| MIMO, which means Multi-Input-Multi-Output, efficiently improvessystem’s capacity and spectral efficiency without improving transmit-power andspectral bandwidth, at the same time, and brings complexing gain and diversity gain.Relay, which can extend communication distance, combat shadowing, is introducedinto the fourth generation mobile communication. Among the relays,amplify-to-forward(AF) relay is more simple and with smaller processing delay sincethe relays just amplify and forward the received signals. It can further improveperformance of system by equipping multiple antennas at the relay node. Femtocell,as the key technique to improve indoor coverage, achieves extensive attention. Afemtocell is a low-power and self-deployed base station, Which improves signalquality and extends battery time of terminal for users, saves their operation-cost andmaintenance-cost for the operator. Multi-user system can bring system’s gain, butcan iccur interference between users. Precoding(Beamforming) and power-control,as techniques to suppress interference, is widely studied. Therefore, we investigatethe precoding in multi-user MIMO relay system with imperfect and perfect channelinformation station and joint beamforming in macro-femto networks.First, this paper proposes a joint transceiver design scheme based on theminimum mean squared error (MMSE) criterion, where the direct link is considered.The proposed scheme employs an iterative algorithm which alternately optimizes therelay precoding and user decoder. Numerical results shows that, our proposedscheme outperforms the existing schemes, e.g., MMSE without direct link andLMMSE schemes in terms of MSE and BER.Then, in practice, the transmitter is difficult to get perfect channel stateinformation, so there is always channel estimation error. Therefore, A joint precodingdesign scheme is proposed for non-regenerative Multi-Input-Multi-Output relaysystem with channel extermination errors which are modeled as Gaussian randomvariables. Based on MMSE(minimum mean squared error) criterion, The proposedscheme employs an iterative algorithm which alternately optimizes the base station precoding and relay precoding. Numerical results shows that, our proposed schemeoutperforms the existing schemes, e.g., joint precoding scheme without channelestimation errors and only relay pre-code scheme with channel estimation errors interms of BER.Beamforming is benefit for interference management, so it is introducedinto macro-femto networks to manage the cross-tier interference. This paper’s objectis the total transmit-power,subjected to the MSE of each macro-user and femto-user.through the theoretical derivation, the problem is transformed into SOCP convexoptimization problem. We put forward iteration algorithm to join the beam formingof macro cell, femtocell and relay. As the simulation result shows, with the user’sMSE threshold increasing, the total transmit-power would be reduced.There are still many problem to be investigated in MU relay system and MUfemtocell networks. In the MU relay system with direct link, there is a need to designrelevant algorithm to choose between direct-link user and relay user.Beside, infemtocell networks, cognitived radio technique should be introduced to manageinterference. |
