Sleep power minimisation using adaptive duty‐cycling of DC–DC converters in state‐retentive systems

Abstract

Aggressive power management techniques, which combine hardware and software solutions, are fundamental for embedded computing platforms today, especially if they are battery operated. This paper proposes an adaptive low-level algorithm, which modulates the DC-DC converter activation for minimising quiescent current consumption. This algorithm allows a discontinuous usage of the DC-DC converter during the sleep time, without requiring modification in the user's main program, by powering the system solely with the internal DC-DC converter capacitor and without using any other additional capacitors as an energy buffer. The algorithm computes the maximum interval between consecutive wake-ups necessary for the capacitor recharging at run-time. Intervals are decided by taking into account both the global leakage and the temperaturedependent variations of the capacitor. The proposed solution significantly enhances the lifetime of applications with a low activity rate, such as wireless sensor networks, while still guaranteeing efficient power delivery for high-current demand intervals.

Self‐tuning indirect adaptive control of non‐inverting buck–boost converter

“…G * (q) = B m A m = b m0 q q 2 + a m1 q + a m2 (19) Therefore, B − = b 0 , B + = q + b 1 /b 0 and the Diophantine equation could be rewritten as below (q 2 + a 1 q + a 2 ) × 1 + b 0 (s 0 q + s 1 ) = q 2 + a m1 q + a m2 (20) and the solution is…”

“…The aim of identification is to determine the dynamics of a system using a given set of data. In parametric identification, the model structure and the experimental input and output data are used to estimate non-measurable parameters and the goal is to optimise the transfer function that fits the best to the system dynamics [17][18][19].…”