Resource allocation and feedback in wireless multiuser networks

THIS thesis focuses on the design of algorithms for resource allocation and feedback in wireless multiuser and heterogeneous networks. In particular, three key design challenges expected to have a major impact on future wireless networks are considered: cross-layer scheduling; structured quantization codebook design for multiuser multipleinput multiple-output (MU-MIMO) networks with limited feedback; and resource allocation to provide physical layer security. The first design challenge is cross-layer scheduling, where policies are proposed for two network architectures: user scheduling in single-cell multiuser networks aided by a relay; and base station (BS) scheduling in coordinated multipoint (CoMP). These scheduling policies are then analyzed to guarantee satisfaction of three performance metrics: symbol error probability (SEP); packet delay; and packet loss probability (PLP) due to buffer overflow. The concept of the τ-achievable PLP region is also introduced to explicitly describe the tradeoff in PLP between different users. The second design challenge is structured quantization codebook design in wireless networks with limited feedback, for both MU-MIMO and CoMP. In the MU-MIMO network, two codebook constructions are proposed, which are based on structured transformations of a base codebook. In particular, the first construction is based on the Householder transform, and the second construction is based on group representation theory. Both constructions are shown to reduce sum-rate loss due to rank-deficient quantized channel matrices when zero-forcing precoding is employed. A rate-outage lower bound is also derived and used to construct optimal base codebooks, which are shown to have the structure of an important class of equiangular tight frames. In the CoMP network, a low-complexity construction is proposed to solve the problem of variable codebook di-