Please use this identifier to cite or link to this item:
|Title:||Limited Feedback Scheduling Schemes for MIMO Broadcast Channels|
|Abstract:||Multiple-input multiple-output (MIMO) and MIMO-orthogonal frequency division multiplex- ing (MIMO-OFDM) broadcast channels (BC) are considered in this thesis for multi-user (MU) environment to leverage the advantages of MIMO and multi-user diversity (MUD) simultane- ously. Multiple independent data streams are transmitted to multiple users simultaneously in the same frequency and time slot by the base station (BS) employing spatial division multiple access (SDMA) technique. Efficient scheduling and precoding are required at BS to mitigate the multi-user interference (MUI) at receivers in SDMA. Dirty paper coding (DPC) is the optimal MUI canceller though computationally expensive technique. The channel state information at transmitter (CSIT) is required at BS to select M number of users with the best channel condition out of K number of users, whereM is the total number of transmit antennas at BS. Users send the information about channel conditions using reverse channel bandwidth which reduces the spectral efficiency. The feedback overhead increases linearly with increase in number of users. Hence, we put our efforts to develop efficient low-complexity and limited feedback user and receive antenna scheduling schemes for MIMO and MIMO-OFDM BC with signal-to-interference-plus-noise ratio (SINR) as the feedback data. We considered both homogeneous and heterogeneous networks for all the proposed scheduling algorithms. Further, the user fairness of all the proposed scheduling algorithms are evaluated for heterogeneous networks. In literature, limited feedback scheduling schemes have been proposed where users send only the maximum SINR value to BS for user scheduling purpose. However, this approach does not always ensure the selection of the best user/receive antenna for data communications. Moreover, the full utilization of all the transmit antennas of BS cannot be guaranteed. Hence, to overcome these limitations, it is proposed that each user requires to send an array of maximum SINRs to BS for effective scheduling and to obtain a better spectrum efficiency. To further reduce the feedback load and ensure throughput comparable to full feedback scheduling, four- bit quantized SINR feedback scheduling scheme is proposed. Four-bit quantized feedback scheduling scheme ensures desired spatial multiplexing gain and MUD gain, which are not obtained by one-bit quantized feedback scheduling. Moreover, selective feedback along with 4-bit quantization process results in a significant reduction in feedback overhead with no loss in system throughput. We obtained the optimum quantization threshold and selection threshold by using genetic algorithm (GA) as these thresholds play vital role in selective multi-bit quantization process. DPC is an exhaustive search algorithm (ESA) where the user encoding sequence is important to transmit data to multiple users. Exhaustive search becomes imperative as the search space grows with number of users and number of transmit antennas in the MU MIMO system. We have demonstrated that combined user and antenna scheduling (CUAS) with binary GA (BGA) adopting elitism and adaptive mutation (AM) achieves about 97% - 99% of system sum-rate obtained by ESA like DPC with significant reduction in computational complexity. It is also interesting to observe that BGA is able to find a solution close to the optimum value quite rapidly. In literature, clustering of adjacent subcarriers is discussed for MIMO-OFDM BC to reduce feedback load. Channel conditions of only the center subcarrier of the clusters are conveyed to BS for scheduling process. However, this method proves to be inefficient for higher cluster size and more channel taps. Hence, we propose a new scheduling scheme which attains higher system throughput due to more efficient user and antenna scheduling without any increment in the feedback overhead by considering the channel condition of all the subcarriers present in a cluster for efficient scheduling process. Moreover, the 4-bit quantized scheduling along with optimum quantization thresholds obtained via GA is shown to attain a reduction in feedabck overhead with no loss in system throughput.|
|Appears in Collections:||03. EE|
Files in This Item:
|Limited Feedback Scheduling Schemes for MIMO Broadcast Channels.pdf||2.63 MB||Adobe PDF||View/Open Request a copy|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.