In this paper, a driving signal-based diagnostic approach for PWM inverter IGBT switch open-circuit faults is proposed. This approach has the capacity to swiftly detect and recognize switch open-circuit faults. The proposed approach has been demonstrated to effectively track the abnormal transitions of circuit operating states prior to and following an open switch fault, thereby acquiring diagnostic variables that reflect these abnormal state transitions. Subsequently, the state variables associated with the abnormal transitions are optimized. After analyzing the logical relationship between the feature set variables and the driving signals in details, a hardware diagnostic circuit is specifically designed for processing signals to obtain fault diagnosis results. This circuit offers the advantages of high speed, low consumption of software resources, and elimination of the impacts of signal delay and switching noise on the diagnostic results. Additionally, it can be integrated with overcurrent detection. The experimental results demonstrate the efficacy of the proposed approach in rapidly identifying the faulty switch.