Method for reducing false detection of successful decoding of cyclic redundancy check codes

A device, comprising:

• one or more memories;
• and
• one or more processors configured to:
• receive, from a base station, a polar encoded physical downlink control channel (PDCCH) payload;
• decode, using a successive cancellation list decoding algorithm, the polar encoded PDCCH payload to obtain candidate PDCCH payloads and to generate path metrics (PMs),
• wherein each PM of the PMs corresponds to one candidate PDCCH payload of the candidate PDCCH payloads;
• perform a cyclic redundancy check on each of the candidate PDCCH payloads to determine, from the PMs, a passing PM;
• determine, based on the PMs, a confidence value associated with the passing PM,
• wherein the one or more processors, to determine the confidence value, are configured to determine the confidence value based on processing the passing PM using a machine learning model trained on historical PM values;
• determine whether the confidence value satisfies a threshold;
• discard, based on determining that the confidence value does not satisfy the threshold, the passing PM;
• or
• output, based on determining that the confidence value does satisfy the threshold, a candidate PDCCH payload corresponding to the passing PM;
• and
• transmit, based on the candidate PDCCH payload, data to the base station.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors, cause the one or more processors to:
• receive, from a base station, an encoded physical downlink control channel (PDCCH) payload;
• decode the encoded PDCCH payload to obtain candidate PDCCH payloads and to generate path metrics (PMs),
• wherein each PM of the PMs corresponds to one candidate PDCCH payload of the candidate PDCCH payloads;
• perform a cyclic redundancy check on each of the candidate PDCCH payloads to obtain, from the PMs, a passing PM;
• determine, based on the PMs and using a machine learning model, a confidence value associated with the passing PM,
• wherein the machine learning model is trained based on historical PM values;
• determine whether the confidence value satisfies a threshold;
• discard, based on determining that the confidence value does not satisfy the threshold, the passing PM;
• or
• output, based on determining that the confidence value does satisfy the threshold, a candidate PDCCH payload corresponding to the passing PM;
• and
• transmit, based on the candidate PDCCH payload, data to the base station.