Spectrum sensing applications cover a wide variety, such as efficient utilization of the frequency spectrum in wireless communication or measurement of the Radio-Frequency Electromagnetic-Field (RF-EMF) in medical applications.
A popular technique used for Spectrum sensing is Energy Detection. The current implementations of this technique suffer from several drawbacks. Wide-band spectrum sensing circuits consume a large power and are not suitable for portable devices. Furthermore they suffer from I/Q imbalance. The choice of the detection threshold has a large impact on the Spectrum sensing performance, namely the probability of detection and the probability of false alarm.
In this thesis, we propose a power-efficient highly digitized RF receiver suitable for portable devices.
The proposed circuit does not suffer from I/Q imbalance. The performance of this circuit has been validated by measurement results from a chip fabricated in a 65 nm CMOS technology. A general methodology has been proposed to determine the detection threshold. The proposed method has been validated by simulation and it has been used to study the impact of non-linearity on the spectrum sensing performance. A circuit implementation of a digital backend of the proposed system is presented. This implementation comprises an efficient down-conversion mixer, decimation filter, custom FFT block, and energy detection module. The implementation was validated on FPGA using the on-chip logic analyzer.
In this work, we also present the first hardware measurement of the I/Q imbalance on spectrum sensing performance using a conventional Software Defined Radio receiver.
In the medical field, we also present for the first time a study of the effect of RF-EMF exposure on newborns by performing simultaneous acquisitions of RF-EMF signals and neonates physiological parameters.