In order to realize a smart home to recognize human activities of daily living (ADL), it is necessary to develop a human activity recognition technology. This study proposes a method of awareness of human activities by detecting state change of room equipment, which includes doors, chairs, desks, a light and a fridge. A characteristic of the proposed method is an indirect recognition method in the sense that it does not employ the human motion data. In this study we deal with an action estimation and a movement locus estimation as the tasks on the awareness of human activities. Two simulation experiments have been done to evaluate the proposed method. As the results, we show that the proposed method can estimate human actions with a precision rate of 66.7% and with a recall rate of 72.7%. Also we show the movement locus of human in a room can be approximately estimated.