使用三維波束成形輔助在異構無線網路的流量卸載

Abstract

波束成形(Beamforming)可看成是一種空間性的濾波器(spatial filtering)，透過在傳輸端對訊號加權處理，形成狹窄的發射波束。能夠加強特定方向的訊號強度，並降低對其他方向上的干擾，有效的增加目標的信號與干擾加雜訊比(SINR)以達到增強系統效能的功用。三維波束成形(3D Beamforming)在水平的維度外多增加了垂直的維度，增長的自由度讓頻率複用有著更好的效果。 現有的流量卸載研究主要分以使用者為主，提供原則或方法讓使用者選擇網路。或是以系統為主，考慮二個以上的系統，目標是最大化整體的效能等等的目標式。但這些研究很少有保證所切換用戶的速率需求，這可能造成雖然系統效能佳，但某些信號較差的使用者無法分得資源。我們論文中考慮具三維波束成形(3D Beamforming)的系統存在於異質網路的的環境。提出在滿足使用者的速率要求下，達到盡可能的服務更多的使用者。若在演算法所排程的一定時間內，使用者均能得到AP發射的訊號服務，則將此使用者加入到我們的系統當中。 我們將目標表示成最佳化的式子，由於最佳解的複雜度過高，所以我們提出了啟發式的演算法來解決所提出的問題。透過小規模的模擬來比較和最佳解的效能差異，以及在大規模環境下各個參數調整對整體系統效能的影響。

Beamforming or spatial filtering is a signal processing technique used in sensor arrays for directional signal transmission or reception.This is achieved by combining elements in a phased array in such a way that signals at particular angles experience constructive interference while others experience destructive interference. Beamforming can improve user throughput by increasing signal-to-interference-plus-noise ratio. Three-dimensional (3D) beamforming combines the horizontal and vertical antenna pattern adaptation. The more degree of freedom make frequency reuse has better results. Existing traffic offload study points to a user-based, providing principles allow users to choose the Radio Access Technology. Or to the system-based, consider heterogeneous networks, the goal is to maximize the overall system throughput, etc. However, these studies rarely ensure the rate requirements of users, which may result in system performance although good, but some poor signal user cannot share resources. We consider a three-dimensional beamforming (3D Beamforming) systems exist in heterogeneous networks environment. Presented at a rate to meet user requirements, the highest possible service to more users. We will formulate target as optimization equation, since time complexity of finding optimal solution is too high, so we propose a heuristic algorithm to solve the problems. Effectiveness difference through small-scale simulation to compare our algorithm and optimal solution, and adjust various parameters impact on the overall system performance in large-scale environments.