Multiple Hops Relay Beam Forming And Power Allocation Technology Research

MIMO (Multiple Inputs Multiple Outputs) relay transmission has become a keytechnology because of the promising capacity in achieving reliable communication andwide-ranging coverage for the next generation of wireless communication. However,the performance of MIMO relay system mainly depends on the signal processing ateach node and beamforming as an effective solution for achieving highpower-efficiency can be utilized in the MIMO relay system. Considering MIMO relaysystem, this paper focuses on the beamforming and power-allocation technologies andderives relaying schemes to meet the performance requirements by using theoptimization theory.Firstly, the existing optimal relaying scheme in single-user multi-hop relay systemis introduced. By using optimization theory, a general framework of optimizing sourceand relay matrices under the power constraints is given, and the power allocationscheme is similar to the water-filling method. Then, numerical simulations are carriedout to compare the performance between the previously described algorithm (objectivefunction-maximizing Mutual Information) and an algorithm called EEAS (End-to-EndAntenna Selection). Finally, a beamforming and power-allocation schemes are finallyintroduced to satisfied QoS requirement in the single-user MIMO relay system.The above researches only focus on relay system with a single user. In practicalsystems, however, a relay must support multiple users. This paper next studies thebeamforming and power-allocation technologies in multiuser MIMO relay system.Two relaying schemes are proposed under the criterion of maximizing the sum-rate.First, an iterative scheme which iteratively searches the optimal source and relaymatrices by deriving the partial derivatives of the sum-rate and applying the gradientsearch algorithm is proposed. Next, in order to reduce the computational complexity,an alternating algorithm utilizing the equivalent channel method is developed. Thismethod also resorts to a so called maximum-SLNR (Signal to Leakage-and-NoiseRatio) that can suppress the CCI (Co-Channel Interference) and noise at the userseffectively. Monte Carlo simulations illustrate the performance of the both schemes.Performance optimization with the constrained power resources is the focus of the above works. However, in certain wireless network with QoS requirement, satisfyingthe predefined performance measure is more important. Motivated by this problem,this paper also studies the optimization problem of maximizing the sum-rate under thepredetermined QoS-measured by SIR (Signal to Interference Ratio) of each user. Thederived algorithm successively optimizes the source matrix or the relay matrix in turnwith the other being fixed. Theoretical analysis shows that the source matrix has itsclose-form expression and the complexity to design it is low. Finally, simulation resultsdemonstrate that the proposed algorithm can satisfy the individual SIR-targets and hasmuch better sum-rate performance in the high power regime.