ftNote - stride monitor

Periodically during your run, ftNote announces stride metrics such as your stride rate and your stride impact index (i.e., average max jerk per stride divided by average speed) which is a measure of the smoothness of your stride.

After your run, you can optionally upload your acquired sensor data to your free Google Drive account for free stride analysis to see even more stride metrics such as how much you are overstriding (i.e., braking on impact). The rotation speed screenshot illustrates several analysis details:

- The analysis is independent of the smartphone's fixed orientation on the shin during acquisition. The two plots are from two smartphones, each on my left shin but occupying different axial ranges (i.e., one is above the other in the sense that one smartphone is close to my knee and the other is close to my ankle) and having different orientations (i.e., the top smartphone's screen is directed out at 2 o'clock and the bottom phone is directed out at 10 o'clock, where 12 o'clock is straight forward).
- The sampling rate is sufficient for stride analysis and is expected to get better. Nexus 4 smartphones acquired the sensor data displayed in the above plots at a sampling rate of 50 Hz. The ftNote stride monitor application implements double buffering in which one buffer gets filled with sensor data while the other buffer gets written to the SD card. The Nexus 5 smartphone acquires sensor data even faster (at about 58 Hz) and the newer smartphones have the ability to acquire sensor data in batch mode to further increase the efficiency of acquiring sensor data.
- During analysis, the sensor data acquired from multiple smartphones can be synchronized to within one sample period. The sensor data displayed in the above plots are from two unsynchronized smartphones but the analysis synchronizes the timing of their acquired sensor data.
- During analysis, the sensor data can be transformed into many different coordinate systems. The sensor data is acquired in the device coordinate system (i.e., x to the right of the screen, y to the top of the screen, and z pointing out of the screen). However, the analysis can transform the sensor data to a shin coordinate system (i.e., x is down the shin, y points forward from the shin, and z points to the right), or to a body coordinate system (i.e., x is forward, y is to the left, and z is up), or to a world coordinate system (i.e., x is East, y is North, and z is up). The above plots are in the shin coordinate system. Oddly enough, some analysis is easier and more accurate when performed in non-inertial reference frames (i.e., the default case for a smartphone strapped to a shin) than when performed in inertial reference frames (i.e., the default case for most sensors in most indoor running labs).

Many of the application's permissions are for the emergency contact feature that automatically calls someone if you become motionless: directly call phone numbers, send SMS messages, control vibration, and change your audio settings. The other permissions (i.e., precise location and modify the contents of your USB storage) are for logging motion sensor data for the stride analysis option.