Energy harvesting techniques with low quiescent current (l_Q), high maximum power-point tracking (MPPT) efficiency become crucial for the Internet of Things (loT) nodes. Meanwhile, the applicability to multiple energy sources is favorable, demanding a wide voltage conversion ratio VCR range. The MPPT is usually achieved by tuning the input impedance of the cascading boost DC-DC converter. The fractional open-circuit voltage (FOCV) [1] is the one of simplest MPPT methods, but the MPPT efficiency is good only when we obtain the precise fraction constant. Hill-climbing (HC, as shown in Fig. 1) [2], [3] is another widely used MPPT scheme that achieves high MPPT efficiency, and is applicable to all energy sources. However, it senses the input voltage and current V_IN and l_lN for the input power P_IN, usually in an analog way and hence demands a high l_Q. Work from [4] proposed an MPPT method based on the Constant-Energy-Packet-Extraction (CEPE), as shown in Fig. 1. Eq. (1) presents the energy obtained in each switching cycle. Using the peak inductor current (l_PEAK) and discontinuous-conduction mode (DCM) control, and assuming (tON > > t(OFF) (when VCR is high), each packet energy ECEPE is almost constant (eq.2). [4] utilizes CEPE to achieve MPPT by finding out the switching frequency (fsw) where duty cycle varies greatly (large ΔD). This may consume a high l_Q, and is only applicable to thermoelectric generators (TEG). Besides, its MPPT efficiency decreases if t_ON is not > > t_OFF (small VCR cases). As given in Fig. 1, the calculated relative error increases when VCR decreases.