The decline in physical function and the deterioration of the neuromusculoskeletal system in older people can easily lead to reduced muscle strength and slower mobility in the joints of the lower limbs, increasing the incidence of chronic diseases such as muscle wasting disorders, osteoporosis, debilitation and fall and fracture. It may also affect the quality of life and functional independence of older people, and in serious cases, even directly threaten their health. This study was conducted to determine the differences in lower limb muscle activation characteristics between static semi-squat (SSS) and dynamic semi-squat (DSS) training in middle-aged and old women at different frequencies and amplitudes and to explore a personalized whole-body vibration (WBV) training instruction program suitable for them. Fifteen healthy middle-aged and old women (60.8 ± 4.18 years old) were recruited for SSS and DSS WBV training. Their muscle activity of the rectus femoris (RF), vastus medialis (VM), vastus lateralis (VL), biceps femoris (BF) and gastrocnemius (GS) were calculated using the BTS FreeEMG300 wireless surface electromyography (EMG), which participants were completed that the two different contraction patterns of squats on WBV training. The knee flexion was maintained at 45° while the subjects were performing the SSS training, while during the DSS training, the knee flexion fluctuates between 10° and 45°. The SSS exercise requires the subject to remain stationary in the squatting position and the DSS to be performed at a rhythm of 4 s/repetition, with 2 s of squatting, 1 s of standing up, and 1 s intervals. The vibration frequencies and amplitudes were changed to the WBV training intensity, and the vibration frequencies were set to 0 Hz, 30 Hz and 40 Hz, and the amplitudes were set to 0 mm, 2 mm and 4 mm. Each subject is randomised to participate in WBV training with 5 combinations of frequency and amplitude in both static and dynamic semi-squat patterns. These were 0 Hz 0 mm, 30 Hz 2 mm, 30 Hz 4 mm, 40 Hz 2 mm, 40 Hz 4 mm for the static and dynamic demi-squat patterns of WBV training. A two-way repeated measures ANOVA was applied to compare the changes in surface EMG of the lower limb muscles in different modes of SSS and DSS with WBV training. (1) Our results showed significantly interaction effects in the frequency × amplitude of root mean square (EMGrms) in GS (P < 0.05), while no significant differences were observed in the interaction effects of SSS/DSS patterns, frequencies and amplitude of RF, VM, VL and BF (P > 0.05). (2) Comparisons between groups showed that the EMGrms of the RF were significantly higher for the DSS than the SSS (P < 0.05). Additionally, the EMGrms of VL and BF at 30 Hz and 40 Hz were greater than 0 Hz (P < 0.05). Also, The EMGrms at 4 mm for the VM, VL and BF were significantly higher than 0 mm, the EMGrms at 4 mm for the VM and VL were significantly higher than 2 mm (P < 0.05), and the EMGrms at 2 mm of VL and BF were significantly higher than 0 mm (P < 0.05). (3) The results showed that WBV stimulation significantly increased the EMGrms of the GS in the SSS compared with the vibration free semi-squat alone (P < 0.05). However, there were no significant differences between WBV training protocols for SSS patterns with different frequencies and amplitudes (frequencies and amplitudes not were 0 Hz and 0 mm) (P > 0.05). Comparison of EMGrms for WBV training of the GS in DSS patterns showed that 40 Hz/4 mm was significantly higher than 0 Hz/0 mm (P < 0.05), but there was no significant difference between the remaining vibration conditions (P > 0.05). WBV training for DSS can significantly improve the activation of the RF compared to SSS pattern. Compared with no vibration, WBV could significantly improv the activity of the lower limb muscles. Additionally, an increase in amplitude from 2 to 4 mm could significantly improve VM and VL activation, while no significant improvement on lower limb muscle activation was observed for increasing vibration frequency from 30 to 40 Hz.