A 0.7-V Sub-mW Type-II Phase-Tracking Bluetooth Low Energy Receiver in 28-nm CMOS

Abstract

We present an architecture of a Bluetooth low energy (BLE)-compliant receiver which, for the first time ever, breaks the 1 mW barrier of power consumption. It is based on a type-II phase-tracking loop and addresses the mutual magnetic coupling between on-chip inductors of a digitally controller oscillator (DCO) and low-noise transconductance amplifier (LNTA), which causes RX performance degradation in the priorart implementations. An inverter-based inductor-free LNTA is employed instead. The resulting adjacent channel rejection (ACR) improves by 1.5/2.5 dB at 2/3 MHz offset. By further leveraging current-reuse and switched-capacitor circuitry, this RX achieves the best-in-class FoM of 183.2 dB with sensitivity of −93.2 dBm. Thanks to the single-channel topology, the proposed RX occupies tiny area of 0.48 mm 2 in 28-nm CMOS. Index Terms-Bluetooth low energy (BLE) receivers (RXs), digitally controlled oscillator (DCO)-based receivers, discretetime (DT) receivers, Internet-of-Things (IoT), phase-tracking RXs (PT-RXs), inverter-based RXs, current-reuse, ultra-low power (ULP), ultra-low voltage (ULV).

A Novel Waveform-Tracking BLE Receiver

“…A behavioral model is built in Simulink to verify the effectiveness of the proposed WT-RX 3 . The simulated output waveforms of PI and OTW are demonstrated in Fig.…”

“…Due to operating in the crowded 2.4-GHz ISM band, the BLE RXs are expected to face more stringent specifications for the nextgeneration IoT nodes. The phase-tracking receivers [1], [2], [3], [4] have been proven superior in energy efficiency, while their selectivity is severely restricted from further improving by the loop delay as a result of the required sharp filtering. Despite the fact that our previous work in [2], [3] meet the BLE specs with a 2∼3-dB margin, they compromise the selectivity compared to the conventional open-loop Cartesian RXs, where the interference rejection ratio is only limited by the filter bandwidth as long as the RX is not saturated.…”

“…The phase-tracking receivers [1], [2], [3], [4] have been proven superior in energy efficiency, while their selectivity is severely restricted from further improving by the loop delay as a result of the required sharp filtering. Despite the fact that our previous work in [2], [3] meet the BLE specs with a 2∼3-dB margin, they compromise the selectivity compared to the conventional open-loop Cartesian RXs, where the interference rejection ratio is only limited by the filter bandwidth as long as the RX is not saturated. A >10-dB specification margin can be easily obtained with a highorder sharp filtering [5], [6] and optimization on the linearity of subblocks [7], [8].…”

Design of High-IF Discrete-Time Receivers for IoT: Demystifying Aliasing Trade-Offs

“…Among the architectures proposed, the most prominent ones are the N-path filter (NPF) [16], [22] and the charge-sharing band-pass filter (CS-BPF) [23]- [25], or even a combination of both topologies [26]. More recently, modifications to the filter or to the RX architecture were also tested to improve the performance of the DT-RX for IoT [4]- [6], [27], [28]. Although both types of filters, NPF and CS-BPF, have been extensively used in the past years, the CS-BPF exhibits a feature that makes it superior to the NPF: While the NPF shows replicas inside the sampling frequency domain, the CS-BPF has only one peak in the interval −f s /2 to f s /2 [7].…”

Design Considerations for a Sub-mW Receiver Front-End for Internet-of-Things

“…[11], [26] have low NF and better IIP3 but power dissipation and area are much higher. Sub-mW RX have a NF at least 5 dB higher [9]- [10] or low in-band IIP3 [15], [48], [68]. DCR solutions with strong BB filtering [9], [37] tend to have very good IIP3 and low power, at the price of increased sensitivity to flicker noise.…”