Identification of gait events using an optimal sensor set and a reliable algorithm would be useful in the clinical evaluation of patients with dropped foot. This article describes a threshold detection method for identifying gait events and evaluating the reliability of a system on ten subjects with dropped foot and three non-impaired controls. The system comprised three sensor units of accelerometers and gyroscopes attached at the thigh, shank and foot of the impaired leg in subjects with dropped foot, and the dominant leg in the controls. A performance index was devised to compare the values of different measuring directions of the sensor units and evaluate the system's reliability. The performance index, with the ideal value equal to 1, depended on the classification accuracy and timing variation of the turning points. These were obtained from the threshold detection method that distinguished the absolute maximum and minimum turning points from local maximum and minimum turning points. It was found that some specific turning points could effectively identify gait events with a high median value in the performance index. These turning points included: the minimum turning point in superior-inferior acceleration on the thigh at loading response (0.972); the minimum turning point in anterior-posterior angular velocity on the shank at pre-swing (0.955) and the maximum turning point in superior-inferior acceleration on the foot at initial swing (0.954). Combining the results of sensor measurements in different orientations and attachment locations could be used for gait event identification. It was shown that the threshold detection method is reliable. Portable gait-monitoring devices can be used for monitoring of daily activities and functional control.