LTE異質網路中佈署著數種不同大小的基地台，為了提供不同的覆蓋範圍及不同的小區容量並不斷的延伸整體網路容量。在LTE異質網路中，由於使用者佈署的天性，像Femtocell這樣為數眾多且開關頻繁的小型基地台造成了嚴重的小區間干擾問題，特別是運作於CSG模式下。3GPP所提出的小區干擾協調概念可概分為功率限制和無線資源限制兩類作法。在本篇論文中我們考慮了功率、時頻及信號觀點，分別提出了基於這兩類的小區干擾協調演算法。受益於使用模糊多屬性決策，所提出的演算法具有低複雜度及高靈活性。模糊理論提供了具有低複雜度及高彈性的近似決策方法，特別是使用在現今的多參數通訊系統上，就像是LTE系統。網絡指標的多樣性有助於模糊系統做出更好的決策。 本文提出的功率限制演算法為Femtocell決定了適當的服務範圍並為頻率重用演算法像是SFR及FFR提供了區分中心區及分類UE的方法且沒有複雜的小區間溝通。本文提出的無線資源限制演算法衡量了小區覆蓋範圍與容量之間的取捨並做出適當的無線資源調度決定，為了避免在無限資源使用上的衝突。特別是，相較於傳統的基於模糊演算法，所提出的方法沒有使用固定的模糊邏輯規則和有形模糊隸屬模型。 模擬結果顯示，相較於現行的鏈路適配演算法，所提出的演算法可以提升49%的femtocell資料率及18%的macrocell資料率。此外，所提出的演算法能達到56%的上界效能及89%的上界無線資源效益。 ;The Long Term Evolution (LTE) Heterogeneous Networks (HetNet) consists of several different type of base station for providing different coverage and capacity and increasing network capacity continuously. In LTE HetNet, the mass deployment and frequently on/off of small cells, such as femtocells, causes severe inter-cell interference problem due to the nature of user deployment without X2 interface, especially in Closed Subscriber Group (CSG) mode. The concept of Inter-Cell Interference Coordination proposed by The 3rd Generation Partnership Project (3GPP) can be achieved by using power restriction or resource restriction methods. In this article, we proposed two distributed fuzzy-based Inter-Cell Interference Coordination (ICIC) algorithms based on the concept of the power restriction and resource restriction. The low complexity and high flexibility of the proposed algorithms is benefited from the fuzzy Multi-Attributes Decision Making (MADM). Fuzzy theory can provide means to make approximate decisions with low complexity and high flexibility, especially in current multi-parameters communication systems, such as LTE system, in which the diversity of network metrics can help fuzzy system to make better decision. The proposed adaptive power restriction algorithm provides an appropriate serving range for femtocells, determining center zone and separating UEs into cell center and cell edge for frequency-reused algorithms, such as Soft Frequency Reuse (SFR) and Fractional Frequency Reuse (FFR), without complicated negotiation among cells. The proposed adaptive radio restriction algorithm weighs the trade-off between coverage and capacity by leveraging three system metrics to make an appropriate scheduling decision to avoid the conflict in radio resource used among cells. In particular, there are no fixed fuzzy logic rules and shaped fuzzy membership model compared to conventional fuzzy-based algorithms. The simulation results show that proposed algorithms provide about 49 % data rate improvement for femtocell and about 18 % data rate improvement for macrocell compared to current link adaptation algorithm. In addition, it can achieve up to 56 % data rate and 89 % radio resource efficiency of the up bound case.
