Resource Allocation Algorithm Designs for OFDM-based communication Systems

• Main Authors: Chun-Ye Lin, 林春羽
• Other Authors: Yin-Yi Lin
• Format: Others
• Language: en_US
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/96112579852987315291

博士 === 國立中央大學 === 電機工程研究所 === 99 === In order to solve the objectives with some constraints in various scenarios for orthogonal frequency division multiplexing (OFDM) based systems, we proposed a couple of adaptive resource allocation algorithms. In downlink multi-user OFDM systems, we utilize the Lagrange multiplier method to propose an approach which uses the related calculation results in the procedure for the power and bit assignment. When the model of the co-channel interference channel in downlink multi-cell OFDM systems or in discrete multi-tone (DMT)-based wired digital subscriber loop (DSL) modems is considered, we developed an iterative interlaced bit and power allocation to minimize the total transmit power of all base stations (BSs) while satisfying the required quality of service (QoS) of each user in the cell. The proposed scheme offers the better power gain for higher required data rates than the existing algorithm while having a less computational complexity. We also propose a sub-carrier, bit, and power allocation algorithm for multiple input multiple output (MIMO) OFDM-based multi-user communication systems on downlink. We exploit a dynamic multi-user MIMO OFDM transmit scheme with two-dimensional (2-D) coder beam-formers in the transmitter to achieve our purpose. We fairly compare the proposed method with several algorithms, such as frequency division multiple access (FDMA), constructive initial assignment (CIA) and Wong’s subcarrier allocation (WSA), under the same conditions. Finally, a resource allocation scheme for multi-access MIMO-OFDM systems in uplink is developed to improve power and spectrum efficiency in the frequency and the space domains. We develop the scheme to remedy some problems for multi-access MIMO-OFDM systems without using multi-user detector (MUD) techniques and the assumption of the identical spatial cross-correlation across all sub-carriers for any pair of spatially separable users.