CDPD機制之研究

Abstract

蜂巢式數位封包數據(Cellular Digital Packet Data；簡稱CDPD)網路以共享AMPS類比式行動電話網路之無線電設備與無人使用之無線電頻道的方式，來提供行動用戶無線數據通訊服務。CDPD使用AMPS的閒置頻道來傳送數據封包，並且在此頻道被指派做語音服務時，自動釋放此頻道。 CDPD使用睡眠模式(Sleep mode)的機制，來節約手機能量的消耗。一個叫做T203的計時器決定手機何時進入睡眠模式，另一個叫做T204的計時器決定何時手機必須醒來。如果我們選擇一個大T204值和一個小T203值，則電池能量的消耗可有效地降低，不過卻必須付出資料框傳輸效能惡化的代價(換句話說，就是遺失的資料框更多、資料框等待時間更長、且等待時間的變異數(variance)更大)。 由於CDPD使用AMPS閒置的無線電資源，因此要如何確保AMPS的正常運作不被CDPD所影響，便成為一個重要的課題。針對這點，CDPD在規格書中定義了數個計時器(如居住計時器dwell timer和中止計時器layoff timer)與系統參數(如CDPD最多可同時佔用無線電頻道的數目)。我們探討這些計時器和系統參數如何影響AMPS/CDPD系統的效能。換句話說，我們研究如何在CDPD的可用率(CDPD availability)和語音電話塞機率之間做妥協。在本論文中，CDPD的可用率定義為：無線電頻道有多少比例的時間，可被CDPD用來傳送數據封包。我們發現，CDPD居住時間太長和CDPD最多可同時佔用無線電頻道數目太大時，會導致語音電話的塞機率惡化，但卻無法改善CDPD的可用率。反之，CDPD中止時間太長會導致CDPD的可用率惡化，但卻無法改善語音電話的塞機率。 AMPS/CDPD的通道選取法則會影響系統的效能。由於AMPS語音需求的優先率高於CDPD，因此當一個語音需求將要使用一個正被CDPD通道流(CDPD channel stream)所佔用的無線電頻道時，此CDPD通道流必須強迫跳躍(forced hop)以離開這個無線電頻道。語音通道選取法則會影響強迫跳躍的次數。我們提出一個分析模式(analytic model)來研究隨機與最少閒置時間(least-idle)這兩個語音通道選取法則對CDPD通道佔用時間的影響。 CDPD也會定期執行定時跳躍(timed hop)以避免封鎖(seal)無線電頻道。針對一個採用強迫跳躍與定時跳躍機制的CDPD系統，我們提出一個模擬模式(simulation model)來研究如何在CDPD通道佔用時間與語音通話未完成率(voice incompletion probability)之間做妥協。此模擬模式也考慮到語音使用者移動率的效應。 總而言之，我們提出分析模式與模擬模式來研究CDPD規格書中定義的計時器與系統參數的效應，以及AMPS與CDPD系統間的交互影響。我們的研究可提供為CDPD網路規劃之參考依據。

Cellular Digital Packet Data (CDPD) networks provide wireless data communications services to mobile users by sharing the radio equipment and unused RF channels with Advanced Mobile Phone Service (AMPS) networks. CDPD uses idle frequency channels of AMPS to transmit packet data, and autonomously releases the occupied RF channels when these channels are about to be assigned for voice requests. CDPD utilizes the sleep mode mechanism to conserve the power of the mobile end systems (M-ESs). A timer called T203 determines when an M-ES enters the sleep mode, and another timer called T204 determines when the M-ES wakes up. A large T204 value and a small T203 value can effectively reduce the power consumption at the cost of degrading the packet transmission performance (i.e., more lost frames, longer frame waiting times, and larger waiting variance). Since CDPD utilizes the idle radio resources of AMPS, it is important to ensure that the normal AMPS activities are not affected by CDPD. Several timers (such as the dwell timer and the layoff timer) and system parameters (such as the maximum number of RF channels that can be simultaneously used by the CDPD channel streams) are defined in CDPD for this purpose. We investigate how these timers and parameters affect the performance of the AMPS/CDPD systems. Specifically, we study the trade-off between CDPD availability and voice blocking probability. In this dissertation, CDPD availability is defined as the proportion of the time that the RF channels can be used to transmit CDPD data. We observe that by selecting a large value of the CDPD dwell time and a large number of CDPD channel streams will degrade the voice blocking probability without improving CDPD availability. On the other hand, by selecting a large value of the CDPD layoff time will degrade the CDPD availability without improving voice blocking probability. Since the AMPS requests have priorities over the CDPD activities, a CDPD channel stream is forced to hop when a voice request is about to use the RF channel occupied by that channel stream. The number of forced hops is affected by the voice channel selection algorithm. We propose an analytic model to investigate the CDPD channel holding time for the the least-idle and random voice channel selection algorithms. CDPD periodically performs timed hop to avoid sealing of RF channels. For the CDPD system with forced hop and timed hop mechanisms, We propose a simulation model to study how channel selection algorithm affects the trade-off between CDPD channel holding time and voice incompletion probability. This simulation model also considers the effects of voice user mobility. In summary, we propose analytic and simulation models to study the effects of the timers and system parameters defined in the CDPD Specification, and the interaction between the AMPS and the CDPD systems. Our study provides guidelines to CDPD network planning.