MEMS motion sensor is low-cost feasible device measuring movement information but sensor noise and drift influence sensor data and movement analysis. The series of related information processing and fusion methods are designed, practised and given with valuable experiment data and analysis using developed MEMS sensor interactive tools. Additionally, portable smart phone sensing application is deep explored towards human daily activity problem covering segmentation and recognition. The smart phone is turned into a cognitive phone by pushing intelligence achieved techniques to the phone. Furthermore, 3D gesture interaction system including three kinds of software platform realization to reduce cognition load caused by communication is illustrated. The developed virtual world navigation interactive tool integrated with game-based learning and spatial thinking training exemplifies constructivist learning application and natural interaction.