| Summary: | 碩士 === 國立中正大學 === 通訊工程研究所 === 105 === Understanding of 5G channel characteristics is a must prior to design 5G communication systems. To meet the explosive growth of mobile traffic demand, the exploitation of large bandwidth in the millimeter wave (mmWave) band is inevitable. However, there are fundamental differences between mmWave and low-frequency channel characteristics. Therefore, it is necessary to analyze 5G channel characteristics in high-frequency band at the beginning of designing 5G systems.
There are several ways to characterize the 5G channel in mmWave band, for example, channel measurements and ray-tracing techniques. In particular, site-specific ray-tracing simulation platform is preferable due to its accuracy. The more detailed the environment construction is, the more accurate the simulation result will be.
In this work, we construct three ray-tracing based simulation platforms for urban, campus and indoor open office, respectively. Especially, we collaborate with ITRI/ICL and department of electrical engineering/CCU in verifying the site-specific simulation platform.
With a small wavelength, propagation rays in the mmWave band are sensitive to blockage by obstacles (e.g., humans and furniture). Blockage by a human penalizes the link budget by 20-30 dB. Consequently, the anti-blockage technique will play a vital role in the design of 5G communication systems.
In this work, in addition to the ray-tracing based simulation platforms mentioned above, we propose three beam selection methods for anti-blockage, including
I. Fixed one transmit beam, multiple receive beams
II. Multiple transmit beams, multiple receive beams under same base station
III. Multiple transmit beams, multiple receive beams under different base stations
As the numerical results show, about 20% of users are in outage without any remedy after being blocked. The first method improves outage probability by 2%, and its advantage is the lowest complexity because switching candidate received beams doesn’t need to inform base station. Method 2 improves outage probability by 8% but it results higher complexity because the switching of transmit beams needs to inform base station. The method 3 has best performance with outage probability decreasing of 7% due to wider angular separation of beam-pair-links (BPLs). However, it also needs inter-base-station cooperation to be operable.
|
|---|
