Research On Resource Allocation In MIMO-OFDM Downlink Systems

In the future of wideband wireless communication systems, there are two severe challenges: multi-path fading and frequency efficiency. Orthogonal Frequency Division Multiplexing (OFDM) make the high speed serial data streams into low speed parallel data streams, and transmit them using the orthogonal subcarriers, because the sub-channels are narrow-band, it is considered that the sub-channels are flat, hence, it can reduce the impact of the multi-path fading. Multiple-Input-Multiple-Output (MIMO) can generate the independent parallel sub-channels and transmit multiple data streams, so the MIMO can increase the capacity of the system. That is the space multiplexing provided by MIMO can promote the frequency efficiency not need to increase the band. Therefore, there are two great results if we combine the OFDM and MIMO. Firstly, the transmission rate is increased. Secondly, the transmission reliable is promoted. Assuming knowledge of channel state information (CSI) is available at the transmitter, the performance can be further improved through the adaptive resource allocation. The resource allocation algorithms Based on the zero-forcing beamforming (ZFBF) in the MIMO-OFDM systems is considered in this paper.In the beginning, we introduce some Margin adaptive algorithms for the OFDM systems. Then, the user scheduling algorithms are reviewed, we emphasis on the semi-orthogonal user selection algorithm (SUS). Based on the user schedule algorithms in MIMO systems and the two-step strategy (resource allocation and subcarrier assignment) in OFDM systems, employing the ZFBF, the NUS and S-NUS algorithms are proposed to allocate the resource in MIMO-OFDM systems. The proposed two algorithms are employed two-step strategy, the Bandwidth Assignment Based on SNR (BABS) algorithm is used to determine the number of subcarriers for each user firstly. Then, the traversing and SUS algorithms are used to assign the subcarrier and power.It is obviously that the calculation complexity of NUS algorithm is large, because it need to traversing all the user sets in each subcarrier. But the performance of S-NUS algorithm depend on the value of the threshold which measures the orthogonal of the channels for users, Hence, we propose a new algorithm called the Greedy algorithm. The Greedy algorithm select a user through calculating the approximate transmit power of the user set after insert the user, we also give the ending condition for user selecting. The Greedy algorithm is not depend on any threshold value and has the low complexity, is fit for the easy-changing communication environment.The three algorithms proposed in this paper have some advantages and disadvantages itself, them can be used for wireless communication environment with different demand. Besides, all the proposed algorithms have considered the QoS of users, play a important role for the actual communication.