利用智慧型天線於階層式蜂巢狀系統干擾管理技術之研究

Abstract

With the increasing demand for high data rates service, macro-cell and small cell coexist hierarchical cellular architecture become an extremely important issue which can improve signal quality of cell edge users and offer huge spectrum efficiency gain in current and future wireless communication systems. In such hierarchical cellular systems, the spectrum efficiency can be further enhanced by managing the inter-cell interference (ICI) between macro-cells and small cells. In this dissertation, we investigate smart multiple-input multiple-output (MIMO) antenna cooperation techniques for interference control in hierarchical cellular systems.

In the first part, we investigate the performance of hierarchical base station cooperation (HBSC) techniques in macro-cell and small cell coexist heterogeneous networks (HetNet) for the 3rd Generation Partnership Project (3GPP) Long-Term Evolution-Advanced (LTE-A) system. HBSC techniques aim at reducing the co-channel interference between a macro-cell and a group of small cells by coordinating the transmissions of distributed antennas in the cell coverage area. We find that joint intra- and inter-site cooperation will significantly reduce the inter-cell interference. Compared to the case without joint intra-/inter-site coopearation, we demonstrate that HBSC techniques in the conventional pentagonal cell architecture can improve spectrum efficiency by $ 65\% $ at the cell edge. When the narrow beam tri-sector cell architecture is considered, the spectrum efficiency can be further improved by $ 16.42\% $ at cell edge.

In the second part, we investigate different beamforming schemes and consider the channel feedback issue in the HBSC systems. Base station cooperation techniques can enhance the system performance by reducing the ICI, but relying on the accurate channel state information (CSI) of the cooperative base stations in the feedback channels. However, when the CSI is used for calculating the beamforming weights, the performance of a cooperative base station is very sensitive to the channel variations. To overcome this issue, we present the design principles of a robust HBSC system by changing the role of the CSI in the feedback channel from calculating beamforming weights to user selection. Because of different purposes and thus having a much margin to tolerate channel variations, the CSI of feedback channels can accurately select the pairs of the transmit antenna of a base stations and the receive antennas of users. By applying a simple beamforming technique at the receiver, such as zero-forcing (ZF) algorithm, the ICI can be effectively reduced. Based on the above idea, we design a HBSC systems, where the underlaying small cells coexist with a macro cell. We also propose a dynamic small cell selection algorithm to obtain a group of appropriate small cells to cooperate with the corresponding macro cell. Our results show that the proposed receive ZF beamforming techniques can improve the spectrum efficiency of the HBSC system at the cell edge by $ 34.6\% $ compared to the transmit ZF beamforming techniques when the standard deviation of channel errors is two times of the average value of the desired signal strength.

In the third part, we apply the receive beamforming technique in the hierarchical cognitive radio (HCR) systems which aim to serve the primary users in a macrocell and the secondary users in underlaying small cells by using the same spectrum. The HCR system considered in this dissertation is defined as a few time-division-duplex (TDD) microcells on top of a frequency-division-duplex (FDD) macrocell, where the microcell is the secondary systems and the macrocell is the primary system. The major challenge for HCR systems lies in controlling the inter-cell interference from the secondary users to the primary systems, and further enhancing capacity of the secondary users. We first propose an effective user scheduling algorithm in the secondary system to mitigate the interference to the primary system. Furthermore, we design an optimal receive beamforming scheme to maximize the uplink transmission rate of the underlaying secondary users in the HCR systems. Our results show that the proposed integrated scheduling and beamforming technique can improve average spectrum efficiency by $ 82.5\% $ compare to the original primary system.

In summary, the main contribution of this dissertation is to investigate the interference management techniques for hierarchical cellular systems. We investigate three kinds of MIMO antenna cooperation schemes to mitigate interference: (1) HBSC systems with joint intra- and inter-site cooperation; (2) HBSC systems with receive beamforming techniques; and (3) HCR systems with optimal beamforming and scheduling. The proposed framework can help analyse the performance of hierarchical cellular systems and provide important insights into the design of base station cooperation techniques, antenna beamforming schemes, and scheduling algorithm to enhanced spectrum efficiency.

With the increasing demand for high data rates service, macro-cell and small cell coexist hierarchical cellular architecture become an extremely important issue which can improve signal quality of cell edge users and offer huge spectrum efficiency gain in current and future wireless communication systems. In such hierarchical cellular systems, the spectrum efficiency can be further enhanced by managing the inter-cell interference (ICI) between macro-cells and small cells. In this dissertation, we investigate smart multiple-input multiple-output (MIMO) antenna cooperation techniques for interference control in hierarchical cellular systems.

In the first part, we investigate the performance of hierarchical base station cooperation (HBSC) techniques in macro-cell and small cell coexist heterogeneous networks (HetNet) for the 3rd Generation Partnership Project (3GPP) Long-Term Evolution-Advanced (LTE-A) system. HBSC techniques aim at reducing the co-channel interference between a macro-cell and a group of small cells by coordinating the transmissions of distributed antennas in the cell coverage area. We find that joint intra- and inter-site cooperation will significantly reduce the inter-cell interference. Compared to the case without joint intra-/inter-site coopearation, we demonstrate that HBSC techniques in the conventional pentagonal cell architecture can improve spectrum efficiency by $ 65\% $ at the cell edge. When the narrow beam tri-sector cell architecture is considered, the spectrum efficiency can be further improved by $ 16.42\% $ at cell edge.

In the second part, we investigate different beamforming schemes and consider the channel feedback issue in the HBSC systems. Base station cooperation techniques can enhance the system performance by reducing the ICI, but relying on the accurate channel state information (CSI) of the cooperative base stations in the feedback channels. However, when the CSI is used for calculating the beamforming weights, the performance of a cooperative base station is very sensitive to the channel variations. To overcome this issue, we present the design principles of a robust HBSC system by changing the role of the CSI in the feedback channel from calculating beamforming weights to user selection. Because of different purposes and thus having a much margin to tolerate channel variations, the CSI of feedback channels can accurately select the pairs of the transmit antenna of a base stations and the receive antennas of users. By applying a simple beamforming technique at the receiver, such as zero-forcing (ZF) algorithm, the ICI can be effectively reduced. Based on the above idea, we design a HBSC systems, where the underlaying small cells coexist with a macro cell. We also propose a dynamic small cell selection algorithm to obtain a group of appropriate small cells to cooperate with the corresponding macro cell. Our results show that the proposed receive ZF beamforming techniques can improve the spectrum efficiency of the HBSC system at the cell edge by $ 34.6\% $ compared to the transmit ZF beamforming techniques when the standard deviation of channel errors is two times of the average value of the desired signal strength.

In the third part, we apply the receive beamforming technique in the hierarchical cognitive radio (HCR) systems which aim to serve the primary users in a macrocell and the secondary users in underlaying small cells by using the same spectrum. The HCR system considered in this dissertation is defined as a few time-division-duplex (TDD) microcells on top of a frequency-division-duplex (FDD) macrocell, where the microcell is the secondary systems and the macrocell is the primary system. The major challenge for HCR systems lies in controlling the inter-cell interference from the secondary users to the primary systems, and further enhancing capacity of the secondary users. We first propose an effective user scheduling algorithm in the secondary system to mitigate the interference to the primary system. Furthermore, we design an optimal receive beamforming scheme to maximize the uplink transmission rate of the underlaying secondary users in the HCR systems. Our results show that the proposed integrated scheduling and beamforming technique can improve average spectrum efficiency by $ 82.5\% $ compare to the original primary system.

In summary, the main contribution of this dissertation is to investigate the interference management techniques for hierarchical cellular systems. We investigate three kinds of MIMO antenna cooperation schemes to mitigate interference: (1) HBSC systems with joint intra- and inter-site cooperation; (2) HBSC systems with receive beamforming techniques; and (3) HCR systems with optimal beamforming and scheduling. The proposed framework can help analyse the performance of hierarchical cellular systems and provide important insights into the design of base station cooperation techniques, antenna beamforming schemes, and scheduling algorithm to enhanced spectrum efficiency.