| Summary: | 博士 === 國立臺灣大學 === 電信工程學研究所 === 103 === In wireless communication systems, transmitters always have to transmit data to receivers under time-varying channels. To achieve optimal performance under time-varying channels, link adaptation is an important mechanism in wireless communication systems. Adaptive modulation and coding (AMC) and hybrid automatic repeat request (HARQ) are two important schemes for link adaptation in modern wireless communication systems. AMC can adjust modulation and coding schemes (MCS) according to immediate channel condition so that the system throughput can be maximized, and HARQ can improve system reliability by retransmitting data packet when the packet is incorrectly decoded. However, in practical system design, the AMC is usually implemented with fixed switching thresholds for MCS, and suffers from performance degradation when channel is varying. Although HARQ is designed to offer robust transmissions and hence alleviate the performance degradation due to the use of unfavorable MCS, it also introduces the problem of retransmission latency. In this dissertation, the problem of throughput maximization is investigated for common wireless communication systems. In addition, The method adaptively adjusts the AMC thresholds according to the HARQ acknowledge (ACK) feedback information in each transmission round so that the AMC mechanism can select the best MCS with nearly optimal thresholds.
To verify the capability of the method proposed in this dissertation, this method is applied in two emerging technologies. The first one is wireless relay systems. The wireless relay system contains broadcast and multiple access channels, and therefore the throughput maximization method is difficult to design. After carefully investigating the channel characteristics, the proposed method is modified for wireless relay systems. The second one is Multicast Broadcast Services (MBS), or Multimedia Broadcast Multicast Services (MBMS) in 3GPP Long Term Evolution (LTE) systems. The properties of the MBS are considered in detail, and then the proposed method is modified to fit the technology so that the system throughput can be optimized.
Simulations for different system settings are conducted to verify the performance of the proposed method. Simulation results show that the throughput of the systems are indeed improved by the proposed method under a variety of system assumptions. The proposed method is very easy to implement, and can perform dynamic optimization each time the channel statistics change.
|
|---|
