預付式行動電話和優先電話服務

Abstract

近年來預付式行動電話服務的用戶數成長快速，成為行動電話中一個重要的應用。目前提供預付式服務方法有四種：快速記帳法（hot billing）、服務節點法（service node）、智慧網路法（intelligent network）和手機計費法（handset-based）。快速記帳法不需對原有的網路架構作太大的更動，而服務節點法是目前預付式服務最受歡迎的方法之一，並被視為演進至未來智慧網路架構的墊腳石。在本論文中，我們研究以快速記帳法和服務節點法為架構的預付式行動電話服務系統之效能。 快速記帳法利用行動交換機產生的通話記錄（CDR）內容，扣除原有的預付金額。這些通話記錄是在一通或數通電話結束後，由行動交換機產生，並傳送至預付服務中心。因此在使用者通話結束時，其預付金額可能已經變成負值。這個負的金額部分稱為可能的呆帳。在快速記帳法中，選擇合適的通話記錄傳送頻率，讓通話記錄傳送/處理成本和呆帳的和最小，這個問題是很重要的。我們提出數學分析模式和利用電腦模擬方式，研究快速記帳法在大量預付金額、少量預付金額及可補充預付金額等三種情形下之系統效能。我們的實驗結果顯示出預付金額和通話費用的變異數如何影響網路信令交通和可能的呆帳大小。根據我們的分析模式，業者可以找出快速記帳法的最佳通話記錄傳送頻率。 與快速記帳法相較，服務節點法提供近乎即時的通話控制，和即時的預付金額監督功能。當手機打出一通預付電話後，服務節點在通話期間內，定期檢查並扣除預付金額，避免巨大的呆帳。理論上藉由升級交換機或服務節點的處理能力，服務節點可以提供即時監督預付金額的功能。但是由於所有的服務控制及通話交換功能均由服務節點負責，因此服務節點的監督及更新使用者預付金額的能力便受到限制。在服務節點方法中，如何規畫服務節點監督及更新使用者預付金額的成本是很重要的。我們提出一個數學分析模式，研究預付金額的檢查次數如何影響服務節點的工作量及業者所能負擔的呆帳大小。 隨著預付服務市場的成長，資源分配問題成為預付服務的一個重要問題。過去有許多以排隊方式為主的優先電話服務方法曾被提出，以改善系統效能。在排隊方法中，若系統的資源不夠，則優先電話將被放在一個佇列內。如果手機在離開基地台範圍前，系統有足夠的資源，則優先電話便可以為系統所服務。相反的對於非優先服務的電話，只要系統的資源不足，便不提供服務。本論文提出數學分析模式及利用電腦模擬，研究優先電話服務系統的效能。我們的研究指出，在優先電話服務中整體的通話阻塞及通話中斷比率，比不具有優先電話服務的系統要低。 此外，優先電話服務與預付式服務可以相互結合，提供行動電話網路的整合服務。在論文中我們利用電腦模擬，研究一個結合優先電話服務，與以服務節點為架構之預付式服務的行動電話系統。我們研究預付電話的比例、系統資源大小及使用者移動性對系統效能的影響。本論文並提出一個收益函數，以比較兩種優先權分配方式的好壞。我們的實驗顯示這個函數可作為選擇優先權分配方式的參考依據。

Prepaid service has become an important mobile application with rapid growth of subscription rate in the recent years. Four solutions have been proposed to implement the prepaid services: hot billing approach, service node approach, intelligent network approach and handset-based approach. The hot billing approach provides a solution without major changes to the network infrastructure. The service node approach is the most widely deployed prepaid solution today and is viewed as a stepping-stone to the intelligent network approach. In this dissertation, we investigate the performance of mobile prepaid services based on these two approaches. Hot billing uses the call detail records (CDRs) produced by the mobile switching center (i.e., MSC) to process the prepaid usage. These records are generated and transported to the prepaid service center when a call or multiple calls complete. The prepaid credit may become negative at the end of a phone call. This negative credit is referred to as the potential bad debt. In the hot-billing approach, it is important to select the CDR sending frequency such that the sum of the CDR sending/processing cost and the bad debt is minimized. We propose analytic and simulation models to study the performance in the large, small credit and the recharged credit cases. Experimental results explain how the amount of prepaid credit and the variance of call charges affect the network signaling and the potential bad debt. Based on our modeling technique, an optimal CDR transmission frequency can be found. Compared with the hot billing approach, the service node approach allows near real-time call control and credit monitoring. When a mobile originates a prepaid call, the service node checks and decrements the prepaid credit periodically during the conversation to avoid large bad debt. Theoretically by upgrading the processing power of the switch or service node, the service node will permit a real-time credit monitoring. However, the capability of the service node may be limited since all service control and call-switching functions are implemented on the service node. It is important to plan the processing budget for a service node. We propose an analytic model to investigate how the number of credit checks affects the workload of the service node and the bad debt that a service provider may bear. With the fast growth of prepaid market, the resource allocation has also become an important issue in mobile prepaid services. Many priority solutions based on the queuing scheme have been proposed to improve the system performance. In the queuing schemes, when the resources are not enough, a priority call is placed in a queue. The call can be served later if the resources become sufficient before the mobile leaves the cell. On the other hand, a call request issued without the priority service is blocked if the resources are insufficient. In this dissertation, we develop analytic and simulation models to evaluate the performance of the priority call service. Our study indicates that, in the priority service, the net effect of call blocking and dropping rates are lower than the non-priority service. Additionally, priority service and prepaid call service can be combined to provide an integrated service in a mobile network. In this dissertation, we conduct simulation experiments to study a mobile phone system that provides priority and prepaid services based on the service nodes. We investigate how the traffic ratio of prepaid calls, the amount of resources and the user mobility affect the system performance. A revenue function is proposed to compare two priority assignment schemes. Our experiments show that this function can be used as a guideline to select the priority assignment scheme.