A Data-aided Synchronization Algorithm Based On TDT For UWB System

1. IntroductionUltra-Wideband (UWB) technology becomes more and more attractive because of its advantages, such as high transmission rate, low power consumption, good security, robustness to multipath and low cost. UWB wireless communication technology uses the principle of overlapping to share the spectrum occupied, it is a novel short-distance wireless communication technique, in which data are transfered using non-sinusoidal pulses with very short duration, so it is fit for using in indoor environments which is a typical multipath environment. And it can be coexisted with the existing wireless communication systems at the same frequency band.In digital communication systems, signal synchronization is essential to the system, and synchronization is the precondition of the transmission of information, the synchronized performance reduces can cause the communications system performance directly decline, even enables the system can't normal work. So is the UWB receiver, the receiver needs to know the propagation delay of the received signal, and make local template signal synchronization with the received signal to accomplish despreading and demodulation. Further more synchronization is an especially difficult task in UWB system which employ spreading codes to distribute the transmitted signal energy over a wide bandwidth. Though synchronization is also a challenge in narrowband systems, its difficulty is accentuated in UWB due to the fact that the information-bearing waveforms are impulse-like and have low power. Moreover, the (oftentimes dense) multipath channel, through which these low-power narrow pulses propagate, is unknown at the receiver during the synchronization stage. These reasons explain why synchronization has received so much emphasis in UWB research.Acquisition schemes for UWB systems can be broadly classified into those which follow detection-based approaches, and those which rely on estimation-theoretic strategies. And the estimation-based schemes can be broadly classified into data aided estimation and nondata aided estimation. Data aided estimation based on pilot symbol, which can lead to the waste of pilot width and power resource, thus decreases efficiency of the system; the nondata aided timing estimation exploit the signal structure inherent in UWB to estimate timing information of the received signal, which overcomes disadvantages of resources waste. In this paper, a estimation-based schemes is presented, and a design of training sequence for data aided estimation is provided.2. A data-aided synchronization Algorithm based on TDT for UWB systemConventional synchronization techniques based on pulse-rate sliding correlation are not only sub-optimum in the presence of dense multipath, but also very slow to converge, due to the prohibitively large number of fine bins (chips) to be searched over. This techniques is obviously acquiring very slowly, and the acquisition time is unacceptable and can not be used in practice. Therefore, we need a fast and effective way to solve the UWB system synchronization. In this paper, a synchronization algorithm for PAM-UWB system which adopts the method of combining the TDT algorithm and cyclostationary approach to achieve synchronization acquisition. Firstly, utilize the TDT algorithm to achieve frame-level synchronization acquisition, then carry on the timing compensation for the received signal, afterwards we apply the cyclostationarity-based timing estimation principle to achieve pulse-level synchronization.Most existing synchronizers are based on the unique maximum of the received pulse's autocorrelation function, which requires a"clean template"of the received pulse to be available. Evidently, the latter is not feasible when the multipath channel is unknown. TDT algorithm relies on the fact that the cross correlation of"dirty templates"extracted from the received waveform exhibits a unique maximum at the correct timing. In this scheme, a symbol length segment of the received waveform is used as a template and correlated with the subsequent symbol length segment, it makes use of cross correlations between adjacent symbols to estimate timing information of the received signal,the template signal is affected by the channel and noise. In the timing synchronization we don't need to know the channel information, so the algorithm does not need to estimate the channel response, so it is low complexity, and is a fast synchronization algorithm. And after the timing compensation we apply the cyclostationarity-based timing estimation principle, we makes pulse-rate sliding correlation in one frame, relative to making pulse-rate sliding correlation in entire symbols the acquisition time is much more less. A symbol often constitutes with several to several dozens frames, the frame number is more, the advantages of this algorithm provided in this paper is more obvious. The proposed algorithm in this paper inherited TDT's merit, such as lowly complexity, fast synchronous speed, in the case of relatively low complexity system could achieve a higher pulse synchronization accuracy. And the simulation results show that the algorithm is more practical and effective.3. Design of training sequence for data aided estimation synchronization AlgorithmData aided estimation based on training symbol,such as M sequence, Gold sequence and Barker Code. Compare to nondata aided estimation, data aided estimation has better acquisition performance, but the problem is the waste of pilot width and power resource, thus decreases efficiency of the system. So we should carry on the reasonable design to the training sequence length and the energy, then enhances the system performance. The good training sequence design may enhance the system synchronization capture accurate. To account for both performance and bandwidth efficiency, we design the training sequence to minimize the channel's mean-square estimation error, and maximize the average capacity.In the design, the question that we mainly considered is for any given size and energy of the burst, how to find out the optimal number of the training symbol and optimizing the training symbol energy. Not only, these parameters influence synchronization capture performance, but also affect the channel estimate and the data demodulation performance, also affect the data transmission speed. These parameter minimize the channel's mean-square estimation error , moreover maximize the average capacity. And the simulation results show that after the training sequence optimized design, the system performance has certain enhancement.