Cooperative Relay Transmission In Wireless Communication Systems

Relay technology has attracted much attention in recent years. With relay station (RS) between base station (BS) and mobile station (MS), the transmit power of the BS is saved while the interference to neighboring nodes is also mitigated. Currently, relaying technology has been considered as one of the promising technologies by the3rd Generation Partnership Project (3GPP). In this dissertation, the physical layer techniques related to cooperative relay system are researched, including time division duplex (TDD) reciprocity in amplify-and-forward (AF) relay systems, adaptive modulation technology, cooperative MIMO precoding design, reference signal design and channel estimation, etc. The detailed research contents and the main contributions of this dissertation are as follows:TDD is usually adopted in relay systems to support asymmetric traffic and acquire the needed channel information utilizing channel reciprocity. When relay is in AF mode, reciprocity only exists within each hop. However, because of the operation at the relay, the whole link channel between BS and MS loses its reciprocity. To overcome this problem and avoid large feedback, this dissertation proposes a novel gain matrix design method. When the uplink (downlink) gain matrix design method is fixed, the downlink one can be designed specifically to maintain the whole link reciprocity. Simulation results show that the proposed method can keep whole link reciprocity without performance loss.To improve the system throughput and cut the signaling overhead, this dissertation proposes a subchannel and modulation joint selection scheme. Regarded as an equivalent channel, the channel between source station (SS) and destination station (DS) is decomposed into multiple parallel independent subchannels by singular value decomposition (SVD). The number of the used subchannels and their modulation schemes are selected and feedback jointly in the proposed algorithm. Compared with the traditional method where each subchannel selects modulation scheme independently, the novel selection scheme can achieve higher throughput and use less signaling bits. Based on the same system model, another sub-channel and modulation joint selection scheme is proposed to reduce the total transmit power of the SS and RS, with given data rate and bit error rate (BER) target.In naive amplify and forward (NAF) relay systems, the gain factor is innovatively expressed as the function of beamforming vector in this dissertation, and a iterative end-to-end beamforming design algorithm is proposed to find locally optimal solution. Compared with the algorithm where considered the gain factor as a constant, the proposed iterative algorithm can achieve better end-to-end signal to noise ratio (SNR) performance.In mobile relay systems, idle users can form clusters of relay nodes to help the nearby active users. To reduce the complexity of the processing in the relay nodes and destination nodes, a novel transmit preprocessing method is proposed for this cluster based multi-user relay system. Block diagonalization (BD) algorithm is adopted twice to remove inter-cluster and intra-cluster interference simultaneously, and the relay nodes are just in simple AF mode. Compared with the conventional signal processing methods, the proposed method can achieve better tradeoff between sum capacity performance and processing complexity in the relay nodes.In order to acquire the whole link channel information and first hop channel information simultaneously, this dissertation proposes a novel reference signal design method. In this method, the relay node just needs to process the received signal in radio frequency (RF). The operation is simple and the processing delay is small.In imperfect frame synchronization two-hop relay orthogonal frequency division multiplexing (OFDM) systems, when cyclic prefix (CP) is removed in advance, the effect of the removal of the CP on the linear interpolation in amplification area and phase area (A-θ interpolation) is studied and compared with linear interpolation in in-phase component and quadrature component (I-Q interpolation), and then a universal expression of mean square error (MSE) about A-θ interpolation and I-Q interpolation is given. Analysis show that the A-θ interpolation method mainly depends on the frequency correlation of the channels, and it is less affected by the preact of the CP removal, while I-Q interpolation is affected by both frequency correlation of the channels and the preact of the CP removal. Thus, A-θ interpolation can also be selected as interpolation algorithm when the performance of synchronization is poor.