be compressed. To allow a single rule, these values are identified by their role (DEV or APP) and not by their position in the frame (source or destination). The SCHC C/D must be aware of the traffic direction (upstream, downstream) to select the appropriate field.
Both ends must be synchronized with the appropriate prefixes. For a specific flow, the source and destination prefix can be unique and stored in the context. It can be either a link-local prefix or a global prefix. In that case, the TV for the source and destination prefixes contain the values, the MO is set to “equal” and the CDA is set to “not-sent”.
In case the rule allows several prefixes, mapping-list must be used. The different prefixes are listed in the TV associated with a short ID. The MO is set to “match-mapping” and the CDA is set to “mapping-sent”.
Otherwise the TV contains the prefix, the MO is set to “equal” and the CDA is set to value-sent.
If the DEV or APP IID are based on an LPWAN address, then the IID can be reconstructed with information coming from the LPWAN header. In that case, the TV is not set, the MO is set to “ignore” and the CDA is set to “DEViid” or “APPiid”. Note that the LPWAN technology is generally carrying a single device identifier corresponding to the DEV. The SCHC C/D may also not be aware of these values.
If the DEV address has a static value that is not derived from an IEEE EUI-64, then TV contains the actual Dev address value, the MO operator is set to “equal” and the CDA is set to “not-sent”.
If several IIDs are possible, then the TV contains the list of possible IIDs, the MO is set to “match-mapping” and the CDA is set to “mapping-sent”.
Otherwise the value variation of the IID may be reduced to few bytes. In that case, the TV is set to the stable part of the IID, the MO is set to MSB and the CDA is set to LSB.
Finally, the IID can be sent on the LPWAN. In that case, the TV is not set, the MO is set to “ignore” and the CDA is set to “value-sent”.
No extension rules are currently defined. They can be based on the MOs and CDAs described above.
To allow a single rule, the UDP port values are identified by their role (DEV or APP) and not by their position in the frame (source or destination). The SCHC C/D must be aware of the traffic direction (upstream, downstream) to select the appropriate field. The following rules apply for DEV and APP port numbers.
If both ends know the port number, it can be elided. The TV contains the port number, the MO is set to “equal” and the CDA is set to “not-sent”.
If the port variation is on few bits, the TV contains the stable part of the port number, the MO is set to “MSB” and the CDA is set to “LSB”.
If some well-known values are used, the TV can contain the list of these values, the MO is set to “match-mapping” and the CDA is set to “mapping-sent”.
Otherwise the port numbers are sent on the LPWAN. The TV is not set, the MO is set to “ignore” and the CDA is set to “value-sent”.
If the LPWAN technology does not introduce padding, the UDP length can be computed from the received data. In that case, the TV is not set, the MO is set to “ignore” and the CDA is set to “compute-UDP-length”.
If the payload is small, the TV can be set to 0x0000, the MO set to “MSB” and the CDA to “LSB”.
On other cases, the length must be sent and the CDA is replaced by “value-sent”.
IPv6 mandates a checksum in the protocol above IP. Nevertheless, if a more efficient mechanism such as L2 CRC or MIC is carried by or over the L2 (such as in the LPWAN fragmentation process (see Section 5)), the UDP checksum transmission can be avoided. In that case, the TV is not set, the MO is set to “ignore” and the CDA is set to “compute-UDP-checksum”.
In other cases, the checksum must be explicitly sent. The TV is not set, the MO is set to “ignore” and the CDF is set to “value-sent”.
A malicious header compression could cause the reconstruction of a wrong packet that does not match with the original one, such corruption may be detected with end-to-end authentication and integrity mechanisms. Denial of Service may be produced but its arise other security problems that may be solved with or without header compression.
This subsection describes potential attacks to LPWAN fragmentation and suggests possible countermeasures.
A node can perform a buffer reservation attack by sending a first fragment to a target. Then, the receiver will reserve buffer space for the IPv6 packet. Other incoming fragmented packets will be dropped while the reassembly buffer is occupied during the reassembly timeout. Once that timeout expires, the attacker can repeat the same procedure, and iterate, thus creating a denial of service attack. The (low) cost to mount this attack is linear with the number of buffers at the target node. However, the cost for an attacker can be increased if individual fragments of multiple packets can be stored in the reassembly buffer. To further increase the attack cost, the reassembly buffer can be split into fragment-sized buffer slots. Once a packet is complete, it is processed normally. If buffer overload occurs, a receiver can discard packets based on the sender behavior, which may help identify which fragments have been sent by an attacker.
In another type of attack, the malicious node is required to have overhearing capabilities. If an attacker can overhear a fragment, it can send a spoofed duplicate (e.g. with random payload) to the destination. If the LPWAN technology does not support suitable protection (e.g. source authentication and frame counters to prevent replay attacks), a receiver cannot distinguish legitimate from spoofed fragments. Therefore, the original IPv6 packet will be considered corrupt and will be dropped. To protect resource-constrained nodes from this attack, it has been proposed to establish a binding among the fragments to be transmitted by a node, by applying content-chaining to the different fragments, based on cryptographic hash functionality. The aim of this technique is to allow a receiver to identify illegitimate fragments.
Further attacks may involve sending overlapped fragments (i.e. comprising some overlapping parts of the original IPv6 datagram). Implementers should make sure that correct operation is not affected by such event.
In Window mode – ACK on error, a malicious node may force a fragment sender to resend a fragment a number of times, with the aim to increase consumption of the fragment sender’s resources. To this end, the malicious node may repeatedly send a fake ACK to the fragment sender, with a bitmap that reports that one or more fragments have been lost. In order to mitigate this possible attack, MAX_FRAG_RETRIES may be set to a safe value which allows to limit the maximum damage of the attack to an acceptable extent. However, note that a high setting for MAX_FRAG_RETRIES benefits fragment delivery reliability, therefore the trade-off needs to be carefully considered.
Thanks to Dominique Barthel, Carsten Bormann, Philippe Clavier, Arunprabhu Kandasamy, Antony Markovski, Alexander Pelov, Pascal Thubert, Juan Carlos Zuniga and Diego Dujovne for useful design consideration and comments.
| [RFC2460] | Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, December 1998. |
| [RFC4944] | Montenegro, G., Kushalnagar, N., Hui, J. and D. Culler, "Transmission of IPv6 Packets over IEEE 802.15.4 Networks", RFC 4944, DOI 10.17487/RFC4944, September 2007. |
| [RFC5795] | Sandlund, K., Pelletier, G. and L-E. Jonsson, "The RObust Header Compression (ROHC) Framework", RFC 5795, DOI 10.17487/RFC5795, March 2010. |
| [RFC7136] | Carpenter, B. and S. Jiang, "Significance of IPv6 Interface Identifiers", RFC 7136, DOI 10.17487/RFC7136, February 2014. |
| [I-D.ietf-lpwan-overview] | Farrell, S., "LPWAN Overview", Internet-Draft draft-ietf-lpwan-overview-07, October 2017. |
This section gives some scenarios of the compression mechanism for IPv6/UDP. The goal is to illustrate the SCHC behavior.
The most common case using the mechanisms defined in this document will be a LPWAN Dev that embeds some applications running over CoAP. In this example, three flows are considered. The first flow is for the device management based on CoAP using Link Local IPv6 addresses and UDP ports 123 and 124 for Dev and App, respectively. The second flow will be a CoAP server for measurements done by the Device (using ports 5683) and Global IPv6 Address prefixes alpha::IID/64 to beta::1/64. The last flow is for legacy applications using different ports numbers, the destination IPv6 address prefix is gamma::1/64.
Figure 22 presents the protocol stack for this Device. IPv6 and UDP are represented with dotted lines since these protocols are compressed on the radio link.
Management Data
+----------+---------+---------+
| CoAP | CoAP | legacy |
+----||----+---||----+---||----+
. UDP . UDP | UDP |
................................
. IPv6 . IPv6 . IPv6 .
+------------------------------+
| SCHC Header compression |
| and fragmentation |
+------------------------------+
| LPWAN L2 technologies |
+------------------------------+
DEV or NGW
Figure 22: Simplified Protocol Stack for LP-WAN
Note that in some LPWAN technologies, only the Devs have a device ID. Therefore, when such technologies are used, it is necessary to define statically an IID for the Link Local address for the SCHC C/D.
Rule 0 +----------------+--+--+--+---------+--------+------------++------+ | Field |FL|FP|DI| Value | Match | Comp Decomp|| Sent | | | | | | | Opera. | Action ||[bits]| +----------------+--+--+--+---------+---------------------++------+ |IPv6 version |4 |1 |Bi|6 | equal | not-sent || | |IPv6 DiffServ |8 |1 |Bi|0 | equal | not-sent || | |IPv6 Flow Label |20|1 |Bi|0 | equal | not-sent || | |IPv6 Length |16|1 |Bi| | ignore | comp-length|| | |IPv6 Next Header|8 |1 |Bi|17 | equal | not-sent || | |IPv6 Hop Limit |8 |1 |Bi|255 | ignore | not-sent || | |IPv6 DEVprefix |64|1 |Bi|FE80::/64| equal | not-sent || | |IPv6 DEViid |64|1 |Bi| | ignore | DEViid || | |IPv6 APPprefix |64|1 |Bi|FE80::/64| equal | not-sent || | |IPv6 APPiid |64|1 |Bi|::1 | equal | not-sent || | +================+==+==+==+=========+========+============++======+ |UDP DEVport |16|1 |Bi|123 | equal | not-sent || | |UDP APPport |16|1 |Bi|124 | equal | not-sent || | |UDP Length |16|1 |Bi| | ignore | comp-length|| | |UDP checksum |16|1 |Bi| | ignore | comp-chk || | +================+==+==+==+=========+========+============++======+ Rule 1 +----------------+--+--+--+---------+--------+------------++------+ | Field |FL|FP|DI| Value | Match | Action || Sent | | | | | | | Opera. | Action ||[bits]| +----------------+--+--+--+---------+--------+------------++------+ |IPv6 version |4 |1 |Bi|6 | equal | not-sent || | |IPv6 DiffServ |8 |1 |Bi|0 | equal | not-sent || | |IPv6 Flow Label |20|1 |Bi|0 | equal | not-sent || | |IPv6 Length |16|1 |Bi| | ignore | comp-length|| | |IPv6 Next Header|8 |1 |Bi|17 | equal | not-sent || | |IPv6 Hop Limit |8 |1 |Bi|255 | ignore | not-sent || | |IPv6 DEVprefix |64|1 |Bi|[alpha/64, match- |mapping-sent|| [1] | | | | | |fe80::/64] mapping| || | |IPv6 DEViid |64|1 |Bi| | ignore | DEViid || | |IPv6 APPprefix |64|1 |Bi|[beta/64,| match- |mapping-sent|| [2] | | | | | |alpha/64,| mapping| || | | | | | |fe80::64]| | || | |IPv6 APPiid |64|1 |Bi|::1000 | equal | not-sent || | +================+==+==+==+=========+========+============++======+ |UDP DEVport |16|1 |Bi|5683 | equal | not-sent || | |UDP APPport |16|1 |Bi|5683 | equal | not-sent || | |UDP Length |16|1 |Bi| | ignore | comp-length|| | |UDP checksum |16|1 |Bi| | ignore | comp-chk || | +================+==+==+==+=========+========+============++======+ Rule 2 +----------------+--+--+--+---------+--------+------------++------+ | Field |FL|FP|DI| Value | Match | Action || Sent | | | | | | | Opera. | Action ||[bits]| +----------------+--+--+--+---------+--------+-------------++------+ |IPv6 version |4 |1 |Bi|6 | equal | not-sent || | |IPv6 DiffServ |8 |1 |Bi|0 | equal | not-sent || | |IPv6 Flow Label |20|1 |Bi|0 | equal | not-sent || | |IPv6 Length |16|1 |Bi| | ignore | comp-length|| | |IPv6 Next Header|8 |1 |Bi|17 | equal | not-sent || | |IPv6 Hop Limit |8 |1 |Up|255 | ignore | not-sent || | |IPv6 Hop Limit |8 |1 |Dw| | ignore | value-sent || [8] | |IPv6 DEVprefix |64|1 |Bi|alpha/64 | equal | not-sent || | |IPv6 DEViid |64|1 |Bi| | ignore | DEViid || | |IPv6 APPprefix |64|1 |Bi|gamma/64 | equal | not-sent || | |IPv6 APPiid |64|1 |Bi|::1000 | equal | not-sent || | +================+==+==+==+=========+========+============++======+ |UDP DEVport |16|1 |Bi|8720 | MSB(12)| LSB(4) || [4] | |UDP APPport |16|1 |Bi|8720 | MSB(12)| LSB(4) || [4] | |UDP Length |16|1 |Bi| | ignore | comp-length|| | |UDP checksum |16|1 |Bi| | ignore | comp-chk || | +================+==+==+==+=========+========+============++======+
Figure 23: Context rules
All the fields described in the three rules depicted on Figure 23 are present in the IPv6 and UDP headers. The DEViid-DID value is found in the L2 header.
The second and third rules use global addresses. The way the Dev learns the prefix is not in the scope of the document.
The third rule compresses port numbers to 4 bits.
This section provides examples of different fragment delivery reliability options possible on the basis of this specification.
Figure 24 illustrates the transmission of an IPv6 packet that needs 11 fragments in the No ACK option, FCN is always 1 bit.
Sender Receiver
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=0-------->|
|-------FCN=1-------->|MIC checked =>
Figure 24: Transmission of an IPv6 packet carried by 11 fragments in the No ACK option
Figure 25 illustrates the transmission of an IPv6 packet that needs 11 fragments in ACK-on-error, for N=3, without losses.
Sender Receiver
|-----W=0, FCN=6----->|
|-----W=0, FCN=5----->|
|-----W=0, FCN=4----->|
|-----W=0, FCN=3----->|
|-----W=0, FCN=2----->|
|-----W=0, FCN=1----->|
|-----W=0, FCN=0----->|
(no ACK)
|-----W=1, FCN=6----->|
|-----W=1, FCN=5----->|
|-----W=1, FCN=4----->|
|-----W=1, FCN=7----->|MIC checked =>
|<---- ACK, W=1 ------|
Figure 25: Transmission of an IPv6 packet carried by 11 fragments in ACK-on-error, for N=3 and MAX_WIND_FCN=6, without losses.
Figure 26 illustrates the transmission of an IPv6 packet that needs 11 fragments ACK-on-error, for N=3, with three losses.
Sender Receiver
|-----W=0, FCN=6----->|
|-----W=0, FCN=5----->|
|-----W=0, FCN=4--X-->|
|-----W=0, FCN=3----->|
|-----W=0, FCN=2--X-->|
|-----W=0, FCN=1----->|
|-----W=0, FCN=0----->|
|<-----ACK, W=0-------|Bitmap:11010111
|-----W=0, FCN=4----->|
|-----W=0, FCN=2----->|
(no ACK)
|-----W=1, FCN=6----->|
|-----W=1, FCN=5----->|
|-----W=1, FCN=4--X-->|
|-----W=1, FCN=7----->|MIC checked
|<-----ACK, W=1-------|Bitmap:11000001
|-----W=1, FCN=4----->|MIC checked =>
|<---- ACK, W=1 ------|
Figure 26: Transmission of an IPv6 packet carried by 11 fragments in ACK-on-error, for N=3 and MAX_WIND_FCN=6, three losses.
Figure 27 illustrates the transmission of an IPv6 packet that needs 11 fragments in ACK-Always, for N=3 and MAX_WIND_FCN=6, without losses. Note: in Window mode, an additional bit will be needed to number windows.
Sender Receiver
|-----W=0, FCN=6----->|
|-----W=0, FCN=5----->|
|-----W=0, FCN=4----->|
|-----W=0, FCN=3----->|
|-----W=0, FCN=2----->|
|-----W=0, FCN=1----->|
|-----W=0, FCN=0----->|
|<-----ACK, W=0-------|no Bitmap
|-----W=1, FCN=6----->|
|-----W=1, FCN=5----->|
|-----W=1, FCN=4----->|
|-----W=1, FCN=7----->|MIC checked =>
|<-----ACK, W=1-------|no Bitmap
(End)
Figure 27: Transmission of an IPv6 packet carried by 11 fragments in ACK-Always, for N=3 and MAX_WIND_FCN=6, no losses.
Figure 28 illustrates the transmission of an IPv6 packet that needs 11 fragments in ACK-Always, for N=3 and MAX_WIND_FCN=6, with three losses.
Sender Receiver
|-----W=1, FCN=6----->|
|-----W=1, FCN=5----->|
|-----W=1, FCN=4--X-->|
|-----W=1, FCN=3----->|
|-----W=1, FCN=2--X-->|
|-----W=1, FCN=1----->|
|-----W=1, FCN=0----->|
|<-----ACK, W=1-------|Bitmap:11010111
|-----W=1, FCN=4----->|
|-----W=1, FCN=2----->|
|<-----ACK, W=1-------|no Bitmap
|-----W=0, FCN=6----->|
|-----W=0, FCN=5----->|
|-----W=0, FCN=4--X-->|
|-----W=0, FCN=7----->|MIC checked
|<-----ACK, W=0-------|Bitmap:11000001
|-----W=0, FCN=4----->|MIC checked =>
|<-----ACK, W=0-------|no Bitmap
(End)
Figure 28: Transmission of an IPv6 packet carried by 11 fragments in ACK-Always, for N=3, and MAX_WIND_FCN=6, with three losses.
Figure 29 illustrates the transmission of an IPv6 packet that needs 6 fragments in ACK-Always, for N=3 and MAX_WIND_FCN=6, with three losses, and only one retry is needed for each lost fragment. Note that, since a single window is needed for transmission of the IPv6 packet in this case, the example illustrates behavior when losses happen in the last window.
Sender Receiver
|-----W=0, CFN=6----->|
|-----W=0, CFN=5----->|
|-----W=0, CFN=4--X-->|
|-----W=0, CFN=3--X-->|
|-----W=0, CFN=2--X-->|
|-----W=0, CFN=7----->|MIC checked
|<-----ACK, W=0-------|Bitmap:11000001
|-----W=0, CFN=4----->|MIC checked: failed
|-----W=0, CFN=3----->|MIC checked: failed
|-----W=0, CFN=2----->|MIC checked: success
|<-----ACK, W=0-------|no Bitmap
(End)
Figure 29: Transmission of an IPv6 packet carried by 11 fragments in ACK-Always, for N=3, and MAX_WIND_FCN=6, with three losses, and only one retry is needed for each lost fragment.
Figure 30 illustrates the transmission of an IPv6 packet that needs 6 fragments in ACK-Always, for N=3 and MAX_WIND_FCN=6, with three losses, and the second ACK is lost. Note that, since a single window is needed for transmission of the IPv6 packet in this case, the example illustrates behavior when losses happen in the last window.
Sender Receiver
|-----W=0, CFN=6----->|
|-----W=0, CFN=5----->|
|-----W=0, CFN=4--X-->|
|-----W=0, CFN=3--X-->|
|-----W=0, CFN=2--X-->|
|-----W=0, CFN=7----->|MIC checked
|<-----ACK, W=0-------|Bitmap:11000001
|-----W=0, CFN=4----->|MIC checked: wrong
|-----W=0, CFN=3----->|MIC checked: wrong
|-----W=0, CFN=2----->|MIC checked: right
| X---ACK, W=0-------|no Bitmap
timeout | |
|-----W=0, CFN=7----->|
|<-----ACK, W=0-------|no Bitmap
(End)
Figure 30: Transmission of an IPv6 packet carried by 11 fragments in ACK-Always, for N=3, and MAX_WIND_FCN=6, with three losses, and the second ACK is lost.
Figure 31 illustrates the transmission of an IPv6 packet that needs 6 fragments in ACK-Always, for N=3 and MAX_WIND_FCN=6, with three losses, and one retransmitted fragment is lost. Note that, since a single window is needed for transmission of the IPv6 packet in this case, the example illustrates behavior when losses happen in the last window.
Sender Receiver
|-----W=0, CFN=6----->|
|-----W=0, CFN=5----->|
|-----W=0, CFN=4--X-->|
|-----W=0, CFN=3--X-->|
|-----W=0, CFN=2--X-->|
|-----W=0, CFN=7----->|MIC checked
|<-----ACK, W=0-------|Bitmap:11000001
|-----W=0, CFN=4----->|MIC checked: wrong
|-----W=0, CFN=3----->|MIC checked: wrong
|-----W=0, CFN=2--X-->|
timeout| |
|-----W=0, CFN=7----->|All-0 empty
|<-----ACK, W=0-------|Bitmap:11110001
|-----W=0, CFN=2----->|MIC checked: right
|<-----ACK, W=0-------|no Bitmap
(End)
Figure 31: Transmission of an IPv6 packet carried by 11 fragments in ACK-Always, for N=3, and MAX_WIND_FCN=6, with three losses, and one retransmitted fragment is lost.
Appendix C illustrates the transmission of an IPv6 packet that needs 28 fragments in ACK-Always, for N=5 and MAX_WIND_FCN=23, with two losses. Note that MAX_WIND_FCN=23 may be useful when the maximum possible bitmap size, considering the maximum lower layer technology payload size and the value of R, is 3 bytes. Note also that the FCN of the last fragment of the packet is the one with FCN=31 (i.e. FCN=2^N-1 for N=5, or equivalently, all FCN bits set to 1).
Sender Receiver
|-----W=0, CFN=23----->|
|-----W=0, CFN=22----->|
|-----W=0, CFN=21--X-->|
|-----W=0, CFN=20----->|
|-----W=0, CFN=19----->|
|-----W=0, CFN=18----->|
|-----W=0, CFN=17----->|
|-----W=0, CFN=16----->|
|-----W=0, CFN=15----->|
|-----W=0, CFN=14----->|
|-----W=0, CFN=13----->|
|-----W=0, CFN=12----->|
|-----W=0, CFN=11----->|
|-----W=0, CFN=10--X-->|
|-----W=0, CFN=9 ----->|
|-----W=0, CFN=8 ----->|
|-----W=0, CFN=7 ----->|
|-----W=0, CFN=6 ----->|
|-----W=0, CFN=5 ----->|
|-----W=0, CFN=4 ----->|
|-----W=0, CFN=3 ----->|
|-----W=0, CFN=2 ----->|
|-----W=0, CFN=1 ----->|
|-----W=0, CFN=0 ----->|
| |lcl-Bitmap:110111111111101111111111
|<------ACK, W=0-------| Bitmap:1101111111111011
|-----W=0, CFN=21----->|
|-----W=0, CFN=10----->|
|<------ACK, W=0-------|no Bitmap
|-----W=1, CFN=23----->|
|-----W=1, CFN=22----->|
|-----W=1, CFN=21----->|
|-----W=1, CFN=31----->|MIC checked =>
|<------ACK, W=1-------|no Bitmap
(End)
The fragmentation state machines of the sender and the receiver in the different reliability options are next in the following figures:
+-----------+
+------------+ Init |
| FCN=0 +-----------+
| No Window
| No Bitmap
| +-------+
| +--------+--+ | More Fragments
| | | <--+ ~~~~~~~~~~~~~~~~~~~~
+--------> | Send | send Fragment (FCN=0)
+---+-------+
| last fragment
| ~~~~~~~~~~~~
| FCN = 1
v send fragment+MIC
+------------+
| END |
+------------+
Figure 32: Sender State Machine for the No ACK Mode
+------+ Not All-1
+----------+-+ | ~~~~~~~~~~~~~~~~~~~
| + <--+ set Inactivity Timer
| RCV Frag +-------+
+-+---+------+ |All-1 &
All-1 & | | |MIC correct
MIC wrong | |Inactivity |
| |Timer Exp. |
v | |
+----------++ | v
| Error |<-+ +--------+--+
+-----------+ | END |
+-----------+
Figure 33: Receiver State Machine for the No ACK Mode
+-------+
| INIT | FCN!=0 & more frags
| | ~~~~~~~~~~~~~~~~~~~~~~
+------++ +--+ send Window + frag(FCN)
W=0 | | | FCN-
Clear local bitmap | | v set local bitmap
FCN=max value | ++--+--------+
+> | |
+---------------------> | SEND |
| +--+-----+---+
| FCN==0 & more frags | | last frag
| ~~~~~~~~~~~~~~~~~~~~~ | | ~~~~~~~~~~~~~~~
| set local-bitmap | | set local-bitmap
| send wnd + frag(all-0) | | send wnd+frag(all-1)+MIC
| set Retrans_Timer | | set Retrans_Timer
| | |
|Recv_wnd == wnd & | |
|Lcl_bitmap==recv_bitmap& | | +------------------------+
|more frag | | |local-bitmap!=rcv-bitmap|
|~~~~~~~~~~~~~~~~~~~~~~ | | | ~~~~~~~~~ |
|Stop Retrans_Timer | | | Attemp++ v
|clear local_bitmap v v | +------++
|window=next_window +----+-----+--+--+ |Resend |
+---------------------+ | |Missing|
+----+ Wait | |Frag |
not expected wnd | | bitmap | +------++
~~~~~~~~~~~~~~~~ +--->+ +-+Retrans_Timer Exp |
discard frag +--+-+---+-+---+-+ |~~~~~~~~~~~~~~~~~ |
| | | ^ ^ |reSend(empty)All-* |
| | | | | |Set Retrans_Timer |
MIC_bit==1 & | | | | +---+Attemp++ |
Recv_window==window & | | | +---------------------------+
Lcl_bitmap==recv_bitmap &| | | all missing frag sent
no more frag| | | ~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~| | | Set Retrans_Timer
Stop Retrans_Timer| | |
+-------------+ | | |
| END +<--------+ | | Attemp > MAX_ACK_REQUESTS
+-------------+ | | ~~~~~~~~~~~~~~~~~~
All-1 Window | v Send Abort
~~~~~~~~~~~~ | +-+-----------+
MIC_bit ==0 & +>| ERROR |
Lcl_bitmap==recv_bitmap +-------------+
Figure 34: Sender State Machine for the ACK Always Mode
Not All- & w=expected +---+ +---+w = Not expected
~~~~~~~~~~~~~~~~~~~~~ | | | |~~~~~~~~~~~~~~~~
Set local_bitmap(FCN) | v v |discard
++---+---+---+-+
+---------------------+ Rcv +--->* ABORT
| +------------------+ Window |
| | +-----+--+-----+
| | All-0 & w=expect | ^ w =next & not-All
| | ~~~~~~~~~~~~~~~~~~ | |~~~~~~~~~~~~~~~~~~~~~
| | set lcl_bitmap(FCN)| |expected = next window
| | send local_bitmap | |Clear local_bitmap
| | | |
| | w=expct & not-All | |
| | ~~~~~~~~~~~~~~~~~~ | |
| | set lcl_bitmap(FCN)+-+ | | +--+ w=next & All-0
| | if lcl_bitmap full | | | | | | ~~~~~~~~~~~~~~~
| | send lcl_bitmap v | v | | | expct = nxt wnd
| | +-+-+-+--+-++ | Clear lcl_bitmap
| | w=expected & +->+ Wait +<+ set lcl_bitmap(FCN)
| | All-1 | | Next | send lcl_bitmap
| | ~~~~~~~~~~~~ +--+ Window +--->* ABORT
| | discard +--------+-++
| | All-1 & w=next| | All-1 & w=nxt
| | & MIC wrong| | & MIC right
| | ~~~~~~~~~~~~~~~~~| | ~~~~~~~~~~~~~~~~~~
| | set local_bitmap(FCN)| |set lcl_bitmap(FCN)
| | send local_bitmap| |send local_bitmap
| | | +----------------------+
| |All-1 & w=expct | |
| |& MIC wrong v +---+ w=expctd & |
| |~~~~~~~~~~~~~~~~~~~~ +----+---+-+ | MIC wrong |
| |set local_bitmap(FCN) | +<+ ~~~~~~~~~~~~~~ |
| |send local_bitmap | Wait End | set lcl_btmp(FCN)|
| +--------------------->+ +--->* ABORT |
| +---+----+-+ |
| w=expected & MIC right| |
| ~~~~~~~~~~~~~~~~~~~~~~| +-+ Not All-1 |
| set local_bitmap(FCN)| | | ~~~~~~~~~ |
| send local_bitmap| | | discard |
| | | | |
|All-1 & w=expctd & MIC right | | | +-+ All-1 |
|~~~~~~~~~~~~~~~~~~~~~~~~~~~~ v | v | v ~~~~~~~~~ |
|set local_bitmap(FCN) +-+-+-+-+-++Send lcl_btmp |
|send local_bitmap | | |
+-------------------------->+ END +<---------------+
++--+------+
--->* ABORT
~~~~~~~
Inactivity_Timer = expires
When DWN_Link
IF Inactivity_Timer expires
Send DWL Request
Attemp++
Figure 35: Receiver State Machine for the ACK Always Mode
+-------+
| |
| INIT |
| | FCN!=0 & more frags
+------++ +--+ ~~~~~~~~~~~~~~~~~~~~~~
W=0 | | | send Window + frag(FCN)
~~~~~~~~~~~~~~~~~~ | | | FCN-
Clear local bitmap | | v set local bitmap
FCN=max value | ++-------------+
+> | |
| SEND |
+-------------------------> | |
| ++-----+-------+
| FCN==0 & more frags| |last frag
| ~~~~~~~~~~~~~~~~~~~~~~~| |~~~~~~~~~~~~~~~~~~~~~~~~
| set local-bitmap| |set local-bitmap
| send wnd + frag(all-0)| |send wnd+frag(all-1)+MIC
| set Retrans_Timer| |set Retrans_Timer
| | |
|Retrans_Timer expires & | | local-bitmap!=rcv-bitmap
|more fragments | | +-----------------+
|~~~~~~~~~~~~~~~~~~~~ | | | ~~~~~~~~~~~~~ |
|stop Retrans_Timer | | | Attemp++ |
|clear local.bitmap v v | v
|window = next window +-----+-----+--+--+ +----+----+
+----------------------+ + | Resend |
+--------------------->+ Wait bitmap | | Missing |
| +-- + | | Frag |
| not expected wnd | ++-+---+---+---+--+ +------+--+
| ~~~~~~~~~~~~~~~~ | ^ | | | ^ |
| discard frag +----+ | | | +-------------------+
| | | | all missing frag sent
|Retrans_Timer expires & | | | ~~~~~~~~~~~~~~~~~~~~~
| No more Frag | | | Set Retrans_Timer
| ~~~~~~~~~~~~~~~~~~~~~~~ | | |
| Stop Retrans_Timer | | |
| Send ALL-1-empty | | |
+-------------------------+ | |
| |
Local_bitmap==Recv_bitmap| |
~~~~~~~~~~~~~~~~~~~~~~~~~| |Attemp > MAX_ACK_REQUESTS
+---------+Stop Retrans_Timer | |~~~~~~~~~~~~~~~~~~~~~~~
| END +<------------------+ v Send Abort
+---------+ +-+---------+
| ERROR |
+-----------+
Figure 36: Sender State Machine for the ACK on error Mode
Not All- & w=expected +---+ +---+w = Not expected
~~~~~~~~~~~~~~~~~~~~~ | | | |~~~~~~~~~~~~~~~~
Set local_bitmap(FCN) | v v |discard
++---+---+---+-+
+-----------------------+ +--+ All-0 & full
| ABORT *<---+ Rcv Window | | ~~~~~~~~~~~~
| +--------------------+ +<-+ w =next
| | +---+---+------+ clear lcl_bitmap
| | | ^
| | All-0 & w=expect| |w=expct & not-All & full
| | & no_full bitmap| |~~~~~~~~~~~~~~~~~~~~~~~~
| | ~~~~~~~~~~~~~~~~~| |clear lcl_bitmap; w =nxt
| | send local_bitmap| |
| | | | +--------+
| | | | +---------->+ |
| | | | |w=next | Error/ |
| | | | |~~~~~~~~ | Abort |
| | | | |Send abort ++-------+
| | v | | ^ w=expct
| | +-+---+--+------+ | & all-1
| | ABORT *<---+ Wait +------+ ~~~~~~~
| | | Next Window | Send abort
| | +-------+---+---+
| | All-1 & w=next & MIC wrong | |
| | ~~~~~~~~~~~~~~~~~~~~~~~~~~ | +----------------+
| | set local_bitmap(FCN) | All-1 & w=next|
| | send local_bitmap | & MIC right|
| | | ~~~~~~~~~~~~~~~~~~|
| | | set lcl_bitmap(FCN)|
| |All-1 & w=expect & MIC wrong | |
| |~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
| |set local_bitmap(FCN) v +--->* ABORT |
| |send local_bitmap +-------+---+--+ |
| +--------------------->+ Wait End +-+ |
| +-----+------+-+ | w=expct & |
| w=expected & MIC right | ^ | MIC wrong |
| ~~~~~~~~~~~~~~~~~~~~~~ | +---+ ~~~~~~~~~ |
| set local_bitmap(FCN) | set lcl_bitmap(FCN)|
| | |
|All-1 & w=expected & MIC right | |
|~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ v |
|set local_bitmap(FCN) +-+----------+ |
+---------------------------->+ END +<----------+
+------------+
--->* Only Uplink
ABORT
~~~~~~~~
Inactivity_Timer = expires
Figure 37: Receiver State Machine for the ACK on error Mode
A set of Rule IDs are allocated to support different aspects of fragmentation functionality as per this document. The allocation of IDs is to be defined in other documents. The set MAY include:
Carles Gomez has been funded in part by the Spanish Government (Ministerio de Educacion, Cultura y Deporte) through the Jose Castillejo grant CAS15/00336, and by the ERDF and the Spanish Government through project TEC2016-79988-P. Part of his contribution to this work has been carried out during his stay as a visiting scholar at the Computer Laboratory of the University of Cambridge.