Transmitted Waveform And Receiver Design In IR-UWB System

Impulse radio Ultra wideband(IR-UWB) based on the transmission of short duration pulses has been one of the most promising short range communication technologies in recent times, which has the virtues of low power consumption and high data rate. The selection of transmitted pulse waveform and corresponding receiver design are key techniques in UWB system.Due to the occupied ultra wide bandwidth, the UWB signals must satisfy given spectral mask so that the coexistence with current systems can be realized. This paper focuses on the improvement of the normalized effective signal power (NESP) of the pulse shape, i.e., the increase of effective transmission power, which will improve the system performance. In order to solve the problem of coexistence of UWB system and current systems, the mathematical model of pulse shaping method which is based on the linear combination of time shifted pulses is introduced, and then this method is translated into the design of a FIR filter and solved using Parks-McClellan algorithm. The NESP obtained using PM algorithm is 74.56% under the condition of low complexity of algorithm. Based on the PM algorithm, two modified algorithms, which are the elimination of effect due to base pulse spectrum envelop and additional combination coefficients, are proposed, and the NESP of the obtained pulse can reach 86.27%.The detection method is then considered after the design of UWB pulse waveform. In digital UWB systems, the sampling rate of A/D converter will be extremely high using traditional receiver in the time domain, like RAKE Receiver, because of the utilization of pulses in nano-second to carry information. In order to realize the receive of UWB signals in lower sampling rate of A/D converter, a receiver algorithm in a frequency domain is consider in this thesis. The information in a time domain can be reconstructed using the spectrum component, and then the energy of multipath is collected to detect the valid signal. Firstly, the theory of receive algorithm in the time domain is introduced from the perspective of signal analysis, and then the basic structure of the receive algorithm is founded. Finally, simulations are given in the AWGN channel and multipath channel. Simulation results show that, in the AWGN channel, the receiver in the frequency domain using 7 fingers has similar performance with the theory when applying the pulse obtained with PM algorithm; in the multipath channel, the receiver in the frequency domain using 10 fingers has the same performance with ideal RAKE receiver using 10GHz sampling rate. When the influence of transmitted pulse to the receive algorithm in the frequency domain is considered, the BER performance using the obtained pulse is superior to Gaussian doublet in the condition of low bit rate, however the performance improvement will be lower in the condition of high bit rate.