Adaptive forward error correction in low-power wide area networks (LPWANS)
In one embodiment, techniques for adaptive forward error correction (FEC) in Low-Power Wide Area Networks (LPWANS) are disclosed. The techniques may include determining, by a process, for a block of messages transmitted through a computer network with forward error correction, whether any unrecovered data loss occurred during transmission; increasing, by the process, a level of forward error correction used to transmit through the computer network in response to unrecovered data loss; and/or decreasing, by the process, the level of forward error correction used to transmit through the computer network in response to no unrecovered data loss.
1 . A method, comprising:
determining, by a process, for a block of messages transmitted through a computer network with forward error correction, whether any unrecovered data loss occurred during transmission based on receipt, from a receiver of the block of messages, of an explicit communication indicating one or more messages of the block could not be recovered after forward error correction decoding;
increasing, by the process, a level of forward error correction used to transmit through the computer network in response to unrecovered data loss; and
decreasing, by the process, the level of forward error correction used to transmit through the computer network in response to no unrecovered data loss.
2 . The method as in claim 1 , wherein the level of the forward error correction is increased or decreased linearly.
3 . The method as in claim 1 , wherein the level of the forward error correction is increased or decreased exponentially.
4 . The method as in claim 1 , wherein the level of the forward error correction is increased or decreased based on a number of losses in the block of messages.
5 . The method as in claim 1 , wherein an upper layer process or application detects data losses.
6 . The method as in claim 1 , further comprising:
using a machine learning model to predict when to increase or decrease the level of forward error correction.
7 . The method as in claim 1 , further comprising:
using a machine learning model to optimize a size of a block used for forward error correction.
8 . The method as in claim 1 , wherein the block of messages includes messages configured as layer 2 frames or layer 3 packets.
9 . The method as in claim 1 , wherein the block of messages includes messages configured as full packets or SCHC fragments.
10 . The method as in claim 1 , wherein determining whether any unrecovered data loss occurred during transmission is based on an ack-on-error acknowledgement signaling an unrecovered data loss has occurred.
11 . The method as in claim 1 , wherein the process is executed on a receiving device or observability controller or agent.
12 . The method as in claim 1 , wherein the level of the forward error correction is restricted from increasing past a maximum forward error correction level or decreasing past a minimum forward error correction level.
13 . A tangible, non-transitory, computer-readable medium having computer-executable instructions stored thereon that, when executed by a processor on a computer, cause the computer to perform a method comprising:
determining for a block of messages transmitted through a computer network with forward error correction, whether any unrecovered data loss occurred during transmission based on receipt, from a receiver of the block of messages, of an explicit communication indicating one or more messages of the block could not be recovered after forward error correction decoding;
increasing a level of forward error correction used to transmit through the computer network in response to unrecovered data loss; and
decreasing the level of forward error correction used to transmit through the computer network in response to no unrecovered data loss.
14 . The tangible, non-transitory, computer-readable medium as in claim 13 , wherein the level of the forward error correction is increased or decreased linearly.
15 . The tangible, non-transitory, computer-readable medium as in claim 13 , wherein the level of the forward error correction is increased or decreased exponentially.
16 . The tangible, non-transitory, computer-readable medium as in claim 13 , wherein the level of the forward error correction is increased or decreased based on a number of losses in the block of messages.
17 . The tangible, non-transitory, computer-readable medium as in claim 13 , wherein an upper layer process or application detects data losses.
18 . The tangible, non-transitory, computer-readable medium as in claim 13 , the method further comprising:
using a machine learning model to predict when to increase or decrease the level of forward error correction.
19 . The tangible, non-transitory, computer-readable medium as in claim 13 , the method further comprising:
using a machine learning model to optimize a size of a block used for forward error correction.
20 . An apparatus, comprising:
one or more network interfaces to communicate with a network;
a processor coupled to the one or more network interfaces and configured to execute one or more processes; and
a memory configured to store a process that is executable by the processor, the process, when executed, configured to:
determine for a block of messages transmitted through a computer network with forward error correction, whether any unrecovered data loss occurred during transmission based on receipt, from a receiver of the block of messages, of an explicit communication indicating one or more messages of the block could not be recovered after forward error correction decoding;
increase a level of forward error correction used to transmit through the computer network in response to unrecovered data loss; and
decrease the level of forward error correction used to transmit through the computer network in response to no unrecovered data loss.