this research is concerned with space-time block coding and OFDM techniques. In this thesis, the design and analysis of space-time and space-frequency block-coded OFDM transmitter diversity techniques are studied. Also investigated are issues related to the channel parameter estimation that is essential to the operation of space-time and pace-frequency block-coded OFDM transmitter diversity systems The remainder of the thesis is organized as follows. In Chapter 2, we provide the necessary background information for presentation of our research. Included in this chapter are brief overviews of characterizations of mobile communication channels, di- versity techniques, and the OFDm modulation scheme. Chapter 3 provides a detailed description of the space-time and space-frequency block -coded OFDM transmitter di versity systems. Chapter 4 provides a detailed description of the bandwidth efficient iterative space-time and space-frequency block-coded OFDM transmitter diversity systems. Chapter 5 describes channel estimation techniques for OF DM systems with transmitter diversity Finally, Chapter 6 summarizes the thesis, its contributions, and outlines future work in this area
Chapter 2 Background In this chapter, we lay the foundation necessary for the discussion and development of pace - time and space-frequency block-coded oFDM transmitter diversity techniques frequency-selective fading channels. a brief overview is provided on the mobile communications channel, diversity techniques, space-time block-coded transmitter diversity, and the OFDM modulation technique 2.1 Mobile Communication Channels Mobile radio propagation channels belong to a class of communication channels known s multipath fading channels. In wireless communications, the transmitted signals generally travel through multiple propagation paths before reaching the receiver. As a result of reflections, scattering, and diffraction, the multipath components arrive at the receiver from many different directions and with diferent attenuations and delays. The multipath components combine at the receiver either constructively or destructively depending on their relative amplitudes and phases. Because of the time-varying properties of the propagation media, the scatterers, and the relative movement between the transmitter and the receiver, the amplitude of the received signal varies(or fades)when observed over time, hence the term multipath fading
2.1.1 Modeling of Multipath Fading Channels The time-varying impulse response of the multipath fading channel can be derived follows [10-12. Consider the transmission of a bandpass signal Rez(t)e j2r」e where r(t)is the complex baseband signal and fe is the carrier frequency. Assuming the multipath fading channel is comprised of a discrete number of propagation paths the received bandpass signal is (t)=Re fy(t)e/ar/ley where the received complex baseband signal y (t)is given by y(t)=2an(t)expf-j2T[efDn(t))(t)-fDn (t)tl](t-n(t)+n(t)(2.3) and where an(t)and Tn(t are the time-varying attenuation factor and propagation delay, respectively, of the n-th propagation path, and n(t) is the channel noise. The Doppler, or frequency, shift fD, induced by mobile receiver movement on the n-th multipath component is given by ∫Dn(t)=fcos6n(t), (24) where fD=v/Ac is the maximum Doppler shift, u is the velocity of the receiver, Ac is the wavelength, and 0n()is the incident angle of the n-th multipath component with respect to the velocity vector of the mobile receiver. The receive complex baseband Sigmat ca )=∑an(t)er(t-n()+n(
here n(t)is the time-varying phase associated with the n-th multipath component given by dn(t)=2π[f∈+fon(t)n(t)-fon()引 (26) From(2.5 ), the equivalent complex baseband channel can be modeled by the time- varying impulse response h(t:)=∑an()e-108(t-n(), (27 where h(t; T)represents the response of the channel at time t to an impulse applied at time t-T and d() is the Dirac delta function Since fe+fD(t)is usually a very large value, even a small change in the path delay Tn(t) can cause a large change in the phase on(t). As a result, the received signal, which is the sum of the multipath components, can vary by tens of dBs with relatively small changes in spatial location. This form of signal variation is referred fast fading or, some mply multipath fading. The mobile exhibits longer term variation in the mean signal level, usually referred to as slow fading. Slow fading is caused by large scale terrain effects such as shadowing by buildings or mountains, so it is also referred to as shadowing. Empirical studies have hown that the mean envelope of slow fading follows a log-normal distribution with a standard deviation dependent on the carrier frequency and environment. Slow fading is usually handled by either increasing the marg transmitter power to cope with worst-case conditions or by employing other system techniques such as the single frequency network, also known as simulcast[13, 14. However, the deep fades associated with fast fading severely degrade the performance of the communication systems and are best mitigated using diversity techniques
The number of multipath components in(2.7)is typically large enough that the central limit theorem can be applied. Hence, the time-varying impulse response of the multipath fading channel can be well approximated by a complex Gaussian random process. In the absence of a line-of-sight( LOS)path or a specular component from a strong scatterer, the impulse response can be modeled as a zero-mean complex Gaus- sian process [15]. The envelope of the impulse response, h(t;T)L, Fluctuates about the local mean with a Rayleigh distribution, and the resulting channel is referred to as a Rayleigh fading channel. For a multipath fading channel containing a specular or LOS component, the impulse response can be modeled as a complex Gaussian process with nonzero mean. The envelope of the impulse response has a Ricean distribution[16 and such a channel is commonly referred to as a ricean fading chan- mel. Other empirical fading channel models such as Nakagami fading[17] have also been proposed. Since Rayleigh fading is usually the more challenging environment for mobile communications, the Rayleigh fading model is assumed throughout this thesis. The fading spectrum of a Rayleigh fading channel is a function of the antenna pattern [15, 18]. For an omnidirectional antenna, which is the most common type of antenna, the theoretical spectral density of the complex envelope of the received signal is given by E2 ∫≤fD; S(f) 2ft (28) where Eo is the root-mean-square(RMS)value of the signal envelope. For other pe of antennas the spectral density has a similar form and often differs only by a scaling constant. Notice that the multipath fading process is bandlimited to the