“…For the negative stage, all of the 𝑛 SMs in upper stack are inserted, while only 𝑚 SMs (0 < 𝑚 < 𝑛) are inserted in lower stack with 𝑛 − 𝑚 SMs bypassed. For the positive stage, 𝑚 SMs in upper stack and 𝑛 SMs in lower stack are inserted [1], [16], [17]. This modulation highly satisfies the demand of voltage amplitude conversion in MMDCs, since the choice of 𝑚 can be utilized as an effective control degree of freedom for voltage regulation [16], namely, a larger 𝑚 provides a higher step-down ratio and vice versa [1], [18].…”