Study On The Crucial Techniques Of Loran-C Digital Signal Processing

Although GNSS is unquestionably the best radio navigation system, it can not be used wildly because its signal is weak and interfered easily. The research result shows that Loran-C should be viewed as an excellent complement to GNSS, and it has not yet reached its technological limit. Contemporary Loran-C receiver performance will be vastly improved in the next 10-20 years. For example, new receivers are all-in-view and track 30-35 Loran transmitters simultaneously in North America. A study on crucial techniques of Loran-C digital signal processing is carried out in this paper.Two approaches are studied in detail in this paper. The first approach, which is the research of crucial techniques of Loran-C digital receiver using DSP and FPGA ,such as digital FIR, adaptive filter, time delay estimation, reducing strength of the algorithm and so on. How to demodulate the timing message which transmitted by the Loran-C data link is the second approach. The followings are the chief tasks.Carry wave interference is studied in this paper. Most of the interference may be suppressed by means of a high-order digital FIR filter. The remaining synchronous interference may be suppressed by notch filter. Firstly, digital FIR filter is designed by Filter Design & Analysis Tool ( Fdatool ) in Matlab. Secondly, the coefficients of FIR can be expressed in Canonical Signed Digit (CSD) representation to minimize the number of additions/subtractions required in each coefficient multiplication. On average, the CSD representation can reduce 33% of the non-zero digits compared with the binary representation. To further reduce hardware complexity, the intensity reduction of the algorithm is studied finally.To suppress synchronous interference, the adaptive filter is searched. At the beginning, two kinds of adaptive algorithm, LMS and NLMS, are researched. Furthermore, according to the character of Loran-C signal,the performance of both algorithms are simulated,analyzed and compared. On the basis of two algorithms, a new algorithm, GNGD, is studied in this paper in the first time. The GNGD represents an extension of the Normalized Least Mean Square (NLMS) algorithm. It can overcome the initial parameter sensitive problem that exists in the predistorter using NLMS algorithm. The proposed predistorter also gets a better linearization performance than the predistorter using NLMS algorithm. The result of simulation shows the validity of this GNGD algorithm.Skywave interference generally affects the performance of Loran-C receivers. The skywave rejected methods employed in current design of receivers are not optimal because they use fixed, worst-case, sample timing. IFFT spectral-division technique is presented for isolating groundwave and skywave in Loran-C receivers in this paper. The full comprehensive analysis of the parameters that might affects its performance is studied. This approach is rejected because of its poor resolution and failure at low signal-to-noise ratios. Therefore, a class of high-resolution is identified : these include MUSIC algorithm,eigen-vector decomposition algorithms. Finally, a new algorithm, modified ESPRIT algorithm,is studied in the first time in this paper. All algorithms are simulated and compared with each other. The improved ESPRIT algorithm is promoted as being suitable for real-time applications because it eliminates the requirement of a time consuming parameter search.Finally,Loran-C datalink is studied in this paper. Loran-C datalink is a data communication system which can broadcast timing message by additional modulation of Loran-C signa1.two kinds of timing message is presented. The timing message , broadcast by xuan cheng station is demodulated using signal process program.