Hybrid Beam Tracking for Millimeter Wave MIMO Systems Using Particle Filter

• Main Authors: Wang, Yan Cheng, 汪彥呈
• Other Authors: Huang, Yuan Hao
• Format: Others
• Language: en_US
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/696muq

碩士 === 國立清華大學 === 電機工程學系 === 105 === Transmission through the millimeter wave has been considered as one of the potential options for the next-generation wireless communication. By exploiting the property of small wavelength of the millimeter wave signal, the multiple-input multiple-output along with the precoding technology can be applied to overcome the critical problem of signal attenuation. Moreover, the hybrid precoding architecture is included to reduce the hardware cost. This study considers the correlation between temporally neighboring channels and adopts the famous sequential Monte Carlo method known as particle filter to track the evolution of channels. By including the modified mixture framework along with a novel output estimation method, a beam-tracking algorithm is proposed to solve the basis selection problem in the hybrid precoding architecture. Based on the commonly used extended S-V channel model, the simulations under the 16x16 mmWave MIMO system indicates that the proposed algorithm can reduce the computational complexity with negligible performance loss comparing to previous works.