Data Estimation Algorithm On HF Channel

In the modern wireless communications, transmission of digital information is inevitable. In the 1980s, HF communication plays an important role as one of the most important communication methods linked to the computer technology. HF channel is influenced directly by three kinds of selective fading respectively on time, frequency, and space which will cause an ISI (Inter Symbol Interference). The performance of HF communication system is influenced directly by the ISI .In digital communication, no matter what kinds of modulation and frame structure it is, the signals must be received rightly in terminal finally, so the data estimation is very important. In the paper, two basic channel estimate algorithms, NDDE and PML are simulated in Matlab. In both NDDE and PML, channel estimate is one of the most important parts, so the channel estimate is discussed as a single part in this paper. Here , HF channel is considered as a filter of limit length. The length of the channel is determined by the autocorrelation of the known symbols. The generally algorithms of channel estimate are LMS and RLS which are compared in the paper. Finally, RLS algorithm is selected to estimate channel based on the HF channel and the frame structure. The technology of NDDE is based on DDE, in which the known data and the unknown data are transmitted alternately. Through simulated, this method can eliminate the ISI effectively. In the end, some simplified NDDE is used to reduce the computational complexity. PML is new algorithm to estimate the data using the ML(the maximum likelihood) criterion., and it is a symbol-by-symbol estimation which has less computational complexity than VA(Viterbi Algorithm). The performance of the PML algorithm is good. In addition, there is a tradeoff between the computational complexity and the performance. The simplified PML has advantages in terms of computational complexity and memory requirements for channels with long CIR(channel impulse response). Each method is based on different criterion, and has different algorithm and different computational complexity. There are advantages and disadvantages in each method.