On The Training Sequence Design And Application For Channel Estimation In IEEE 802.11ay System

In wireless communication system, the channel estimate accuracy has a decisive influence on the achieved detection performance at the receiver. Most of the systems assume the training sequence based channel estimate scheme, wherein a variety of channel estimate algorithm, such as the maximum likelihood (ML) estimation, linear minimum mean square error (LMMSE) estimation, can be utilized. However, with the development of wireless communication protocol and the demand for higher throughput, the utilized frequency band is gradually shifted to the millimeter wave band. In this thesis, we focus on the channel estimation technology for the next generation of millimeter wave wireless communication protocol of IEEE 802.11 ay. Although the existing channel estimation methods do provide us many choices, the related computational complexity, especially the channel estimate in multiple antenna scenarios, can not fulfill the requirement of simple but efficient channel estimate by IEEE 80211ay.In order to realize simple but efficient channel estimate, we focus on the binary zero correlation zone (ZCZ) based training sequence and its associated channel estimate for IEEE 802.11ay system in this thesis. Secondly, accurate channel estimate is critical for the realized communication performance, which requires enough training payload. Nonetheless, too much training payload will degrade the transmission efficiency. Hence, this thesis tries to analyze the optimal channel estimation training sequence design in terms of the lower channel capacity based on the different channel estimation error. The primary works in this thesis can be summarized as follows. At first, the functional requirements and current research progress on the next generation of millimeter wave communication standard of IEEE 802.11 ay are briefly reviewed. Meanwhile, the detailed design of the physical layer frame structure in IEEE 802.11 ay for different transmission scenarios are also presented to highlight the related training sequence design. Secondly, the definition and the construction of the binary ZCZ sequence, as well as the ZCZ based channel estimate training sequence design for IEEE 802.11ay system are presented. More specifically, the parameter choices of the ZCZ sequence are addressed to highlight the training sequences design for the given transmission mode in the IEEE 802.11 ay standard. At the same time, the comparisions in terms of the channel estimation proformance and the required sequence overhead are presented to show the difference for both the proposed training sequences by IEEE 802.11 ay and the ZCZ based channel estimate sequence. Finally, the channel estimate variance by using the ML algorithm and the LMMSE one is presented. Based the lower bound of the channel capacity affected by noise, the optimal structure of the training sequence, the optimal transmit power of the training sequence and the optimal length of the training sequence are studied to show the influences of these parameters on the lower bound of channel capacity.The analysis in this thesis can provide reference for the further exploration of the channel estimate training sequence design in the future millimeter wave communication system.