裝置對裝置間干擾感知下的模式選擇

Abstract

在本篇論文中，我們在有多個裝置與裝置對 (D2D pair) 的情況下考慮在支撐式細胞網絡下裝置與裝置間的通訊，並著重於新加入網絡D2D pair的模式選擇，我們採用了兩種模式選擇方法，分別是依據SINR和capacity，每一種模式選擇方法都會搭配兩組不同的預編碼器來做討論，第一組預編碼器是cellular users使用Block Diagonalization (BD) 預編碼器，D2D users使用Zero Forcing (ZF) 預編碼器，第二組則是cellular users和D2D users都使用signal-to-leakage-and-noise ratio (SLNR) 預編碼器但有各自不同的算法。最後，我們的模擬結果顯示使用SLNR預編碼器不僅只需要滿足比BD和ZF更少的條件還能夠達到接近最佳的結果。此外，無論現存D2D pairs是在underlay或overlay模式，對於新加入網絡的D2D pair來說underlay 模式是最佳的模式。最後，資源分配對於提升整個cell的sum rate來說是個非常重要的因素。

In this work, we consider device-to-device (D2D) communication underlaying a cellular network and focus on the mode selection of the new arrival D2D pair with existing D2D pairs. We also utilize two mode selection rules which is based on signal to interference noise ratio (SINR) and capacity. Each selection rule are discussed with two different sets of precoders. The first set is block diagonalization (BD) precoder for cellular users and zero forcing (ZF) precoder for D2D users. The second set is signal-to-leakage-and-noise ratio (SLNR) precoder with two different derivations for cellular users and D2D pairs, respectively. Our evaluation results show that utilizing SLNR precoders can not only achieve a suboptimal performance but also require much less limitations than BD and ZF. The simulation also shows that underlay mode is the best mode for new arrival D2D pair whether the existing D2D pairs are in underlay or overlay mode. In addition, resource allocation is a critical factor for improving the sum rate of the whole cell.