Dynamic self-correction of message reliability in LDPC codes

An electronic system, comprising:

• memory;
• a controller to control access to the memory;
• and
• logic communicatively coupled to the controller and the memory, the logic to:
• detect unreliable messages between check nodes and variable nodes in association with an error correction operation;
• determine respective degrees of unreliability for the unreliable messages;
• and
• reduce an influence of the unreliable messages on the error correction operation, as compared to an influence of reliable messages between the check nodes and the variables nodes, based on the determined respective degrees of unreliability;
• wherein:
• the logic to detect the unreliable messages comprises the logic to determine a first reliability confidence value, associated with a first message, based on a comparison of a first hard decision value and a second hard decision value;
• the first hard decision value is based on an identification of a respective maximum value of a first log-density ratio (LDR) vector which is communicated, in a first iteration of multiple iterations of a decode algorithm, via a first edge between a first node of a decoder and a second node of a decoder;
• and
• the second hard decision value is based on an identification of a respective maximum value of a second LDR vector which is communicated, via the first edge, in a second iteration of the multiple iterations, the second iteration after the first iteration.

A method of controlling memory, the method comprising:

• detecting unreliable messages between check nodes and variable nodes in association with an error correction operation;
• determining respective degrees of unreliability for the unreliable messages;
• and
• reducing an influence of the unreliable messages on the error correction operation, as compared to an influence of reliable messages between the check nodes and the variables nodes, based on the determined respective degrees of unreliability;
• wherein:
• detecting the unreliable messages comprises determining a first reliability confidence value, associated with a first message, based on a comparison of a first hard decision value and a second hard decision value;
• the first hard decision value is based on an identification of a respective maximum value of a first log-density ratio (LDR) vector which is communicated, in a first iteration of multiple iterations of a decode algorithm, via a first edge between a first node of a decoder and a second node of a decoder;
• and
• wherein the second hard decision value is based on an identification of a respective maximum value of a second LDR vector which is communicated, via the first edge, in a second iteration of the multiple iterations, the second iteration after the first iteration.