This letter presents the first neural front-end specifically targeting intravascular recording with direct frequency-domain encoding for body-coupled communication. It employs a differential amplifier and voltage-controlled oscillator (VCO)-based amplitude-to-frequency encoding. Fabricated in TSMC 180 nm CMOS technology, the proposed architecture integrates a fully differential capacitive-feedback low-noise amplifier with a current-starved ring VCO to directly translate microvolt-level neural signals into frequency variations. The design occupies 0.0025 mm2 and operates from externally supplied ±0.6V DC rails, intended to be delivered through the catheter in the target application, without any on-chip negative-voltage generator. Measured results demonstrate linear amplitude-to-frequency conversion up to 2mVpp input with a minimum detectable signal of 20 μVpp. The measured bandwidth is 6 Hz–6.7kHz. The total power consumption is 11 μW under PCB loading conditions. The integrated equivalent input-referred noise is 5.55 μVrms over the spike band (300 Hz–6 kHz). By prioritizing extreme miniaturization and frequency-domain robustness, the proposed design enables neural interfaces compatible with catheter-based deployment and body-coupled transmission.

A 0.0025 mm2 CMOS Neural Front-End With VCO-Based Frequency Encoding for Body-Coupled Communication / Meimandi, A., Crovetti, P.S., Citraro, E., Andriulli, F.P., Carrara, S.. - In: IEEE SENSORS LETTERS. - ISSN 2475-1472. - ELETTRONICO. - 10:9(2026), pp. 6008404-6008404. [10.1109/LSENS.2026.3716537]

A 0.0025 mm2 CMOS Neural Front-End With VCO-Based Frequency Encoding for Body-Coupled Communication

Crovetti P. S.;Citraro E.;Andriulli F. P.;
2026

Abstract

This letter presents the first neural front-end specifically targeting intravascular recording with direct frequency-domain encoding for body-coupled communication. It employs a differential amplifier and voltage-controlled oscillator (VCO)-based amplitude-to-frequency encoding. Fabricated in TSMC 180 nm CMOS technology, the proposed architecture integrates a fully differential capacitive-feedback low-noise amplifier with a current-starved ring VCO to directly translate microvolt-level neural signals into frequency variations. The design occupies 0.0025 mm2 and operates from externally supplied ±0.6V DC rails, intended to be delivered through the catheter in the target application, without any on-chip negative-voltage generator. Measured results demonstrate linear amplitude-to-frequency conversion up to 2mVpp input with a minimum detectable signal of 20 μVpp. The measured bandwidth is 6 Hz–6.7kHz. The total power consumption is 11 μW under PCB loading conditions. The integrated equivalent input-referred noise is 5.55 μVrms over the spike band (300 Hz–6 kHz). By prioritizing extreme miniaturization and frequency-domain robustness, the proposed design enables neural interfaces compatible with catheter-based deployment and body-coupled transmission.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/3014874
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo