Age-of-Information-Minimized Scheduling Policies for Updates of Distributed Measurements over OMA and NOMA Wireless Links

• Main Authors: Tan, Yu-Hung, 譚玉宏
• Other Authors: Chen, Po-Ning
• Format: Others
• Language: en_US
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/49d58f

碩士 === 國立交通大學 === 電信工程研究所 === 108 === To tackle the challenge of smart metering over an unreliable wireless link, this thesis investigates the best scheduling policies that minimize the average age of information (AoI) under capacity-achieving repetitive transmissions, where the minimum number of transmissions required to complete a measurement update for each sensor is obtained via the channel capacity formula. In such case, superposition transmissions for the updates of distributed measurements, together with the successive interference cancellation (SIC) receiver at the data center, becomes a novel technology option, which we refer to as the non-orthogonal multiple access (NOMA) setting for simplicity. In order to compare the NOMA technology option with the conventional OMA setting, the best preemptive and non-preemptive scheduling policies that forbid superposition transmissions are determined. The necessary and sufficient condition, under which a non-preemptive OMA-based scheduling policy outperforms a preemptive OMA-based one, is then established, where the condition dictates that the minimum number of transmissions required to compete a measurement update for one sensor cannot be six times larger than that for the other sensor. We end this thesis by showing that when the ratio of the channel gains experienced by two sensors is larger than a certain threshold, adopting the NOMA- based scheduling policy can further reduce the average AoI of the best OMA-based scheduling policy.