Table 1.
Features used for identifying the inertial sensors
| Feature | |||
|---|---|---|---|
|
Description |
a |
ω |
α |
| RMS of the | |||
| -magnitude |
RMS{ ||a||} |
RMS{ ||ω||} |
RMS{ ||α||} |
| -x-component |
RMS{ ax} |
RMS{ ωx} |
RMS{ αx} |
| -y-component |
RMS{ ay} |
RMS{ ωy} |
RMS{ αy} |
| Variance of the | |||
| -magnitude |
Var{ ||a||} |
Var{ ||ω||} |
Var{ ||α||} |
| -x-component |
Var{ ax} |
Var{ ωx} |
Var{ αx} |
| -y-component |
Var{ ay} |
Var{ ωy} |
Var{ αy} |
| -z-component |
Var{ az} |
Var{ ωz} |
Var{ αz} |
| Sum of cc’s of a sensor with all other sensors of the | |||
| -magnitude |
Σcc{ ||a||} |
Σcc{ ||ω||} |
Σcc{ ||α||} |
| -x-component |
Σcc{ ax} |
Σcc{ ωx} |
Σcc{ αx} |
| -y-component |
Σcc{ ay} |
Σcc{ ωy} |
Σcc{ αy} |
| -z-component |
Σcc{ az} |
Σcc{ ωz} |
Σcc{ αz} |
| The maximum value of the cc’s of a sensor with all other sensors of the | |||
| -magnitude |
Max{cc{ ||a||}} |
Max{cc{ ||ω||}} |
Max{cc{ ||α||}} |
| -x-component |
Max{cc{ ax}} |
Max{cc{ ωx}} |
Max{cc{ αx}} |
| -y-component |
Max{cc{ ay}} |
Max{cc{ ωy}} |
Max{cc{ αy}} |
| -z-component |
Max{cc{ az}} |
Max{cc{ ωz}} |
Max{cc{ αz}} |
| The inter-axis cc’s of a sensor between the | |||
| -x- and y-axes |
cc{ ax,ay} |
cc{ ωx,ωy} |
cc{ αx,αy} |
| -x- and z-axes |
cc{ ax,az} |
cc{ ωx,ωz} |
cc{ αx,αz} |
| -y- and z-axes | cc{ ay,az} | cc{ ωy,ωz} | cc{ αy,αz} |
All 57 (19 ·3) features are given as input to the decision tree learner. The C4.5 algorithm automatically chooses the features that split the data most effectively.