الفهرس 
WIRELESS COMMUNICATIONS OVERVIEWOnly 14 pages are availabe for public view
 |  | from | 174 |  |  |
| | | from | 174 | | |
Abstract
The explosive growth of wireless communication created a demand for high speed, reliable, and spectrally efficient wireless communications. Signal processing is a key research area in wireless communication systems. The recent years have known a real explosion in research addressing different aspects of wireless communications signal processing. This area is continuously expanding with emerging applications and services such as interactive multimedia and Internet. Signal processing has to meet the new challenges presented to future wireless communication systems such as very low bit error rates, very high transmission rates, very high spectral efficiency, real-time multimedia access, and differential Quality of Service (QoS).
In high-speed data transmissions, the channel is severely frequency-selective due to the presence of many interfering paths with different time delays. A promising wireless transmission technique that can overcome the channel frequency-selectivity and even take advantage of this selectivity to improve the transmission performance is Multi-Channel (MC) transmission, as in Orthogonal Frequency Division Multiplexing (OFDM) and Multi-Input Multi-Output (MIMO). The combination of the two MC transmission techniques, OFDM and MIMO, benefits from the superior advantages of them to support very high-speed communications and has incorporated in the recent Fourth Generation (4G) wireless cellular and wireless broadband communication systems.
In the conventional MIMO-OFDM transmission, all sub-channels employ the same signal constellation size and Uniform Power Allocation (UPA). However, the overall error probability is dominated by the sub-channels with the worst performance. To improve the MIMO-OFDM transmission performance and get a higher throughput and spectral efficiency, adaptive power allocation algorithms can be employed, where the sub-channels signal constellation size and power allocation vary according to their measured Signal to Noise Ratio (SNR) values. Recently, design of efficient power allocation algorithms for MC transmission, which is the main topic of this thesis, has attracted much attention due to its advantage of higher transmission capacity and spectral efficiency.
Firstly, the thesis presents efficient and low complexity power allocation algorithms for single-user OFDM systems, in which the multi-path and frequency diversities are exploited by combining the Maximum Ratio Combining Pre-coding (Pre-MRC) and an
V
optimal Greedy Power Allocation (GPA) algorithm to maximize the throughput and spectral efficiency of the Single-User (SU) OFDM system. The proposed optimal Pre-coded GPA (Pre-GPA) algorithm suffers from high computational complexity, so low complexity GPA-based algorithms are proposed to simplify the complexity of the optimal GPA algorithm by grouping the sub-channels into Quadrature Amplitude Modulation (QAM) levels and applying the GPA algorithm per-QAM group. The excess power of each QAM group is exploited by GPA-based algorithms to more improve the throughput performance of the SU OFDM system.
Secondly, the spatial dimension is introduced for improving the throughput and spectral efficiency of the SU OFDM system by incorporating multiple antennas at each end of the communications link. The two signal processing techniques of the MIMO transmission, which are spatial diversity and spatial multiplexing, are considered. Three proposed spatial power allocation algorithms are developed for the SU MIMO-OFDM systems. In the first proposed spatial power allocation algorithms, the low-complexity Pre-GPA algorithms, suggested for the SU OFDM systems, are developed for the SU MIMO systems. The sub-carriers of the SU OFDM system are replaced by spatial sub-channels produced by applying the Singular Value Decomposition (SVD) on the MIMO channel matrix. The simulation results show that the proposed low-complexity spatial Pre-GPA algorithms outperform the traditional spatial GPA algorithms with a low computational complexity. In the second proposed spatial power allocation algorithms, the spatial and frequency diversities are exploited by adaptively allocating the system sub-carriers to the available spatial sub-channels followed by Per-Spatial GPA (PSGPA). The proposed algorithms exhibit more system throughput and spectral efficiency than traditional spatial diversity, in which the sub-carriers are fixed allocated per-spatial sub-channels. In the third and last proposed power allocation algorithms, the spatial multiplexing-based MIMO is considered, in which the spectral efficiency of the system is linearly increased by the minimum number of the MIMO system antennas. An optimal two-dimensional Spatial-Frequency GPA (SFGPA) algorithm is proposed to efficiently improve the average system spectral efficiency. The high computational complexity of the optimal SFGPA solution is simplified by proposing a low-complexity near-optimal one-dimensional PSGPA algorithm with moving the per-spatial excess power downwards to enhance the spectral efficiency of the spatially multiplexing-based SU MIMO-OFDM systems.
VI
Thirdly, the multi-user dimension is introduced to exploit the multi-user diversity, in which the probability of having a sub-channel suffers deep fading for all activated users is very low. In Multi-User OFDM (MU OFDM), also called Orthogonal Frequency Division Multiple Access (OFDMA), the available sub-carriers must be efficiently allocated to the activate users before the power is allocated. In this thesis, an efficient and low complexity GPA algorithm for MU OFDM system with user rates proportionality guarantee is proposed to satisfy improved system spectral efficiency and acceptable users’ rates fairness. The proposed proportional rate adaptive GPA algorithm is compared with a related published algorithm, where the proposed algorithm outperforms the published one with the same computational complexity.
Finally, three-dimensional (Spatial, Frequency, and Multi-User) GPA algorithms for the MU MIMO-OFDM systems are proposed. The proposed algorithms start with spatial sub-carrier allocation followed by optimal SFGPA. The proposed three-dimensional GPA algorithms exploit the spatial dimension to efficiently allocate sub-carriers to the active users, i.e. the sub-carriers allocation is performed in two-dimensional scenario (spatial and multi-user), and thereby it exploits both of the spatial and multi-user diversities. The SFGPA step of the proposed algorithms exploits both the spatial and frequency diversities. So, spatial, frequency, and multi-user diversity are exploited by the proposed three-dimensional GPA algorithm to more maximize the throughput and spectral efficiency of the MU MIMO-OFDM systems.