HARQ-ACK Signal Detection for ACK/NACK and DTX Separation

The detection process is segmented into two distinct stages: first, a threshold-based algorithm provides rapid initial detection; second, a machine learning classifier refines the detection result by analyzing features that distinguish ACK/NACK from DTX. This segmentation allows the system to benefit from both the speed of threshold-based methods and the precision of machine learning, resolving the contradiction between detection speed and accuracy.

A machine learning classifier acts as an intermediary between the threshold-based detection and the final decision. The classifier takes the threshold detection output and additional signal features as input, processes them through learned patterns, and produces a refined detection result. This intermediary component enables the system to overcome the limitations of pure threshold-based methods while maintaining overall detection speed.

The machine learning classifier serves as an intermediary that enhances reliability without significantly increasing device complexity. By using pre-trained models and efficient feature extraction, the classifier can be integrated into existing detection architectures with minimal additional computational burden, while substantially improving the reliability of distinguishing ACK/NACK from DTX signals.

The system changes the parameter space by introducing multiple features for analysis beyond simple threshold comparisons. These features include signal energy, correlation metrics, and other characteristics that the machine learning classifier processes to make more reliable detection decisions. This parameter expansion enables more accurate differentiation between similar signals without requiring complex hardware changes.