Long-echo Suppression Channel Estimation Algorithm Optimization And VLSI Implementation In DTMB System

With the double drives of the techs and policy level, digital TV has welcomed the popularization period. In the three transmission methods, the satellite, cable and ground transmission, due to the most complex ground channel environment, the channel estimation of ground transmission is huge challenge. Especially, in the long echo channel, siginals are influenced with long interference and intensity of larger diameter, so it is very important to eliminate multipath effects. DTMB is our country’s independent standard design with an unique form of PN sequence frame head which is more easily influenced by long echo channel. So the research of the algorithm and hardware implementation of channel estimation has a strong theoretical and practical significance. This thesis proposes a new design and study how to eliminate the impact of long echo channel, through the reconstruction of receiving data to obtain more accurate channel estimation response value. With slight performance loss, the new proposal can reduce hardware cost.The existing research on channel estimation based on DTMB system has certain achievements, but there is still has considerable room to improve hardware cost.The improved algorithm proposed in this paper, on the basis of the existing research results, has not changed the basic idea of iterative feedback, but is operated in time domain. In this algorithm, all the calculation are completed by only one convolution module, which can greatly reduce the amount of computation module (such as multiplier and FFT/IFFT).The calculation modules and their corresponding RAMs take up most of the hardware cost, so to reduce the use of these is the key method to improve the DC area output. On the other hand, through the reasonable optimized sequence arrangement, again optimizes the amount of RAM, eventually making the output of hardware cost of the proposed design to achieve improvements than before.