在低軌衛星通訊為現在及未來重要的通訊技術中的一部分,這項技術可以應用於海上通訊、山上通訊、偏僻地區等基地台建設無法普及的地方。近幾年來每年以超過 1000 顆以上的速度來生產低軌衛星,而每年生產低軌衛星的速度也有增長的趨勢,因此在天空中低軌衛星的密度也跟著飛速增長。但現今主要的衛星通訊技術仍以單衛星傳輸技術為主,因此本論文提出全新的資源分配模式,將衛星數視為資源的一部分。使用者選擇合適的兩顆衛星來進行 joint 傳輸,不僅讓使用者能夠實現他所期望的 data rate,同時也能夠降低衛星傳送端所需要傳送的功率,達成一個雙贏的局面。 在傳送的過程中,我們考慮較為實際的情況,讓使用者可以為不同衛星選擇合適的 MCS 來傳送。並且透過 SNR 回推的方式來計算衛星端傳送的功率,使傳送訊號的 BLER 維持在 10%,確保一定的服務品質(QoS)。並且我們透過一些技巧來降低使用者選擇衛星的時間複雜度。在最後,我們考慮衛星間功率分配的議題,設法在傳送端的功率最小化的同時,也能使使用者實現他所期望的 data rate。 ;Low Earth Orbit (LEO) satellite communication is an integral part of current and future communication technologies. This technology can be applied in areas where traditional base station infrastructure is impractical, such as maritime and remote regions. In recent years, the production rate of LEO satellites has been exceeding 1000 satellites per year, and this trend is expected to continue. Consequently, the density of LEO satellites in the sky is rapidly increasing. However, the predominant satellite communication technology still relies on single satellite transmission. Therefore, in this paper, we propose a novel resource allocation scheme that treats the number of satellites as part of the available resources. Users can select two suitable satellites for joint transmission, enabling them to achieve their desired data rates while reducing the transmit power required at the satellite, resulting in a win-win situation. During the transmission process, we consider practical scenarios where users can choose appropriate modulation and coding schemes (MCS) for different satellites. We calculate the transmit power at the satellite by reverse engineering the signal-to-noise ratio (SNR) thresholds to maintain a block error rate (BLER) of 10% and ensure a certain level of quality-of -service (QoS). We also employ techniques to reduce the time complexity associated with user satellite selection. Lastly, we address the issue of power allocation among the satellites, aiming to minimize the transmit power at the transmitter while enabling users to achieve their desired data rates.