A Novel Energy-Efficient Approach for Human Activity Recognition

. 2017 Sep 8;17(9):2064. doi: 10.3390/s17092064

Algorithm 2 Context-based classification Algorithm

1: Initialize the data buffer ${R e s u l t_{1}, R e s u l t_{2}, \dots, R e s u l t_{2 k + 1}}$
2: while $R$ is input from H-SVM do
3: {
4: If n < 2k + 2 then
5: $R e s u l t_{n} = R (n = 1, 2, \dots, 2 k + 1)$
6: $R W_{T} = R t$
7: else
8: ${M o d e_{1}, M o d e_{2}, \dots} = M a j o r i t y O f {R e s u l t_{1}, R e s u l t_{2}, \dots, R e s u l t_{2 k + 1}}$
9: if $R e s u l t_{k + 1} \in {{M o d e_{1}}, {M o d e_{2}}, \dots}$ then
10: $R W_{t} = R e s u l t_{k + 1} = R_{t}$
11: else
12: $R W_{t} = M o d e_{1}$
13: }
14: end while