An algorithmic framework for reconstruction of time-delayed and incomplete binary signals from an energy-lean structural health monitoring system

•A new structural health monitoring strategy based on time-delayed binary signals is presented.•A machine learning (ML) framework merging matrix completion and pattern recognition is developed.•A data fusion model is introduced to pre-process the incomplete signals for the ML framework.•A statistical approach is employed to detect damage with time-delayed signals.•The approach is evaluated for plates through numerical and experimental tests.