Premium
Analysis and design of an integrated RF energy harvester for ultra low‐power environments
Author(s) -
Caselli Michele,
Tonelli Matteo,
Boni Andrea
Publication year - 2019
Publication title -
international journal of circuit theory and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.364
H-Index - 52
eISSN - 1097-007X
pISSN - 0098-9886
DOI - 10.1002/cta.2637
Subject(s) - rectifier (neural networks) , electrical engineering , energy harvesting , radio frequency , power (physics) , cmos , voltage , capacitor , maximum power point tracking , maximum power principle , electronic engineering , engineering , computer science , physics , stochastic neural network , quantum mechanics , machine learning , inverter , recurrent neural network , artificial neural network
Summary Radio‐frequency (RF) energy harvesting must cope with the limited availability and high variability of the energy source. In this paper, the available RF power in three typical environments (urban, semi‐urban, and rural) is investigated. Measurements show that in the surveyed urban and semi‐urban environments, an average input power above −22 and −29 dBm, respectively, is available in the [700, 1,000] MHz band. A mathematical model of the interface between the RF rectifier and the DC‐DC converter is provided. The analysis demonstrates that the energy can be efficiently transferred to the external accumulator coupling the rectifier with a strobed, input control DC‐DC converter. Based on the measurements and the analysis, an RF harvester architecture has been designed in 65 nm Complementary Metal‐Oxide Semiconductor (CMOS) technology to operate over the [−40, 85] o C temperature and the [1.1, 2.5] V battery voltage ranges. The input control strategy adopted for the converter allows the adaptation of the harvester to the available RF power and enables a real maximum power point tracking (MPPT). Post‐layout simulation of the harvester, recharging a large capacitor, precharged at 2 V, at 950 MHz of input frequency returned a 33.4% peak efficiency with an input power of 15 μ W (−18 dBm). The minimum input power leading to a positive energy balance is −30 dBm with an output voltage of 1.1 V.