Mitigation techniques for mobile communications over fading channels
Date of Issue2016-06-01
School of Electrical and Electronic Engineering
The explosive expansion of wireless communications and people’s growing need for information exchange have imposed a great demand on developing wireless communication systems with high spectrum utilization, high energy efficiency and high quality of service. After nearly one hundred years’ development, with the evolution from regional systems to cellular systems, from analogue modulation to digital modulation, difference generations of communication systems have witnessed how scientists and engineers are pushing the transmission performance of wireless communication systems to the Shannon theoretical limit. Signal fading is one of the most crucial factors that impair system performance. In this dissertation, the main causes of fading are studied. Generally speaking, we have path loss factor, shadowing factor and multipath factor forming the causes of signal fading. The first two factors are classified as large-scale fading, and the last one is regarded as small-scale fading. Signals suffering from large-scale fading would have path loss as a function of distance and shadowing. Signals suffering from small-scale fading would have amplitudes and phases changed rapidly in a very small scale. After deducing the channel impulse response function, we would use a famous multipath channel model, called Jakes’ model, to analyze the small-scale fading in detail. There are two phenomena associated with small-scale fading, namely, signal dispersion and time-variant behavior. The signal dispersion results in frequency-selective fading or flat fading. The time-variant behavior produces fast fading or slow fading. As for frequency-selective fading, it means that the signal bandwidth is larger than the channel coherence bandwidth in the frequency domain and the signal period is smaller than the root-mean-square delay spread. The fast fading is due to varying Doppler shifts on different paths. The shift is always directly proportional to the relative velocity of the mobile station in the arrival direction of the received signals. For fast fading, signal period is larger than coherence time in the time domain and signal bandwidth is smaller than Doppler frequency spread in the frequency domain. The study of channel fading will help us understand possible degradations in mobile transmission over fading channels. In the second part of the dissertation, mitigation techniques, especially the multiple-input multiple-output (MIMO) scheme, are applied to reduce the negative effects on mobile communications based on the spatial channel extended model (SCME) provided by the 3rd-generation partnership project (3GPP). MIMO is a popular technique in 4G mobile communication systems, which can be used to combat multipath fading. Channel state information (CSI) can be used by MIMO precoding and beamforming to improve the MIMO system capacity and reduce the bit-error-rate. In a time-division duplex (TDD) systems, take LTE-A TDD for example, channel reciprocity is widely used to obtain CSI. However, ideal channel reciprocity doesn't exist in practical systems. The time-varying property of channels, which may lead to fast fading, would cause serious damage to channel reciprocity and weaken advantages of TDD systems. After analyzing the error performance of the TDD-MIMO systems using precoding and beamforming for the ideal case, we show that the error performance of the system would be worse with .the increase in channel changing rate. In response to the reciprocity loss caused by a time-varying channel, we utilize a compensation algorithm based on the auto-regressive channel prediction model and design an application method for the algorithm in TDD systems. Simulation results show that this compensation algorithm could effectively compensate the loss of reciprocity in TDD-MIMO channels.
DRNTU::Engineering::Electrical and electronic engineering