Resource Allocation Technique In Multi-cell Cooperation System And Device-to-Device Cellular Networks

With the rapid increasing demand of spectrum resources in wireless communication,enhancing the spectrum efficientcy is a key way to improve the performance of wireless communication systems due to the scarcity of spectrum resources.The multi-cell cooperation and device-to-device(D2D)communication techniques are two important ways that can fully ultilize the spectrum resources in wireless communication systems.The multi-cell cooperation technique can combat the effects of inter-cell interference(ICI),and eliminate the spatial isolation of the cells that share the same frequency to improve the spectrum efficientcy.By introducing the direct link between the devices,D2 D communication technique can let the devices share the frequency resources of the cell at the cost of low interferences,which can improve the frequency resue in time and space dimensionalities and enhance the system capacity significantly.However,in both multi-cell systems and D2 D networks,there are a lot of resources that need to be allocated,such as the power resources and spectrum resources.The proper resource allocation is the key to guarantee the system performances.Aiming at impropving the energy efficiency(EE)or spectral efficiency(SE),this dissertation studies the resource allocation problems for different situations in multi-cell systems and D2 D networks.Firstly,the power allocation(PA)and feedback bit allocation(FBA)problem for a limited feedback coordinated multi-cell downlink transmission system over composite fading channel is addressed,and an energy-efficient joint PA and FBA algorithm is proposed.Comparing with the small scale fading channel or large sacle fading channel,the composite fading channel can present the real wireless channel better.In this thesis,the average sum rate is replaced by a modified performance measure referred to as the virtual sum rate to facilitate the analysis.The approximate closed form of the lower bound on the virtual sum rate is derived over the composite fading channel.The influences of imperfect factors such as quantization error,feedback delay,and estimation error on the channel information are considered.Based on the approximate closed form of the lower bound on the virtual sum rate,a joint PA and FBA algorithm is proposed to maximize the EE.Specially,an alternating iterative algorithm is applied to solve PA sub-problem and FBA sub-problem iteratively.The FBA sub-problem can be transformed into the geometric programming problem(GPP).An iterative algorithm based on the Dinkelbach method is proposed to solve the PA sub-problem.Numerical results show that the proposed strategy can converge in a few iterations and generally yield better EE performance than the conventional FBA schemes.Secondly,the problem of power allocation and mode selection in a hybrid coordinated multi-cell downlink transmission system with capacity limited backhaul capacity is studied,and an energy-efficient power allocation and mode selection algorithm is proposed.Compared with Interference Channel(IC)mode and multi-cell Multiple-input Multiple-output(MIMO)mode,the hybrid coordinated multi-cell system has these two modes,and can achieve better performance by proper power allocation and mode selection.In the thesis,the joint power allocation and mode selection problem maximizing EE is modeled.The joint power allocation and mode selection problem can be approximately transformed into two sub-problems,which have the one-way relationship.Moreover,we prove that the mode selection has an influence on the power allocation,but the power allocation does not affect the mode selection.Furthermore,an energy efficient power allocation scheme with closed form expressions is proposed,which can significantly reduce the implementation complexity.An energy-efficient mode selection metric is proposed.Based on the proposed metric,the optimal mode selection strategy maximizing EE is obtained.Numerical results show that our hybrid multi-cell scheme outperforms the traditional single cooperation mode scheme in terms of EE.Thirdly,the channel access and power control problem for the uplink transmission of D2 D underlaid cellular networks is addressed.A novel base station(BS)-assisted,device-decided power and channel access control scheme is proposed.The proposed scheme can control the interference from D2 D users to the cellular user accurately and guarantee the performances of both the cellular link and D2 D links.Moreover,the proposed scheme is semi-distributed and has low information feedback overhead.For the proposed method,we apply the stochastic geometry tool and derive analytic expressions including the coverage probabilities of both the cellular link and D2 D links.To maximize the area spectral efficiency,the D2 D user access problem is formulated as a multi-criteria optimization problem.The weighted sum method is applied to convert the multi-criteria optimization with high complexity into a simpler single objective optimization problem,and the closed form expression of optimal access threshold is obtained.Different from the existing works,the performances of the cellular link and D2 D links are both considered in obtaining the optimal access threshold.Simulation results show that the proposed method is efficient to mitigate intra-tier and cross-tier interferences caused by the underlaid scenario,and can improve the area spectral efficiency significantly when high data rate and high quality are required in communication links.Finally,for the uplink transmission of D2 D underlaid heterogeneous cellular networks,the channel access problem for D2 D users and the power control problem for both femto cellular users and D2 D users are considered.By applying the stochastic geometry tool,the analytic expressions including the coverage probabilities of macro cellular link,femto cellular links,and D2 D links are derived.Moreover,the macro BS-assisted semi-distributed channel access and power control scheme is proposed.The proposed power control scheme not only can control the interferences from femto cellular users and D2 D users to the macro cellular user accurately,but also can further mitigate the interferences while meeting femto cellular users/ D2 D users performance constraints.According to the proposed scheme,the closed form expression of the optimal D2 D access threshold maximizing the area spectral efficiency is obtained,which considers the performances of the macro cellular link,femto cellular links and D2 D links.Simulation results show that the proposed method is efficient to mitigate intra-tier and cross-tier interferences caused in the D2 D underlaid heterogeneous cellular networks,and can improve the area spectral efficiency significantly compared to existing schemes.