Improved Ultra-wideband Pulse Shaping Technique Based on Spectrum Shifted Gaussian Waveforms

In the ultra-wideband (UWB) communication system, Gaussian monocycle is the most frequently used due to its simplicity. However, in a Gaussian monocycle, the most of signal energy is concentrated in low frequency band, in which direct current components exist. Therefore, it cannot fully meet the U.S. Federal Communication Commission (FCC) on radiation template requirements. Meanwhile, in order to avoid the interference with other wireless systems, good pulse design is particularly important. Accordingly, several pulses were proposed, such as Scholtz's pulse, Hermite pulse, and spectrum shifted Gaussian waveforms (SSGW) pulse. However, it was shown that coefficient adjustment of SSGW pulse requires too many computations or iterations. In this paper, in order to overcome this drawback, random selection (RS) and Least Square Error (LSE) algorithms are introduced to the SSGW pulse design. Correspondingly, an improved SSGW waveform design is presented. Our analysis and results show that the improved approach has significant flexibility in designing and therefore it can meet the requirements of FCC spectrum mask better.