
215
Power amPlifier linearization
is hampered and the CGMP technique is trying to decrease the error
in the regions where the input power is near the saturation point, as
shown in Figure6.13, the CGMP is going to be nonconvergence.
It can also be seen that the value of the error vector is negative and
this is true because the AM-AM characteristicof thePA is bending,
0.70.60.50.4
Input Voltage (Scaled)
Voltage Error
0.3
0.20.10
–0.06
–0.04
–0.02
0.02
0.04
0.06
0.08
0
Figure 6.13 Error voltage of the power amplifier after five iterations.
0.70.60.50.4
Input Voltage (Scaled)
Voltage Gain
0.3
1
2
0
0.2
–2
–1
0.10
–3
Figure 6.12 Complex gain memory predistortion ...