Aalo, Valentine A.

Based on the theoretical method developed by Clark and Greenstein for frequency-selective Rayleigh fading channel, we develop a general model for frequency-selective Nakagami fading channel. We derive analytical expressions of the average bit-error-rate in an ideal space diversity mobile radio receiver using the matched filter bound. Our simulation results show that the influences of the diversity order, the shape of the multipath delay profile, and delay spread of the multipath delay profile. Five shapes are considered in our simulation. Our simulation results highlight the importance of the ratio the normalized delay spread d. The results show that the delay profile is of no importance for $d<0.3,$ but can have a profound influence for $d\geq0.3.$

Closed form results for the symbol error probabilities are derived for M-ary, L branch Selection diversity schemes and Partial Decision combining schemes in a Nakagami fading environment using noncoherent frequency-shift keying modulation. The Maximum SNR selection, the Maximum output selection, the Optimal partial decision, and the Majority voting schemes are considered in this analysis. The analysis is not limited to integer values of the Nakagami fading parameter m. Results for the Rayleigh fading channel are obtained and presented as special cases of those of the Nakagami fading model. The symbol error probability when the diversity branches undergo different fading statistics with unequal mean SNR are considered for the Maximum SNR selection scheme. Outage probabilities are also calculated for some of the diversity combining schemes. A comparison of all the four schemes are discussed using the performance curves.

This thesis is concerned with the performance analysis of a DS/CDMA packet radio system under Rician fading channel conditions. Analytic expressions are derived for the probability density functions and cumulative distribution functions of the total signal-to-interference ratio when considering mixed-type of interference sources. The users are divided into two groups: those that apply closed-loop power control and those that apply open-loop power control. Four different scenarios of power control were studied. The outage probability is evaluated for the four different cases of power control. A packet transmission protocol with forward error correction capabilities is considered, and the probability density functions of the signal-to-interference ratio are used to calculate the average block error probability as well as the outage probability of the system.

The performance aspects of conventional cellular (FDMA/TDMA) and CDMA systems with micro- and macroscopic diversity reception are investigated in a severe mobile communication environment which is characterized by path loss, correlated lognormal shadowing, multipath fading, background noise and interference. Under a co-channel interference-limited assumption, an exact analytical expression for the co-channel interference (CCI) probability is presented for a macroscopic diversity system with an arbitrary number of correlated macroscopic branches. For noise-limited systems, the average bit-error-rate (BER) and outage probability performances of a narrowband mobile communication system with micro- and macrodiversity reception are evaluated. In the relevant analysis, both Nakagami and Rician fading channels are considered. When both co-channel interference and noise coexists, the results for a Nakagami fading channel show that diversity reception can be used to reduce the effects of interference while combating fading and shadowing. Micro- and macroscopic diversities are also applied to a multicell DS-CDMA system. In a conventional cellular system with macroscopic diversity, the mobile user is usually connected to the closest base station. However, a base-station selection scheme based on a least attenuation criterion is shown to provide a significant performance improvement over the conventional system. In this case, the system performance is examined in terms of BER and outage probability, while accounting for the effects of path loss, correlated shadowing, multipath fading, multiple access interference, and imperfect power control.

This dissertation investigates the performance of different space diversity combining techniques for the wireless mobile communication systems employing Direct Sequence-Code Division Multiple Access (DS-CDMA). It covers two research topics, all falling under the umbrella of diversity combining techniques. The first part deals with diversity reception of wideband DS-CDMA signals in which the diversity branches experience some correlation. This analysis is performed without the usual assumption that diversity branches are independent and hence uncorrelated. In this case, the analysis is limited to the conventional diversity techniques such as maximal ratio combining, equal gain combining and selection diversity. In particular, the effect of correlation on system performance using two correlation profiles namely, the constant and the exponential correlation profiles is explored. In the constant correlation, the correlation between adjacent antenna are equal regardless of the separation between them, while in exponential correlation, it is assumed that the level of correlation between adjacent antennas decreases as their separation increases. The second topic deals with the development of new combining techniques. Two new techniques--the Generalized Selection Diversity Combining (GSDC) and the Maximal Ratio-Selection Diversity Combining (MR-SDC)--are introduced and analyzed in this dissertation. Analytical models which can be used to evaluate the performance of these new techniques are indicated. Results show that GSDC perform significantly better than the conventional selection diversity when two or more larger signals are selected. The MR-SDC technique accounts for the possibility of using a large array of antennas and for occasions when the receiving antennas may not be collocated. Also, it is shown that for the large array of antennas, better performance is achieved when the MR-SDC is employed with the maximum number of subgroups. The figures of merit used in this dissertation are the average Bit Error Rate (BER) and the probability of outage for a given threshold probability of error. Exact and approximate expressions are derived for the average bit error probability as well as for the outage probability while accounting for the effect of multipath fading, multiple access interference and background noise.

The main focus of this dissertation is to analyze the performance of linear diversity schemes operating in generalized gamma fading channels. The generalized gamma fading model is a versatile fading envelope that generalizes many of commonly used statistical models that describe signal fluctuations due to multipath, shadowing, or a mixture of such processes. The traditional linear diversity combining techniques such as maximal ratio combining (MRC), equal gain combining (EGC), and selection combining (SC) are addressed with reference to generalized gamma fading environments. For the special case of Nakagami fading, new expressions for outage probability and error-rate performance of linear diversity schemes with arbitrary fading parameters are derived in terms of the Lauricella function. Effects of correlated fading are also studied. Their fruitful application to third generation (3G) wideband code division multiple access (WCDMA) systems, particularly for multiple-input multiple-output (MIMO) and 2D-RAKE receivers are demonstrated. The results exhibit a finite integral representation that can be used for fast and accurate numerical computation. A detail study is also done on multivariate generalized gamma fading environments. Relevant statistical characterization of the sum of independent generalized gamma random variables is derived and expressed in terms of the multivariable Fox's H-function. Since numerical evaluation for the multivariate Fox's H-function is difficult, simpler numerical computations are developed using moment generating function and characteristic function approaches. Since some wireless applications may not have enough space among diversity branches, the statistical characterizations of multivariate correlated generalized gamma fading are relevant in such cases. An investigation on the outage performance for multi-branch selection combining is performed for the correlated multivariate generalized gamma channel. Finally, the dissertation summarizes the main results and explores some directions for further studies.

This dissertation is concerned with studies on the performance aspects of mobile LEO satellite cellular systems. Relevant performance measures studied in the dissertation include new call blocking probability, handoff failure probability, call termination probability, and call dropping probability. The analytical teletraffic models available in the literature are generalized in this work in order to characterize the 3G systems more realistically. The effect of earth rotation on cell residence time is also investigated and a new cell residence time model is developed. The new model proposed uses right-truncated gamma distribution to describe the statistics of residence time in the origination cell; and, the generalized beta distribution is used to model the residence time in subsequent cells. A closed-form expression to determine the premature call termination and call dropping probabilities is proposed. The proposed expression requires only the cumulative distribution function of the call holding time and the first four moments of the cell residence time. In all the above probabilistic considerations, the arrival process is regarded as: (i) Poissonian implying voice-like traffic and (ii) non-Poissonian depicting inhomogeneous mix of voice, data, and video transmissions expected on a trunk traffic. The effect of factors that may affect the signal intensity on the performance of mobile LEO satellite cellular systems is also investigated. In particular, a mathematically tractable expression to approximate the moments of cell residence time in origination cell as well as in subsequent cells is developed based on the received signal power.

This dissertation is concerned with studies on the performance aspects of digital cellular radio systems operating in non-Gaussian multipath fading environments. The multipath fading channel, modeled as a superposition of sinusoidal random vectors, is the main focus of this work. Elementary phase distributions, which cause quadrature components of the composite received vector to be correlated, are studied and relevant envelope distribution for the resulting non-Gaussian quadrature components is investigated. The Student-t distributed random process is chosen to model the quadrature components in an indoor multipath fading channel when the number of sinusoidal random vectors is small. For the correlated bivariate non-Gaussian quadrature components, the exact probability distribution function, corresponding to those elementary phase distributions as well as the Gaussian approximation for resultant envelope are evaluated. The scenario where the elementary envelope is beta distributed is also considered. Spherically invariant random process (SIRP) is also used to model the multipath fading channel. The performance analysis based on the spherically invariant multipath fading channel, is then evaluated. The system performance, specified in terms of outage probability and average error probability, significantly depends on the choice of characteristic probability distribution function of the random process that describes the RF ambient. It is shown that the optimality of the optimum combining and maximal ratio combining schemes in interference-limited environments is still retained under the spherically invariant multipath fading channel model.

The research presented investigates the use of cumulants in conjunction with a spectral estimation technique of the signal subspace to perform the blind separation of statistically independent signals with low signal-to-noise ratios under a narrowband assumption. A new blind source separation (BSS) algorithm is developed that makes use of the generalized eigen analysis of a matrix pencil defined on two similar spatial fourth-order cumulant matrices. The algorithm works in the presence of spatially and/or temporally correlated noise and, unlike most existing higher-order BSS techniques, is based on a spectral estimation technique rather than a closed loop optimization of a contrast function, for which the convergence is often problematic. The dissertation makes several contributions to the area of blind source separation. These include: (1) Development of a robust blind source separation technique that is based on higher-order cumulant based principle component analysis that works at low signal-to-noise ratios in the presence of temporally and/or spatially correlated noise. (2) A novel definition of a spatial fourth-order cumulant matrix suited to blind source separation with non-equal gain and/or directional sensors. (3) The definition of a spatial fourth-order cumulant matrix-pencil using temporal information. (4) The concept of separation power efficiency (SPE) as a measure of the algorithm's performance. Two alternative definitions for the spatial fourth-order cumulant matrix that are found in the literature are also presented and used by the algorithm for comparison. Additionally, the research contributes the concept of wide sense equivalence between matrix-pencils to the field of matrix algebra. The algorithm's performance is verified by computer simulation using realistic digital communications signals in white noise. Random mixing matrices are generated to ensure the algorithm's performance is independent of array geometry. The computer results are promising and show that the algorithm works well down to input signal-to-noise ratios of -6 dB, and using as few as 250 x 103 samples.