On Adaptation Technique For MIMO Wireless Communications

The support of high speed multimedia data transmission is an irresistible trend of wireless communication system. To achieve this goal the key point is to increase channel capacity. However it may not be practicable to expand capacity by simply increasing the bandwidth and transmit power. MIMO (Multi-Input Multi-Output) tech-niques employ multiple antennas at both ends of wireless link. By space-time signal processing methods, it can achieve much higher spectral efficiency and transmission reliability than that of the single-antenna link.The dissertation focuses on several issues of adaptation in wireless MIMO imple-mentations. The main contributions are as follows.1. In adaptive modulation system with transmit power constraint, there is always power residual due to limited constellation. We proposed power and bit adap-tation algorithms to utilize the residual power in spatial domain as well as in space-time domain for adaptive MIMO systems. Simulation results indicate that the proposed methods can effectively improve spectral efficiency. The algorithm is of low complexity and easy to implement.2. An adaptive MIMO transmission technique in IEEE 802.11n clustered MIMO channels is proposed. In the strategy a compound antenna structure composed of uniform linear array, uniform circular array and star array, and two fundamen-tal MIMO transmission schemes including spatial multiplexing and beamforming are considered. The mobile station and base station utilize the instantaneous and spatially statistical channel information to cooperatively implement adapta-tion among different transmission schemes and antenna array types. Compared with the method only using statistical channel information, the proposed algo-rithm can achieve better capacity performance approximating that of the optimal strategy. Moreover, the algorithm is of low complexity and feedback cost.3. We proposed a distributed scheduling algorithm for multiuser MIMO (MU-MIMO) downlink with dynamic variation of transmission mode and feedback cost. With the algorithm each mobile station (MS) chooses transmission scheme accord-ing to its channel state and then feeds back channel state information (CSI). Based on received CSI, base station (BS) selects user subset to communicate with and employs precoding to eliminate co-channel interference (CCI). User scheduling and transmission mode adaptation are carried out by MS and BS corporately. The system feedback cost depends on each MS's channel state and dynamically adapt to communication environment. Compared with existing methods the proposed algorithm can achieve higher system capacity and better BER performance with appropriate feedback load.4. We investigated the communication strategies for MU-MMIO downlink with re-duced feedback and proposed two adaptive transmission schemes. Thresh-old policy in company with transmission mode adaptation is employed. Each MS carries out channel estimation, selects appropriate transmission mode and feeds back channel information adaptively. Users whose channel quality is be-low the given threshold would keep silent and make no attempt to access chan-nel resources. With feedback information BS schedules one user out of can-didate MSs in each time slot. The first strategy employs single-level threshold from which the transmission mode selection criterion is derived. The second one employs two-stage thresholds with which a tight one is used for feedback load reduction and a loose one is to guarantee good outage performance, re-spectively. In evaluating the strategies, theoretical analysis and numerical sim-ulation are used. Results show that compared with strategies without thresh-old, single-level threshold strategy can achieve high throughput with low system feedback cost. Moreover, compared with the strategy with single-level thresh-old the double-level one can further reduce feedback and maintain low outage probability at the cost of negligible throughput loss.5. We proposed an MU-MIMO downlink transmission strategy to solve the trans-mission problem for MIMO broadcast channels without available CSI. With the scheme, BS selects and transmits to a set of MSs simultaneously. Each transmit antenna at BS corresponds to an MS in the active user subset. The MS applies independent component analysis (ICA) algorithm to demix different components from the received mixing signals and identifies its own. Simulation results show that applying the proposed strategy, MU-MIMO downlink transmission can be implemented successfully without CSI.