Researches On Non-coherent Receiver Technology For Impulse Radio Ultra-wideband Signals

With the development of short-range wireless networks, such as WPAN and WBAN, the market needs a novel physical layer technology that can resolve the conflict of short-range high-speed access demand and crowded spectrum resource allocation. The unique properties of UWB such as large bandwidth, very low power spectral density and fine time-resolution provide higher channel capacity, reduced fading effects and position location capability. Thus, UWB is considered an important solution of short-range wireless applications for its above characteristics.From a theoretical viewpoint, the coherent RAKE receivers are optimal for UWB multi-path signals in an AWGN environment. However, the coherent RAKE receivers are faced with formidable implementation difficulties due to high cost, stringent requirements on synchronization and channel estimation. In the past few years, there has been growing interest toward the noncoherent UWB receivers (such as energy detection receivers and autocorrelation receivers) for their simple implementation and good performance-complexity tradeoffs. Compared with coherent receivers, the noncoherent receivers do not require complex channel estimation, be robust to synchronization and have simpler structures. However, these benefits are obtained at the cost of some error performance loss. Therefore, how to improve the error performance of noncoherent receiver has been becoming a current research hotspot in the UWB domain. In this thesis, we address the issues of noncohrent receiver design and performance optimization approaches combined with the characteristics of UWB signals and channels. The main works and contributions are concluded as:1. Research on UWB weighted energy detection approach.For OOK-UWB signal, we propose a quadratic form weighted energy detection receiver based on fractionally-sampled scheme, and derive a closed-form expression for the bit error probability using the characteristic function method. Numerical analysis and computer simulation results are presented to validate the derived theoretical performance. Moreover, from the viewpoint of statistical signal detection, a comparative study between the OOK-UWB and PPM-UWB weighted energy detection receiver is carried out. The study results show the signal detection problem of this two receivers are essentially the Gaussian 2-ary signal maximum likelihood detection problem, and respectively correspond to the case of unequal means, equal variances and the case of unequal means, unequal variance under 2-ary hypotheses.2. Research on UWB blind energy detection receiver.To avoid the weight coefficients estimation, we propose a novel MSDD-based UWB blind energy detection receiver, which performs joint detection of a block of multiple symbols by exploiting the channel quasi-static assumption (i.e. the channel coherence time is less then the data block time). Simulations in multi-path fading scenarios show that, the proposed receiver performs considerably better than the conventional energy detection receiver, and with the increase of symbol block size, its performances approach those of the ideal weighted energy detection receiver.3. Research on reduced-complexity MSDD scheme.As far as the computational complexity of MSDD is concerned, we propose a reduced-complexity sub-block joint search algorithm by employing a two-step search strategy, which can reduce the search space efficiently and be implemented simply. The proposed algorithm preserves major desired properties of the optimum MSDD, while resulting in low complexity at the expense of some performance degradation. We also study the parameters optimization in the proposed algorithm through the computer simulations.4. Research on UWB auto-correlation receiver combined with block-coded modulation (BCM) scheme.Traditional UWB auto-correlation receivers suffer from energy loss and data rate reduction due to the transmission of non-data-bearing reference pulses. Based on the generalized likelihood ratio test (GLRT) optimality criterion, we propose a novel UWB non-coherent autocorrelation receiver combined with BCM scheme. Compared with BCM energy detection receiver, the proposed receiver adopts the autocorrelation structure that can be implemented in a more flexible and simpler manner. Moreover, based on the derivation of pair-wise error probability, the closed-form upper bound expression for the bit error rate is presented. Simulations in multi-path fading scenarios show that this upper bound is quite tight for large signal to noise rate (SNR) and can be treated as a good approximation of BER in high SNR regions. The optimized integral time and MAI suppression capability are studied through computer simulation.