Analog-to-Digital Converters (ADC) are a key component in a digital radio receiver. Modern wireless standards, such as Wi-Fi and 5G, push the limits for the resolution and bandwidth of ADCs to achieve higher data-rates. Time Interleaving (TI) becomes the dominant architecture at high sampling speeds, needed to match the requirements set by the standards. This thesis presents an in-depth analysis on TI ADCs and the impact on their non idealities on a wireless receiver performance. Calibrations are conventionally used in TI ADCs to enhance their performance but, as demonstrated in this work, they are not enough to guarantee a good performance when decoding a wireless signal. Randomization techniques are shown to be a solution, and the study focuses on both pseudo-random chopping-aided offset calibration and channel-order randomization. The two techniques are compared, analyzing both their advantages and disadvantages considering also the implementation non idealities. Channel-order randomization is found to be the most suitable solution, and an implementation of the key block to this technique, the phase generator, is proposed for the adaption on a 4 channel TI ADC with a sampling frequency of 2GHz using the 28nm TSMC technology node. Finally, a new timing skew mismatch calibration is proposed, which improves existing State-of-the-Art techniques for correcting sampling time errors when the channel order is randomized.
I convertitori Analogico-Digitali (ADC) sono un componente fondamentale in un ricevitore radio digitale. Gli standard wireless moderni, come il Wi-Fi e il 5G, spingono i limiti in termini di risoluzione e larghezza di banda degli ADC per raggiungere velocità di trasmissione più elevate. L’interlacciamento temporale (Time Interleaving) diventa l’architettura dominante alle alte velocità di campionamento, necessarie per soddisfare i requisiti imposti dagli standard. Questa tesi presenta un’analisi approfondita dei TI ADC e dell’impatto delle loro non idealità sulle prestazioni di un ricevitore wireless. Le calibrazioni sono convenzionalmente utilizzate nei TI ADC per migliorarne le prestazioni ma, come dimostrato in questo lavoro, non sono sufficienti a garantire buone prestazioni nella decodifica di un segnale wireless. Le tecniche di randomizzazione sono presentate come possibile soluzione, e lo studio si concentra sulla tecnica di calibrazione dell’offset pseudo-random chopping-aided e sulla channel-order randomization. Le due tecniche vengono confrontate, analizzandone vantaggi e svantaggi considerando le non idealità implementative. La channel-order randomization risulta essere la soluzione più adatta, e viene proposta un’implementazione del blocco chiave di questa tecnica, il generatore di fasi, adattato per un TI ADC a 4 canali con una frequenza di campionamento di 2 GHz, utilizzando la tecnologia TSMC a 28 nm. Infine, viene proposta una nuova calibrazione per la correzione degli errori di skew, che migliora le tecniche allo stato dell’arte esistenti quando l’ordine dei canali è randomizzato.
Analysis of calibration and randomization techniques in TI ADCs for wireless applications
AIROLDI, LEONARDO
2024/2025
Abstract
Analog-to-Digital Converters (ADC) are a key component in a digital radio receiver. Modern wireless standards, such as Wi-Fi and 5G, push the limits for the resolution and bandwidth of ADCs to achieve higher data-rates. Time Interleaving (TI) becomes the dominant architecture at high sampling speeds, needed to match the requirements set by the standards. This thesis presents an in-depth analysis on TI ADCs and the impact on their non idealities on a wireless receiver performance. Calibrations are conventionally used in TI ADCs to enhance their performance but, as demonstrated in this work, they are not enough to guarantee a good performance when decoding a wireless signal. Randomization techniques are shown to be a solution, and the study focuses on both pseudo-random chopping-aided offset calibration and channel-order randomization. The two techniques are compared, analyzing both their advantages and disadvantages considering also the implementation non idealities. Channel-order randomization is found to be the most suitable solution, and an implementation of the key block to this technique, the phase generator, is proposed for the adaption on a 4 channel TI ADC with a sampling frequency of 2GHz using the 28nm TSMC technology node. Finally, a new timing skew mismatch calibration is proposed, which improves existing State-of-the-Art techniques for correcting sampling time errors when the channel order is randomized.File | Dimensione | Formato | |
---|---|---|---|
2025_07_Airoldi_Tesi_01.pdf
non accessibile
Dimensione
29.31 MB
Formato
Adobe PDF
|
29.31 MB | Adobe PDF | Visualizza/Apri |
2025_07_Airoldi_Executive_Summary_02.pdf
non accessibile
Dimensione
2.83 MB
Formato
Adobe PDF
|
2.83 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/240537