Research On Physical-Layer Multicast Beamforming

The mobile data traffic has been growing rapidly.Meanwhile,the types of wireless services are also experiencing a fundamental shift from the traditional end-to-end communications to content delivery.Physical-layer multicasting is a promising solution to simultaneously deliver a common message to multiple receivers,which has received increasing interest from academia and industry.Physical-layer multicasting via beamforming can further boost the system performance by exploiting the channel state information at the transmitter.It can properly design an individual beamforming vector for each signal to improve the signal-to-interference-plus-noise ratio at the receiver side,thus to improve the spectrum and energy efficiencies.However,the existing research on multicast beamforming is far from mature comparing with the traditional unicast beamforming.Specifically,multicast beamforming problem is NP-hard in general.Most of the existing algorithms suffer from high computational complexity and are not scalable in large-scale wireless systems.In the current 3GPP evolved multimedia broadcast multicast service(e MBMS)specification,single-frequency network(SFN)transmission is adopted for multicast services,which requires dedicated time/frequency resources that are orthogonal to unicast services.However,such scheme with SFN transmission and orthogonal resource sharing suffers from low spectrum efficiency and poor flexibility.In addition,as a common physical-layer technology,multicast transmission is currently limited to multimedia broadcast and multicast services in traditional cellular networks.Its great potentials under the emerging wireless network architectures,such as wireless caching networks and self-backhauled wireless networks,have not yet been fully explored.This thesis aims to develop efficient and effective multicast beamforming algorithms to improve the network performance,and explore multicast potentials and applications in the emerging wireless network architectures.The main contributions are as follows.1.In order to address the issue that the existing multigroup multicast beamforming algorithms suffer from high computational complexity and are not scalable in largescale wireless systems,we develop a fast algorithm with high-performance and low-complexity for multigroup multicast beamforming by utilizing convex-concave proce-dure(CCP)and alternating direction method of multipliers(ADMM).By exploiting the specific structure of the considered problem and introducing a few auxiliary variables,the large-scale problem in each iteration of the proposed algorithm is then decomposed into multiple small-size subproblems,which can be solved in parallel with closed-form solutions.Numerical results show that the proposed fast algorithm maintains the same favorable performance as the state-of-the-art algorithms but reduces the simulation running time by one to two orders of magnitude,which is very suitable for large-scale wireless systems.2.Due to the fact that the transmission of multicast services in the current e MBMS specification requires dedicated time/frequency resources,which suffers from low spectrum efficiency and poor flexibility,we propose a non-orthogonal multicast and unicast transmission framework based on layered-division multiplexing(LDM)to efficiently support hybrid multicast and unicast services in cooperative multi-cell cellular networks with limited backhaul capacity.Under this transmission framework,we formulate a joint multicast and unicast beamforming problem with adaptive base station(BS)clustering that aims to maximize the weighted sum of the multicast rate and the unicast rate under per-BS power and backhaul constraints.We develop an algorithm to find its global optimum.We also reformulate the problem as a sparse beamforming problem and propose a low-complexity algorithm to find a near-optimal solution.Simulation results demonstrate the significant superiority of the proposed LDM-based non-orthogonal scheme over the orthogonal scheme in terms of the achievable multicast-unicast rate region.3.We present a content-centric transmission mechanism in the cache-enabled cloud radio access network architecture by incorporating multicasting and caching,which can effectively avoid redundant content delivery in the network,thus significantly reduce the network cost.We formulate a joint content-centric BS clustering and multicast beamforming problem with fixed caching to minimize the weighted sum of backhaul cost and transmit power cost.We develop an efficient algorithm to obtain an approximate solution that is very close to the optimum.Three heuristic caching strategies are proposed and compared through comprehensive simulations under different network settings.The results provide some useful insights into the design of more advanced caching strategies.The superiority of the proposed content-centric transmission mechanism is also demonstrated.4.We propose a user-centric joint access-backhaul transmission framework for full-duplex self-backhauled wireless networks by leveraging multicasting.User-centric clustering is adopted in the access link to mitigate the inter-cell interference,while multicast transmission is adopted in the backhaul link to reduce the traffic load.We first study the joint access-backhaul beamforming and BS clustering design problem to maximize the network throughput when global channel state information(CSI)is available.We develop an efficient algorithm to solve it.We then extend the study to the stochastic joint access-backhaul beamforming with partial CSI.Simulation results demonstrate the effectiveness of the proposed algorithms for both full CSI and partial CSI scenarios.The results also show that the transmission design with partial CSI can greatly reduce the CSI overhead with little performance degradation.