Font Size: a A A

Research On Key Techniques Of OFDM Systems Under Low SNR Condition

Posted on:2012-10-06Degree:DoctorType:Dissertation
Country:ChinaCandidate:T WangFull Text:PDF
GTID:1118330371962607Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
Orthogonal Frequency Division Multiplexing (OFDM) is a digital multi-carrier modulation technique, which uses several parallel orthogonal sub-carriers to transmit/receive a high data rate signal. OFDM shows many advantages over single-carrier transmissions, such as high frequency efficiency; robustness against frequency-selective fading in a multipath channel; flexibility to determine modulation manner and power of sub-carriers; natural ability to cooperate with Multiple Input Multiple Output (MIMO). At present, the researches on OFDM mainly focuses on the applications of high Signal-to-Noise Ratio(SNR) and high speed. Meanwhile, it shows an increasingly demand of OFDM transmissions under low SNR conditions, such as extending distence Ultra-Wide Band(UWB) systems, low detectable probability adaptive speed military communication systems and data transmissions between the brim and the base station in mobile communications.It is significant to develop the OFDM techniques under low SNR conditions. The major work and contribution of this dissertation are focus on the key techniques of low SNR OFDM systems, it summarized as follows:1. A new OFDM preamble design method is presented. It has been testified that if the OFDM frequency domain preamble has constant amplitude at all sub-carriers, the time domain preamble (IDFT form of frequency domain preamble) shows an ideal correlation performance, this kind of special OFDM preamble is named"Constant Amplitude Frequency Domain Preamble"in this dissertation. It is easy to generate the CAFDP sequence, and for real CAFDP sequence, the time domain form is central conjugate symmetrical.2. Two joint symbol timing and frequency offset synchronization methods are proposed: the differential cross-correlation synchronization method and the auto-correlation synchronization method. They are both multi-stage processes, the only difference between them is at the symbol timing stage. The former introduces differential operation to cross-correlation to eliminate the influence of large frequency offsets, the latter utilizes the conjugate symmetrical structure of CAFDP to get sharp auto-correlation peak. Computer simulations in the AWGN(additive white Gaussian noise) channels and the Rayleigh fading channels show that the proposed synchronization methods achieves superior performance to the existing methods.3. Aimed at obtaining better performance and fast synchronization convergence, a time and frequency two dimension joint synchronization method is present. It has excellent performance: high timing precision, large integer frequency estimation range, fast convergency speed etc. At the same time, the realization of the method needs great hardware resources. Based on the analysis of the mathematic expressions, two promoted algorithms are proposed which can realize the method with the same performance but saving much hardware resources, they are the parallel FFT synchronization algorithm and the parallel Chirp-Z synchronization method.4. Utilzing the cyclic correlation character of the CAFDP and the multipath delay distribution information achived at the timing synchronization stage, a time domain channel estimation method is proposed. This method shows several advantages over the existing channel estimation method: first, it uses the same preamble with the synchronization method, that means less systems consumed cost; second, it shows superior perfomace over existing methods; last, it is a low computing complexity algorithm that leads small hardware resources cost and fast processing speed.
Keywords/Search Tags:OFDM, constant amplitude frequency domain preamble, differential cross-correlation, time and frequency two domain jointed synchronization, parallel FFT, parallel Chirp-Z, time domain channel estimation
PDF Full Text Request
Related items