John W. McCorkle

The importance of fractional bandwidth in ultra-wideband pulse design

Ultra-wideband (UWB) communications systems provide extremely high data rates while using low transmit power relative to narrowband systems. Previous research concludes that such systems have characteristics that make them appropriate for high-speed wireless network applications. One of the distinguishing aspects of the UWB signal is an extremely high fractional bandwidth when compared with conventional narrowband systems. This paper describes how a large fractional bandwidth leads to lower worst-case fading in the presence of multipath for UWB communications systems, thereby providing a significant advantage for a low-power wireless systems. Also included are measurements made using an actual UWB communications systems showing the magnitude of signal strength variations due to multipath interference.

Ultra wide bandwidth (UWB): gigabit wireless communications for battery operated consumer applications

Regulations legalizing UWB provide the first ever opportunity to build radios that suffer no Rayleigh fading. This fact has tremendous ramifications in classical thinking on communication architectures. For example, in all other comparisons between DS and OFDM, both systems had to operate in Rayleigh fading. But with UWB, the comparison is new. DS is not Rayleigh faded, giving it a marked advantage in terms of prime power, complexity, and speed*distance2 performance. With a low complexity (no multiply), rake and decision feed-forward/feed-back equalizer, DS is able to utilize more channel energy than its OFDM counterpart. And the equalizer allows operation at Gbps+ speeds. Though the DFE in DS-UWB chips is small (only 3% of the die, and 3% of the power budget in the XSI-110 chip set), beware, DS does not work at high speed without it. This talk addresses the history of UWB and the theoretical basis for its advantages in short-range high-speed applications. It also addresses the challenges the technology poses for implementation in a VLSI product, particularly regarding the implementation impacts of the two candidate proposals being considered for the IEEE 802.15.3a standard at the time of this writing; direct sequence (DS-UWB) (Kohno, McLaughlin and Welborn) and multi-band OFDM (MB-OFDM) [Batra et al.]. For example, to maximize network capacity and minimize power drain, it is advantageous for each user to burst at the highest possible data-rate, and thus leave more time for other users, and more time to sleep. But this bursting requires higher voltages which makes the design very sensitive to the peak-to-average power ratio (PAPR) at the IC pin. BPSK DS has the same PAPR at the IC pin as a sine wave-much lower than the noise-like signal of OFDM-allowing it to operate at much lower voltages and without a power amplifier. In a unique convergence of attributes serving the needs of multiple UWB applications and markets, DS eliminates fading without hardware, scales to Gbps+ speeds, is capable of full-power high-speed bursts in low-voltage low-cost silicon processes, and does these in a low-complexity small die. Continuing process refinements in low cost sub-micron CMOS, SiGe, and SOI solidify UWB as the technology of choice for battery powered ultra-high-speed short-range high-capacity wireless networks.