Ofdm System Beamforming

Orthogonal frequency division multiplexing (OFDM) is a promising technique which can be thought as a hybrid of multicarrier modulation (MCM) and frequency shift keying (FSK) modulation and it is the main candidate technology for the new generation communication system. The OFDM system with multiple antennas realizes the space diversity by adopting antenna arrays. There are three diversity technologies such as time, frequency and space which make the wireless communication system have stronger tolerance to noise, interference and multipath. This dissertation considers a hot problem of smart antenna in OFDM system with multiple antennas: Beamforming.The OFDM beamforming system based on reference signals is studied first. Usually, the reference signals used in this kind of OFDM beamforming system are pilot subcarriers. However, the limited resource of pilot subcarriers in some OFDM system make the performance of the system can not be improved further. Then, an algorithm based on pilot subcarriers and virtual subcarriers is proposed. More, from the views of convergence, power spectrum and packet error ratio, a complete analysis is conducted on the influence induced by the new algorithm. Results demonstrate that better performance can be reached obviously without influence on the efficiency of bandwidth.The beamforming algorithm based on eigen analysis for OFDM system have great potential in reducing the system complexity and improving the system performance. The first and key problem of this algorithm is to detect the number of eigen multipath channel. It decides the number of demodulation modules required in receiver directly. After analyzing the basic principle of this algorithm, the concept called effective eigen channel order is put forward at first. Secondly, combining the characteristic that theeffective eigen channel order is related to the amplitudes of paths and noise, the detection algorithm based on constant false alarm rate (CFAR) is proposed. The superiority of the new algorithm is demonstrated by extensive simulations in a stochastic multipath model and results show that it is benefit to system performance.The performance of extensively used beamforming algorithm based on the direction of arrival (DOA) is good. However, its good performance must have the basis that the estimation of desired DOA is always accurate. In reality, it is often difficult because of noise and other interferences. It influences the robustness of the algorithm. To solve this problem, a robust beamforming algorithm based on extended Bayesian filter is proposed. In this algorithm, the desired DOA is regarded as a random variable consists of some samples. The optimized weight is arrived by weighted the minimum variance distortionless response (MVDR) weights of these samples with their post probabilities. In addition, the samples with little post probabilities will be resampled and it improves the robustness of beamforming effectively.Space-time encoded OFDM (STC-OFDM) have the characteristic of great capacity and high bit rate. It has good foreground in future wireless communication system. However, the main disadvantage of this system is serious co-channel interference (CCI). Beamforming can be used to relieve the CCI with angular diversity. In this paper, we adopt the beamforming algorithm based on extended Bayesian filter to suppress the CCI and arrive in better performance compared to conventional algorithm.