Abstract

Wireless mobile communications are growing in an exponential manner, especially in terms of the number of users. Also, the demand for high Quality of Service (QoS) has become essential. Nowadays, subscribers are using more applications such as the internet, video conferencing, and high quality TV. These applications require high data rates. The Space Division Multiple Access (SDMA) is the key element that can enable reusing of the
same channels among different users in the same cell to meet this demand. For the application of SDMA in an efficient way, it is required to identify the users’ positions and directions in the cell. The Direction of Arrival (DOA) algorithms can estimate
the incident angles of all the received signals impinging on the array antenna. These algorithms give the DOAs of all relevant signals of the user sources and interference sources. However, they are not capable of distinguishing and identifying which one is the direction of the desired user. In this thesis, we have proposed to use a Reference Signal (RFS) known by the transmitter and the receiver to identify which one of the estimated DOAs is the DOA of the desired user in the cell. Using a RFS and applying the correlation concept, we can distinguish the desired signal from the others. Moreover, we have considered the Affine Projection Algorithm (APA) to enhance the accuracy of the estimated direction and to form a beam towards the desired user and nulls towards the interferers. Our simulation results assure that, in the presence of the RFS, the DOA algorithms can identify the direction of the desired user with high accuracy and resolution. We have investigated this concept on different DOA algorithms such as MUltiple Signal Classification (MUSIC), ROOT MUSIC, and Estimate the direction of arrival of Signals Parameters via Rotational Invariance Technique (ESPRIT) algorithms.
Moreover , we have introduced an approach for using the smart antennas (SA) to exploit the space diversity for the next generations of mobile communication systems. We have applied
a combination of the MUSIC and the Least Mean Squares (LMS) algorithms. We have proposed the MUSIC algorithm for finding the directions of the users in the cell. In addition, we have considered the LMS algorithm for enhancing the accuracy of the DOA, performing the beam generation process, and keeping track of the users in the cell. Furthermore, we have proposed a scheduling algorithm that performs the scheduling in terms of the generated beams.
The space diversity, together with the time and frequency diversities of LTE (Long Term Evolution) results in a large capacity increase in the next generations of wireless mobile communication systems. Simulation results show that the proposed algorithm called MUltiple
Signal Classification and Least Mean Squares (MLMS), has the capability to converge and completely follow the desired user signal with a very high resolution. The convergence and the accurate tracking of the desired signal user take place after 13 iterations while in the traditional LMS, the convergence needs 85 iterations to take place. This means an 84.7% improvement over the traditional LMS algorithm for the same number of calculations in each iteration. In contrast to the traditional LMS algorithm, the proposed algorithm can work in the presence of high level of interference. Furthermore, the proposed scheduling scheme based on beamforming shows a gain of 15% in the total aggregated throughput for each 10o decrease in the beam size. The proposed model provides an optimum, complete, and practical design for the next generations of the mobile communication systems. In this model, we have proposed a mechanism to find the direction of each user in the cell, enhance the accuracy of the obtained DOAs, and perform scheduling based on the generated beams. In addition, we have presented an approach for Frequency Reuse (FR) based on beamforming for 5G. We have implemented a synthesizer in order to smartly form the desired beam shape and make the nulls deeper. We have taken the advantage of the SAs, beamforming capabilities, and the radiation pattern (RP) synthesizing techniques to build up a FR scheme for 5G. Also, we have developed a formula for calculating the Signal to Interference and Noise Ratio (SINR) in terms of the desired and the interferers directions. The objective is to maintain the SINR at the minimum acceptable levels required by the LTE while reducing the beam sizes, and hence increase the FR factor. The simulation results show that with a Uniform Linear Antenna (ULA) of 11 elements, we can achieve the desirable SINR levels using beams of 100 width, which improves the FR factor from 1 to 18, and subsequently increases the number of mobile users.