Research On Multicarrier Modulation Technology In TETRA

TETRA digital mobile radio communication system is designed by the European Telecommunications Standards Institute (ETSI) to satisfy the require from Private Mobile Radio sectors in Europe. TETRA standard has developed TETRA Release2which is suitable for IP-based high-speed multimedia applications. TETRA Release2adopts a set of up-to-date technologies. The physical layer adopts filter-bank-based multicarrier modulation Filtered Multitone Modulation (FMT). TETRA system has been used widely in China’s public security, transportation and so on. But the equipments and technologies are monopolized by foreign countries. That makes the costs too expensive and limits the continued development of digital trunking system in our country. So developing TETRA independently has important significance both in theoretical and practical.This paper foruses on FMT modulation technology and its channel equalization and estimation. The fundamental of FMT modulation and the implementation of the filter banks in the multi-carriers modulation have been researched. Then paper presents the effective achieve way of non-stricted sampled FMT modulation, and analyzes its BER performance which is also be compared to OFDM in the AWGN. After studied FMT subchannels, paper concludes that the output signal is the product of the input signal and the frequency response of the transmitted channel under specific conditions. As a result, paper uses the post DMT in channel equalization, and analyzes the basic DFE, post DFE and two different linear equalization algorithms which are respectively based on MMSE and zero-forcing criterion through stimulation. The result of the stimulation presents the performance of single tap frequency domain equalization is worse than base DFE, but it has less computation and more practical. At last, paper researches on the inset mode of pilot signal in TETRA. And paper combines this mode to use two different channel algorithms, one is two-dimensional linear interpolation and the other is bayesian estimation in time domain and linear interpolation in frequency domain. The result of stimulation indicates the latter one has better performance.