




Abstract
Ultra-wideband (UWB) technology has been received significant interest in recent years due to the steep growth in demand and deployment of wireless personal area networks. Many open questions still exist in the design and implementation for UWB systems. Transceiver implementations are particularly challenging due to regulation, interference, synchronization, and channel estimation issues. This dissertation is intended to investigate the potential promises and proposes possible solutions to the challenges for the design of transceivers and optimization of system parameters.
In February 2002, the Federal Communications Commission (FCC) authorized UWB devices to operate in the frequency band currently occupied by existing radio services under specific emission restrictions, referred to as the spectral mask. Impulse radio is a popular technique for UWB transmission, which communicates using a baseband signal composed of sub-nanosecond pulses. In this dissertation, a new UWB pulse shape is proposed that satisfies the FCC spectral mask. Furthermore, a potential implementation of this pulse is proposed. Using this pulse, the link budget is calculated to quantify the relationship between data rate and distance.
A core idea using UWB is that the UWB signal can cohabit in the existing spectrum with existing applications without causing harmful interference to them. This fact opens promising opportunities for a variety of applications. However, narrowband interference emitted by existing networks in close proximity has to be suppressed. In this dissertation, a reduced-rank adaptive filtering technique is applied to the problem of interference suppression in UWB communications. The reduced-rank combining method, in particular the eigencanceler, is proposed which requires a shorter data record than a minimum mean square error Rake receiver.
A promise of UWB is that a UWB channel can be resolved into a significant number of multipath components. A Rake receiver can be employed to exploit the high multipath diversity. However, practical Rake receivers require knowledge of multipath delays and amplitudes. In practice, that is estimated through pilot aided channel estimation, producing imperfect channel state information, which leads to degraded performance. In this dissertation, the impact of non-ideal estimates on system performance is investigated when path delays and path amplitudes are jointly estimated. Further, this analysis is applied to determine optimal transceiver parameters, such as the fraction of a transmitted packet constituted by the pilot symbols, signal bandwidth, and the number of diversity paths combined at the receiver. Towards this goal, pilot symbols are transmitted, and the Cramér-Rao bounds (CRBs) of the variance of path delay and path amplitudes are derived. Using the errors obtained from the CRB, the effective signal-to-noise ratio and bit error rate (BER) are analyzed for a Rake receiver employing maximum ratio combining. Subsequently, the BER is used as part of a binary symmetric channel and the achievable information rates are evaluated.



