階層式行動通信系統之通道指派策略

Abstract

本論文是研究階層式行動通信系統之通道指派策略，吾人首先提出合成通道指派策略。本策略之主要目標是提高系統之通道使用率，減少呼叫之阻塞率。為達成上述目標，所提出合成通道指派策略是結合升流、降流和逆流策略。升流策略是呼叫發生在微細胞與巨細胞重疊區域時，若是微細胞已經沒有通道可以使用時，則可以將此呼叫升流到巨細胞使用它的空閒通道。降流策略是相鄰的巨細胞中的呼叫交遞到微細胞與巨細胞重疊區域時，若是巨細胞已經沒有通道可以使用時，則可以將此交遞呼叫降流到微細胞使用它的空閒通道。逆流策略是為了提高微細胞之通道使用率，可以將巨細胞中的呼叫逆流至本身所涵蓋的微細胞。 一般而言，客戶對於進行中的呼叫被中斷，其受到抱怨的程度比新呼叫被阻塞的程度高。為了減低客戶抱怨的程度，傳統式之通道指派方法是保留部份通道給優先順序較高之交遞呼叫使用。但是究竟要保留多少通道才能維持交遞呼叫的服務品質，而且通道使用率是最有效率。往往在一個負載隨機變動的系統是很困難達成的，它不是不可能做到而需要經由很複雜的系統模式與數學推導。吾人設計之乏晰通道指派控制器不僅可以達成維持交遞呼叫的服務品質保證，而且又能達成高的通道使用率。乏晰通道指派控制器之設計觀念是使用二層式乏晰邏輯控制方式，能簡化設計之複雜度，並且能解決複雜的通道指派問題。乏晰通道指派控制器已經經由系統模擬得到其結果與傳統式之通道分派方法比較，乏晰通道分派控制器的確能維持交遞呼叫的服務品質及得到較高的通道使用率。 由於未來的無線通信系統將可以提供多媒體的服務，因此系統必須具有高速率及高容量的能力。針對高速率的階層式行動通信系統，吾人提出類神經乏晰通道指派控制器，它具有類神經學習能力及吸納乏晰理論的優點。為了達成較高的通道使用率以及維持交遞呼叫的品質保證，它的之設計觀念是可以做彈性的允諾控制以及軟性的速率管理。類神經乏晰通道指派控制器之設計方法亦是使用二層式控制方式，第一層是乏晰通道選擇器，它選擇有效的系統訊息作為乏晰通道選擇器之輸入變數，然後應用最小-最大推論方式，最後決定呼叫要分派到巨細胞或微細胞。第二層是類神經乏晰通道處理器，它應用五層類神經乏晰架構並結合補強式學習方式來執行彈性的允諾控制以及軟性的速率管理。系統模擬顯示，類神經乏晰通道指派控制器可以維持交遞呼叫的品質，得到較高的通道使用率，以及高的客戶滿意指數。

In this dissertation, channel assignment schemes of hierarchical cellular systems with overlaying macrocells and overlaid microcells are studied. First, a combined channel assignment (CCA) mechanism is proposed. The objective of the proposed CCA mechanism is to reduce the new-call blocking probability, decrease forced termination probability, and increase the channel utilization for the whole system. The proposed CCA mechanism combines overflow, underflow, and reversible schemes to reduce both the new-call blocking probability and forced termination probability. Also, it adopts the buffering scheme and considers load balancing between macrocell and microcell to obtain a high channel utilization. Next, a novel scheme, a fuzzy channel allocation controller (FCAC), is proposed for hierarchical cellular systems. The objective of the proposed FCAC scheme is to increase the channel utilization for the whole system and to guarantee the required QoS of handoffs as well. The FCAC mainly contains a fuzzy channel allocation processor (FCAP) which is designed to be in a two-layer architecture: a fuzzy admission threshold estimator in the first layer and a fuzzy channel allocator in the second layer. The FCAP chooses the handoff failure probability, defined as quality-of-service (QoS) index, and the resource availability as input linguistic variables for the fuzzy admission threshold estimator. The Sugeno's position-gradient type reasoning method is applied in the fuzzy admission threshold estimator to adaptively adjust the admission threshold for the fuzzy channel allocator. For achieving channel allocation with flexible admission and soft rate control, a neural fuzzy channel allocation controller (NFCAC) is designed for multimedia services in hierarchical cellular system. The design concept of NFCAC is to support flexible resource allocation and soft rate control so that the high blocking rate in a hierarchical cellular system providing multi-channel services can be reduced. The NFCAC mainly contains a neural fuzzy channel allocation processor (NFCAP) which is designed to be in a two-layer architecture that consists of a fuzzy channel selector (FCS) in the first layer and a neural fuzzy call-admission and rate controller (NFCRC) in the second layer. FCS is a fuzzy control logic, and it can properly determine which channel in either macrocell or microcell to be allocated. The NFCRC is a five-layer neural fuzzy controller with the reinforcement learning architecture to perform the flexible admission control and soft rate management. 2. Hierarchical Cellular Systems 3. A Combined Channel Assignment Mechanism 4. A QoS-Guaranteed Fuzzy Channel Allocation Controller 5. A Neural Fuzzy Channel Allocation Controller for Multimedia Services 6. Conclusions