232
|
6 Applications and Methods in Biosignal Processing
0
1
2
3
4
5
6
7
8
9
10
Time t/s
300
400
500
Force F/N
a) Findpeaks Force Data
0
1
2
3
4
5
6
7
8
9
Time t/s
0
100
200
Angle /°
b) Findpeaks Angle Data
0
1
2
3
4
5
6
7
8
9
Time t/s
0
100
200
Angle /°
c) First Derivative Zeros in Angle Data
0
1
2
3
4
5
6
7
8
9
Time t/s
0
100
200
Angle /°
d) Second Derivative Zeros in Angle Data
0
1
2
3
4
5
6
7
8
9
10
Time t/s
300
400
500
Force F/N
e) First Derivative Zeros in Force Data
0
1
2
3
4
5
6
7
8
9
10
Time t/s
300
400
500
Force F/N
f) Second Derivative Zeros in Force Data
Fig. 6.12: Results of the analysis approaches by means of the MATLAB functions findpeak() and
gradient().
added to an array. These force values are further referred as local minima. The start
and end points of the local minima are used to calculate the trigger points for the
synchronisation. According to additional conditions of the characteristic pattern of
the 3 knee bends [11], [73], the number of the local minima in the force signal must be
six (cf. Figure 6.13, a). If the amount of the local minima is above six, which is very
likely starting with a threshold factor of 95%, the threshold factor has to decrease by
1% and the determination of local minima is performed again until the number of six
local minima is reached. The determination of the local minima of the knee joint angle
data follows the same principle. But due to the different characteristic pattern of the
knee joint angle signal the desired number of local minima is three (cf. Figure 6.13, b).
The result of the threshold manipulation and the stored values of the local minima
of the force and the knee joint angle can be seen in Figure 6.13 (c and d), where the
start and end points of the extracted local minima are marked. After determining the