**7. Future concepts for smartphones and portable media devices, sensor fusion**

In the near future the application of sensor fusion will likely be achieved broadly throughout the domain of smartphones and portable media devices as wearable and wireless accelerometer and gyroscope sensors. Sensor fusion involves the tandem operation of multiple inertial sensors, such as an accelerometer and gyroscope. With the combined signal of both accelerometer and gyroscope signal the actual spatial location in terms of displacement, velocity, and acceleration of the sensor mounting position can be determined [57–59]. For example, the spatial representation of foot displacement can be ascertained during gait [57].

gyroscope signal data. Through the application of a multilayer perceptron neural network illustrated in **Figure 11** considerable classification accuracy was achieved for distinguishing between scenarios with Virtual Proprioception for eccentric training and eccentric training

**Figure 11.** Multilayer perceptron neural network applied for classifying between scenarios of eccentric phase for a set of biceps curls respective of Virtual Proprioception for eccentric training and without feedback through Virtual

**Figure 10.** Gyroscope signals for eccentric phase of a biceps curl respective of Virtual Proprioception for eccentric

A wobble board is a therapy device that promotes rehabilitation of the ankle foot complex. A portable media device as shown in **Figure 12** can be readily mounted to the wobble board to provide advanced acuity as a wireless gyroscope platform. The rotation of the wobble board is notably disparate regarding the hemiplegic affected ankle and unaffected ankle upon observation. The observed disparity can be quantified through the wireless gyroscope platform from the portable media device and conveyed as an email attachment through wireless connectivity to the Internet. Post-processing consolidates the gyroscope signal into a feature set. With a multilayer perceptron neural network considerable classification accuracy was achieved for differentiating between the hemiplegic affected ankle and unaffected ankle [53].

without feedback through Virtual Proprioception [54].

training (a) and without feedback (b) [54].

16 Smartphones from an Applied Research Perspective

Proprioception [54].

Traditional sensor fusion however requires a considerable sampling rate that exceeds the feasible sampling threshold for smartphones and portable media devices as wearable and wireless systems [1–3, 57]. Post-processing resources are generally not portable, and techniques, such as the Kalman filter are required to serve as an orientation filter [57–59]. An orientation filter that satisfies the computational bounds associated with smartphones and portable media devices would be desirable for sensor fusion applications.

Madgwick has proposed a gradient descent algorithm as an orientation filter that is feasible for smartphones and portable media devices for the scope of sensor fusion. The IMU version requires only on the order of roughly 100 computational operations per filter update. The filter is also amenable to reduced sampling rates, such as 10 Hz [58, 59]. These capabilities make the gradient descent algorithm a realistic orientation filter with regards to sensor fusion for smartphones and portable media devices.

Sensor fusion also requires a novel application for smartphones and portable media devices that can simultaneously record the accelerometer and gyroscope signal for the same sample. Cognition Engineering has recently developed a smartphone and portable media device application suitable for acquiring both the accelerometer and gyroscope signal in a simultaneous manner. An example of the Cognition Engineering application has been demonstrated with regards to a person with essential tremor conducting a reach and grasp task with a deep brain stimulator in 'On' and 'Off' mode [60].
