This report presents "through-the-wall" (TTW) radar measurements of a person, performing various movements in an enclosed room. Based on data from these measurements a new signal processing method has been developed and tested. The work is part of a joint program between Sweden (FOI) and France (ONERA and DGA Celar), "New models for radar targets and environment". The Swedish participation is financed within the project "Penetrating radar systems for urban operations". The aim of the collaboration is to investigate and develop technical and signal processing methods that enable reliable and robust detection of moving objects behind walls. The TTW measurements were conducted at FOI's test range Lilla Gåra in October 2008. We have found that the recorded motions of the person in the enclosed room can be detected. Even a person who is standing still and holding his breath is relatively easy to detect with the radar signal processing technique used. An MTI (Moving Target Indication) signal processing algorithm, based on coherent subtraction between different frequency sweeps, has been developed and tested. A dynamic threshold level that is related to the returned signal strength has been implemented, which enables detection of a wide spectrum of human movements. By plotting the detected movements in a range-time diagram we get detection tracks that correspond to the movement. The obtained detection tracks can be explained by direct reflections from the person or multiple reflections from the walls and the person. The stationary background is well suppressed by the algorithm. Depending on the type of human movement, the appearance of the detection plots varies. We can recognize some specific motions (e.g., a moving arm) in the offline analysis, while more subtle motions (e.g., moving the head back and forth) cannot be captured. The robust detection algorithm can run in real time and the interpretation of the detection plots can probably be automized. Hence, this indicates that our MTI-based signal processing approach has potential to be implemented in future handheld TTW systems.