Bluetooth low energy profile for MPU9150 IMU data transfers

An increasing number of commercial and scientific solutions for unmanned vehicles, aeronautical/aerospace navigation systems, health and sports monitoring solutions, has been prompted by the development of Micro-Electro-Mechanical Systems based inertial sensors. These sensors, with small form factor and low cost, present an advantage for computing three-dimensional space object orientation. Also radio frequency communication (RF) devices are frequently associated with inertial sensors and embedded systems. Radio frequency communication protocols allow immediate data transfer and processing without cumbersome cabling. With special relevance for sports monitoring, gait analysis and general movement assessment inertial sensors have been embedded together to form one device with several sensors. The so called Inertial Measurement Units (IMU), that include an accelerometer, a gyroscope and sometimes may include one magnetometer, are increasingly used, despite their inherent inaccuracy, to determine a body orientation and/or movement profile. These sensors present therefore an advantage for computing three-dimensional space object orientation. The inherent inaccuracy is addressed by using sensor fusion algorithms that increase system reliability. In the work presented in [1], a survey of a number of relevant sensor fusion algorithms was presented. In [2] a Yaskawa Motoman robotic arm is used to implement a set of predefined movements that granted accuracy and repeatability conditions to a framework that is able to test systems that are able to determine a body position. In both [1] and [2], a Bluetooth Low Energy based CC2541 Texas Instruments device has been used to transfer the required data from the IMU sensors to a PC computer running a Windows operating system. This paper presents the BLE proprietary profile that has been implemented to transfer data in wireless mode. Moreover, the paper describes the BLE protocol basics and the relevant options so that implemented systems are low power and fully compatible with the protocol. Also the future use of the developed device is addressed within the context of the Parkinson disease early diagnosis and Canoeing sport monitoring.