The Relay In The Cellular Communication System And Its Performance Investigation

Cellular network is used to cover different geographical regions. On one hand, they should cover high-density buildings in urban areas with indoor users, on the other hand cellular network should provide wireless access for the rural in a large geographic area. In both cases, to cover the whole service area is challenging. Due to either the shadow or too long of the radio connection from base station, the radio propagation faces more challenge. In order to cope with different radio propagation conditions, multi-hop communications was proposed. Through intermediate nodes, such as relay/repeaters, radio links are divided into two or more hops, which is better than the direct connection, which increases the link quality improves the throughput and cell coverage.In this thesis, the MIMO relay system is first studied with co-located case and distributed case in the context of the relay gain. In the co-located case, all the antennas are located in the same relay mode, while in the distributed case, the antennas can be located in different relay nodes, which can have co-operation. The performance with different relay positions, the performance with different number of relay antennas, and the performance of the antenna with different correlation is investigated. Both decode-and-forward relay and amplified-and-forward relay are considered. When the relay nears the base station, the capacity is limited by the second hop, and there is no large relay gain difference between the co-located case and distributed case. When the relay nears the mobile station, the capacity is limited by the first hop, while the capacity of first hop is related to the number antennas in each relay. In this case, the co-located case has better performance. In most of the case, the relay gain for the co-located case is larger than the gain in the distributed case.The filter matrix design for MIMO AF relay is further investigated in the co-located case and the distributed case. According to channel state information of the first and second hop, the blind relay, the match filter relay and the SVD relay are considered. Further a spatial filter for MIMO repeater is designed based on first-hop CSI and end-to-end SNR. The proposed filter is a product of a diagonal matrix and the conjugate transpose of the left singular matrix of the first-hop channel matrix. The diagonal matrix is obtained by maximizing an upper bound of the ergodic channel capacity. Simulation results indicate that the proposed scheme outperforms the scheme without CSI, especially when the number of repeater antennas is greater than that of the base station antennas. The introduction of such a repeater may not need modifications on the base station nor the mobile station. Hence the transparency property of a repeater can be preserved.For multi-relay case, the thesis investigates the relay selection performance. The relay selection algorithm based on the average capacity (ergodic capacity) or the outage capacity is proposed. The average capacity (ergodic capacity) or outage capacity can be obtained based on the average channel information with Taylor series. For the relay system, the relay can easily obtain the instantaneous channel state information in the first hop and the average channel state information on the second-hop. Here relay selection based on the average channel state information (or the second-order statistics of the channel) in the second hop is investigated. The case for the multi-relay with one user and the case with multi-relay and more users are considered. The simulation results shows that relay selection based on the average channel state information (or the second-order statistics) of the second hop is a simple and accurate method.Finally, the performance of the two-hop decode-and-forward (DF) relay and radio frequency (RF) repeater in the cellular system are compared. In the cellular system, there is inter-cell interference, which is more complicated than the single link, leading that the conclusions of a single link may not be applicable to the cellular system. This thesis focuses on the performance of relay and repeaters in the cellular system. Here the decode-and-forward relay is mainly L2/L3 relay. The current relay can not receive and transmit at the same time with the same resource, and the relay capacity gain mainly comes from frequency multiplexing or spatial diversity. In the cellular system, the RF repeaters can receive and transmit at the same time with the same resource, but it brings more interference since the RF repeater amplifies the inter-cell interference and noise when it amplifies the signal. For the cells with small radius, the RF repeater can bring more gain than the DF relay. For the cells with large radius, the DF relay can bring more gain than the RF repeater. The deployment of the multiple relay/repeater in the cellular system is also proposed.