Research On Adaptive Radio Resource Allocation In Ofdm Systems

OFDM (orthogonal frequency-division multiplexing)/OFDMA (orthogonal frequency-division multiple access),recognized as one of the most advantage modulation/multiplex techniques, can provide a high performance physical layer. In the well known4G wireless systems, OFDM is regarded as the fundamental technique affecting the network deployment. Benefiting from the multi-carrier transmission scheme and the orthogonality among subcarriers, the wireless resource allocation in OFDM systems is more flexible and efficient compared to the traditional solutions. Due to the limitation happened when employing the existing resource allocation algorithms in a real OFDM wireless system, practical resource allocation in OFDM wireless systems considering implementation issue and user requirement is investigated in this thesis. Furthermore, as a result of ever-increasing demand for green radio design, energy efficiency enhancement in OFDM systems is also investigated. Overall, three main aspects are covered:1) A low-complexity suboptimal resource allocation algorithm with discrete rate sets constraint.According to the effect of MCS (Modulation and Coding Schemes) on data loading in a real OFDM wireless system, a low-complexity suboptimal resource allocation algorithm with discrete rate sets constraint is developed assuming that MCS determination policy is executed based on SINR (Signal to Interference plus Noise Ratio) measurement. The proposed algorithm can not only reach a high system throughput/spectral efficiency, but also fulfill user minimum rate requirements. Comparing to the iterative-based algorithm in literature, the proposed algorithm is less complex via two-step subcarrier/subchannel allocation. Furthermore, the proposed algorithm is more flexible to be implemented in a real system via MCS related consideration.2) Spectral-efficient resource allocation in OFDM wireless systems with subcarrier grouping constraint.According to the implementation issue of MCS in a real OFDM wireless system, i.e., the subcarriers assigned to one user always use the same MCS, low-complexity suboptimal resource allocation schemes are proposed with subcarrier grouping constraint, including:a downlink resource allocation algorithm considering user rate allowed region, a downlink resource allocation algorithm considering proportional rate constraints, and a low-complexity resource allocation algorithm which carry out subcarrier assignment and power allocation jointly. The proposed algorithms fulfill the requirements of a real OFDM wireless system, reduce the complexity of resource allocation algorithms implementation, and also simplify related device implementation of both the base station and user side. Furthermore, system faireness and various user rate requirements are satisfied, and the system throughput/spectral efficiency is enhanced as well.3) Energy-efficient resource allocation in OFDM wireless systems with system faireness considerationAccording to system fairness requirement in a best-effort-service-based scenario, an energy-efficient resource allocation algorithm in OFDM wireless system with proportional rates constraint is proposed. Users are ensured achieving rate levels which equal to a predetermined proportion of the overall system throughput, and meanwhile the system energy efficiency is enhanced. The proposed algorithm includes three steps:energy-efficient transmit power estimation under assumption of flat fading and subcarrier sharing, max-min-based suboptimal spectral-efficient subcarrier assignment, and bisection-based optimal energy-efficient power allocation. Through the above steps, the complexity of the proposed algorithm is linearly increasing with respect to the number of users and that of subcarriers, compared to exponentially increasing with the number of subcarriers for optimal solution.In this thesis, adaptive resource allocation in OFDM systems is investigated based on the above3aspects. According to the requirement of resource allocation in a real OFDM system, the algorithms are proposed to reduce computational complexity and inhance implementation flexibility. Furthermore, according to system fairness requirement in an energy-efficient OFDM system, the algorithm is proposed to make energy-efficient resource allocation more reasonable.