Blind Channel Parameter Estimation For Multi-carrier Modulation Systems

With the rapid development of mobile communication technologies, the voice service is notthe only requirement for mobile phone users. Currently, a series of mobile services has becomeessential for users, such as mobile phone videos, online games, etc. The emergence of theseservices is accompanied by a rapid increase in data service, multimedia of mobile business, andhigh-speed and broadband data transmission. The technologies such as Orthogonal FrequencyDivision Multiplexing (OFDM) and Multiple-input-multiple-output (MIMO) have become hotresearch topics in wireless communications because of their characteristics of high bandwidthefficiency, fast transmission rate, high bandwidth and large capacity. Moreover, the physical layeralgorithms in the wireless communication systems require accurate channel parameter informationsince the accurate channel parameter information is helpful for improving the performance ofcommunication systems. Thus, the channel parameter estimation problem has also attracted muchattention of the researchers.The communication process will be affected by external interferences due to the specialcharacteristics of transmission in mobile communications. Moreover, along with the problems ofthe scarcity of spectrum and the low spectrum utilization, blind channel parameter estimation hasbecome hot research topic. In this dissertation, we focus on a number of channel parameterestimation problems under the external interference environment, which include blind signal-to-noise ratio (SNR) estimation, blind signal-to-interference-plus-noise ratio (SINR) estimation andblind channel estimation. The main contributions of this dissertation are the following.1. Based on the cyclostationarity statistics of OFDM signal, a blind SNR estimation method inMIMO-OFDM systems is proposed. In this method, cyclic prefix and suffix are introduced intoeach symbol. By setting the different cyclic suffix lengths, the cyclic correlation of transmit signalof each antenna will have different zeros. Then, the MIMO channel system can be transformed intoa number of single-input single-output (SISO) channel systems by selecting appropriate cyclicfrequencies. After that, the blind SNR estimation method is formulated by analyzing the energydistribution of cyclic correlations. The new method can provide accurate SNR estimate for eachchannel, and performs better than the existing methods. 2. A blind SINR estimation method is proposed for OFDM systems in the presence ofimpulsive interference. By analyzing the energy distribution of cyclic correlations of OFDM signal,impulsive interference and Gaussian noise, then selecting appropriate delay variables and cyclicfrequencies, the received signal power is decomposed into the transmitted signal power and theinterference plus noise power. According to the definition of SINR, a novel blind SINR estimationmethod is formed. Compared with the Two Order and Four Order (M2M4) method, the newalgorithm has higher accuracy and lower computational complexity.3. By combining the cyclostationarity characteristics of the OFDM signal and applyingpretreatment of the received signal, a blind channel estimation method in the presence of impulsivenoise is proposed. Computer simulations show that this method has higher estimation accuracy andsmaller bit error rate.