Skip to main content

Table 1 Features derived from smartphone sensor signals. Acceleration due to gravity = (Xgrav, Ygrav, Zgrav), linear acceleration = (Xlin, Ylin, Zlin), SD = standard deviation

From: Evaluation of a smartphone human activity recognition application with able-bodied and stroke participants

Signal Feature Formula Abbreviation
Simple moving average of sum of range of linear acceleration (4 windows) \( \frac{{\varSigma^4}_{\mathrm{i}=1}\left[\left( range\left(Xli{n}_i\right)\right) \pm \left( range\left(Yli{n}_i\right)\right) \pm \left( range\left(Zli{n}_i\right)\right)\right]}{4} \) L-SMA
Difference to Y YgravZgravXgrav DifftoY
Sum of range of linear acceleration (range(Xlin i )) + (range(Ylin i )) + (range(Zlin i )) SOR
Sum of standard deviation of linear acceleration (SD(Xlin i )) + (SD(Ylin i )) + (SD(Zlin i )) SoSD
Maximum slope of simple moving average of sum of variances of gravity \( \frac{SM{A}_{var} = {\varSigma^4}_{\mathrm{i}=1}\left(Var{(Xgrav)}_i \pm Var{(Ygrav)}_i \pm Var{(Zgrav)}_i\right)}{4} \)
max(SMA var (2)–SMA var (1), SMA var (3)–SMA var (2), SMA var (4)–SMA var (3))
G-SMAvar