Predictive context-based decoder correction
Methods, systems, and machine-readable mediums utilizing context information to create decoding feedback information to improve decoder accuracy and/or performance. In some examples, the context information is from layers of a network stack above the layers in which the decoders are present. The context information may be or be based upon information about previously received and decoded data and/or information about the sender to provide decoding feedback information to the decoder that is used either to correct a previous decoding error or to inform the decoder on which of a plurality of decoding choices is more likely to be correct. This may increase decoding performance by decreasing errors and in some examples, reducing the complexity of choices by eliminating certain decoding possibilities and thus increasing decoder efficiency.
1 . A method of improving decoding performance using predictive feedback information, the method comprising:
receiving, encoded communication data encoded with an error correction code, the encoded communication data being communications from a communicating user;
identifying predictive feedback information of the communicating user based upon previous communications of the communicating user;
creating, using the predictive feedback information, decoding feedback information, the decoding feedback information comprising encoded data that is predictive of communications sent by the communicating user;
decoding, by a decoder and using the decoding feedback information, the encoded communication data to produce decoded communication data, the decoding using the decoding feedback information to select one or more of a plurality of probable decodings; and
outputting the decoded communication data.
2 . The method of claim 1 , wherein the predictive feedback information comprises a word frequency database describing word frequency usage from past communications of the communicating user, and wherein creating the decoding feedback information comprises:
identifying a first word and a second word in the word frequency database, the first word having a higher frequency of usage than the second word;
encoding the first word and the second word to create a first encoded word and a second encoded word; and
creating the decoding feedback information using the first encoded word and the second encoded word and information indicating a frequency of usage of the first encoded word and the second encoded word.
3 . The method of claim 2 , wherein identifying the first word and the second word comprises:
identifying that the first and second words are within a threshold similarity to each other and: a difference in the frequency of usage of the first word and the second word exceeds a first threshold or the frequency of usage of one of the first word or the second word is below a second threshold and the frequency of usage of the other of the first word or the second word is above a third threshold.
4 . The method of claim 3 , wherein identifying the first word comprises identifying the first word in a portion of decoded communication data.
5 . The method of claim 3 , wherein the decoded communication data comprises one of: decoded chat data; decoded instant message data; decoded file data; or decoded voice communications.
6 . The method of claim 1 , wherein the decoded communication data is a document and wherein a sender of the document is a different person than the communicating user.
7 . The method of claim 1 , wherein the encoded communication data is one of:
received over a communication medium or retrieved from a storage device.
8 . The method of claim 1 , wherein decoding, by the decoder and using the predictive feedback information, the encoded communication data to produce the decoded communication data, the decoding using the decoding feedback information to select one or more of the plurality of probable decodings comprises:
adjusting a probability of a decoding result in a turbo decoder based upon the decoding feedback information.
9 . The method of claim 1 , wherein decoding, by the decoder and using the predictive feedback information, the encoded communication data to produce the decoded communication data, the decoding using the decoding feedback information to select one or more of the plurality of probable decodings comprises:
adjusting a path cost of a trellis-based decoder based upon the decoding feedback information.
10 . A computing device for improving decoding performance using predictive feedback information, the computing device comprising:
a hardware processor;
a memory, storing instructions, which when executed by the hardware processor, causes the hardware processor to perform operations comprising:
receiving, encoded communication data encoded with an error correction code, the encoded communication data being communications from a communicating user;
identifying predictive feedback information of the communicating user based upon previous communications of the communicating user;
creating, using the predictive feedback information, decoding feedback information, the decoding feedback information comprising encoded data that is predictive of communications sent by the communicating user;
decoding, by a decoder and using the decoding feedback information, the encoded communication data to produce decoded communication data, the decoding using the decoding feedback information to select one or more of a plurality of probable decodings; and
outputting the decoded communication data.
11 . The computing device of claim 10 , wherein the predictive feedback information comprises a word frequency database describing word frequency usage from past communications of the communicating user, and wherein the operations of creating the decoding feedback information comprises:
identifying a first word and a second word in the word frequency database, the first word having a higher frequency of usage than the second word;
encoding the first word and the second word to create a first encoded word and a second encoded word; and
creating the decoding feedback information using the first encoded word and the second encoded word and information indicating a frequency of usage of the first encoded word and the second encoded word.
12 . The computing device of claim 11 , wherein the operations of identifying the first word and the second word comprises:
identifying that the first and second words are within a threshold similarity to each other and: a difference in the frequency of usage of the first word and the second word exceeds a first threshold or the frequency of usage of one of the first word or the second word is below a second threshold and the frequency of usage of the other of the first word or the second word is above a third threshold.
13 . The computing device of claim 12 , wherein the operations of identifying the first word comprises identifying the first word in a portion of decoded communication data.
14 . The computing device of claim 12 , wherein the decoded communication data comprises one of: decoded chat data; decoded instant message data; decoded file data; or decoded voice communications.
15 . The computing device of claim 10 , wherein the decoded communication data is a document and wherein a sender of the document is a different person than the communicating user.
16 . The computing device of claim 10 , wherein the encoded communication data is one of: received over a communication medium or retrieved from a storage device.
17 . The computing device of claim 10 , wherein the operations of decoding, by the decoder and using the predictive feedback information, the encoded communication data to produce the decoded communication data, the decoding using the decoding feedback information to select one or more of the plurality of probable decodings comprises:
adjusting a probability of a decoding result in a turbo decoder based upon the decoding feedback information.
18 . The computing device of claim 10 , wherein the operations of decoding, by the decoder and using the predictive feedback information, the encoded communication data to produce the decoded communication data, the decoding using the decoding feedback information to select one or more of the plurality of probable decodings comprises:
adjusting a path cost of a trellis-based decoder based upon the decoding feedback information.
19 . A non-transitory machine-readable medium, storing instructions for improving decoding performance using predictive feedback information, the instructions, when executed by a machine, cause the machine to perform operations comprising:
receiving, encoded communication data encoded with an error correction code, the encoded communication data being communications from a communicating user;
identifying predictive feedback information of the communicating user based upon previous communications of the communicating user;
creating, using the predictive feedback information, decoding feedback information, the decoding feedback information comprising encoded data that is predictive of communications sent by the communicating user;
decoding, by a decoder and using the decoding feedback information, the encoded communication data to produce decoded communication data, the decoding using the decoding feedback information to select one or more of a plurality of probable decodings; and
outputting the decoded communication data.
20 . The non-transitory machine-readable medium of claim 19 , wherein the predictive feedback information comprises a word frequency database describing word frequency usage from past communications of the communicating user, and wherein the operations of creating the decoding feedback information comprises:
identifying a first word and a second word in the word frequency database, the first word having a higher frequency of usage than the second word;
encoding the first word and the second word to create a first encoded word and a second encoded word; and
creating the decoding feedback information using the first encoded word and the second encoded word and information indicating a frequency of usage of the first encoded word and the second encoded word.