Degree Name

Master of Engineering Research


School of Electrical, Computer and Telecommunications Engineering - Faculty of Informatics


Ultra-wideband (UWB) Communication is currently considered as a key technology of the next generation wireless personal area network and wireless local area network. The use of very wide transmission bandwidth brings significant advantages in terms of high speed as well as low power transmission compared to traditional narrow band technologies. Ultra-wideband operation does not require a spectrum license but ultra-wideband devices are required to share spectrum with licensed narrow band users. As a concept to solve coexistence issues with other devices, cognitive radio used with ultra-wideband now becomes a new hot topic. Two mainstream development directions of ultra-wideband, MB-OFDM and DS based UWB, have all provided some cognitive solutions. However, both of them involve complicated computations such as IFFT/FFT and multiple pulse combination, not only increasing system complexity but also increasing manufacturing cost and power consumption.

This thesis presents a novel system for DS based UWB. The proposed dynamic bandwidth direct sequence (DBDS) system, focused on exploring a new idea rather than concentrating on impulse manipulation, provides a cognitive solution for DS based UWB in much simpler and more efficient way. Enhanced from the original MCIDS algorithm, this system is able to transfer data under a fraction of original spread spectrum signal bandwidth and different spectral shapes while maintaining the same data rate. This system does not require generating specific impulse for working environment, therefore significantly reduces system complexity. Different types of filters in the system enable a variety of potential transmission spectrums to satisfy cognitive radio needs. This thesis also introduces a symbol combining mechanism over the original MCIDS to guarantee system performance under multipath channel.

Simulation results demonstrate that the DBDS has a very exciting performance. Even received with different bandwidths and different spectral shapes, the data information can still be fully recovered at the same data rate. The BER-Eb/No curve in ideal Gaussian channel exactly matches the theoretical curve, indicating that this system performs lossless with partial signal bandwidth. In multipath channel, the DBDS system still offers an excellent performance, incurring only a slight loss due to the cyclic prefix compared to the result in ideal Gaussian channel.

02Whole.pdf (601 kB)