An effective AMS Top-Down Methodology Applied to the Design of a Mixed-Signal UWB System-on-Chip Marco Crepaldi, Mario R. Casu, Mariagrazia Graziano and Maurizio Zamboni VLSI Laboratory, Dipartimento di Elettronica, Politecnico di Torino, Italy Email: marco.crepaldi@polito.it Abstract The design of Ultra Wideband (UWB) mixed-signal SoC for localization applications in wireless personal area net- works is currently investigated by several researchers. The complexity of the design claims for effective top-down methodologies. We propose a layered approach based on VHDL-AMS for the first design stages and on an intelli- gent use of a circuit-level simulator for the transistor-level phase. We apply the latter just to one block at a time and wrap it within the system-level VHDL-AMS description. This method allows to capture the impact of circuit-level de- sign choices and non-idealities on system performance. To demonstrate the effectiveness of the methodology we show how the refinement of the design affects specific UWB sys- tem parameters such as bit-error rate and localization esti- mations. 1. Introduction Ultra-Wideband (UWB) impulse-based communication received recent attention since the Federal Communica- tions Commission released the spectrum between 3.1 and 10.6 GHz for unlicensed use in 2002. Sub-nanosecond base-band UWB pulses can be directly sent to a wideband antenna without carrier modulation. Among the others, per- haps the most attractive feature of UWB is the locationing capability, enabled by the possibility of isolating the first echo of the signal received through a multipath channel. The large bandwidth, and the corresponding short time du- ration, is the key for such accurate time-domain resolution, which translates into an accurate distance measurement [3]. Providing UWB transceivers with locationing capabilities may open the way to a number of applications within the WPAN field (e.g. package tracking, search-and-rescue func- tions). An IEEE standardization task group is currently working toward an alternative physical layer of the 802.15.4 standard with the aim of enabling high precision localiza- tion (on the order of 1 meter) [4]. The final goal is the complete integration of UWB transceivers with locationing functions in the same System-on-Chip (SoC), possibly us- ing a standard CMOS technology. This paper thus deals with the design, modeling and simulation of UWB Mixed- Signal Systems-on-Chip (SoC). Recent papers discuss UWB design topics [7][10][9][11], however none of them focus on the in- teraction between system-level issues and circuit-level design and on the importance of using a uniform methodol- ogy. With the aim of closing this gap, we proposed the use of VHDL-AMS as a common description language used at both levels [1][2]. While not new as an approach to the description of telecom SoC [8][5], we applied it for the first time in the UWB context. However, once we pass from modelization to CMOS transistor-level design, VHDL-AMS is no more sufficient, and using Spice-like netlists and complex MOSFET models becomes mandatory. The risk when dealing with such fine- grain details is that of losing the system-level view and the impact that specific choices taken at this lower level may have on it. In this work we focused then on a methodol- ogy which permits the evaluation of the effect on system- level parameters, like bit-error rate (BER) or localization accuracy, of MOS-level design choices of relevant blocks of a UWB device based on the “energy detection” principle. We first summarize in section 2 the architecture of a UWB transceiver to which the simulation and design methodol- ogy, described next in section 3, has been applied. Section 4 presents the design of a relevant block in a CMOS tech- nology. Then we show in section 5 how system-level pa- rameters are affected by design choices and second-order effects by comparing pure VHDL-AMS results with mixed VHDL-AMS and Spice simulations. Finally section 6 sum- marizes the paper achievements and indicates future works in this field. 2. UWB receiver architecture The architecture of the transceiver based on energy de- tection of a 2-PPM modulated train of UWB pulses is de- 978-3-9810801-2-4/DATE07 © 2007 EDAA