近年來無線網路通訊已經倍受重視並且由於行動化是未來網路最重要的趨勢，因此許多的無線網路技術已經被提出並標準化，其中藍芽技術在短距離無線網路中以低價位、低消耗功率、與低干擾吸引眾多的目光。微網是藍芽的基本網路元件，藉由共用連結微網的藍芽裝置可以將數個微網組成一個藍芽散網，這些微網間的裝置又稱為橋，主要負責傳送微網間的封包，由於橋不能同時參與兩個以上的微網，因此如何利用時間分割來分配橋在各個微網間的時間長度就是微網間的排程演算法。

　　在本篇論文中，我們提出一種新的以保持模式為基礎的微網間資料排程演算法，目的在降低網微間的封包延遲時間與提高橋的利用率，然而微網間的封包延遲時間主要的原因在於橋浪費太多時間處於閒置的狀態而造成其他微網的微網封包必須等待這段時間，因此為了降低橋閒置的機率，我們所提出的微網間的排程演算法根據網路的流量情形來決定橋必須停留在不同微網的時間長度。在我們所提出的微網間資料排程演算法，並不需要去修改藍芽的標準或新增網路元件，所以此排程演算法也在實作上較容易達成。最後，由模擬的結果可以得知我們所提的微網間資料排程演算法提供了低延遲、高產能，並適用於多元化的網路拓撲的效能。

　　In recent years, much attention has been given to the wireless communication and the mobility is the most important trend in the future. Thus, many wireless technologies have been proposed and standardized. Among these novel technologies, Bluetooth attracts much attention in short range wireless technology through low cost, low battery power consumption, and low interference. Bluetooth scatternets are formed by piconets, the basic network unit of Bluetooth, sharing common device(s). These inter-piconet devices, i.e. the bridges are responsible for the forwarding of inter-piconet traffic but they can not participate in two or more piconets simultaneously. Thus, this require inter-piconet scheduling algorithm to do time division multiplex for their presence in each piconet.

　　In this paper, we proposed a new inter-piconet scheduling algorithm based on HOLD mode, a power saving mode of Bluetooth to reduce the average delay of inter-piconet packets and increase the utilization of bridge. The main reason of delay in inter-piconet traffic is that bridge wastes time to be idle in a empty piconet and the inter-piconet packets must wait this time period in other piconets. Thus, our proposed scheme decides the time duration which the bridge should spend in different piconet according to the traffic pattern that decrease the probability of bridge is idle. In proposed inter-piconet scheduling algorithm, there is no need to modify the specification of Bluetooth and add any other device so that it is easy to implement and join to specification of Bluetooth. Finally, the result of simulation presents the low average delay, high throughput, and the scalability in our proposed inter-piconet scheduling algorithm.

[1]A.S. Bhargava, D.M. Gruenbacher, “Using Piconet Avoidance Techniques to Reduce Interference in Bluetooth Networks,” IEEE CCNC 2004, Page(s):65 - 70.

[2]A. Racz, G. Miklos, F. Kubinszky and A. Valko, “A Pseudo Random Coordinated Scheduling Algorithm for Bluetooth Scatternets,” ACM MOBIHOC 2001.

[3]Bluetooth SIG, http://www.bluetooth.com.

[4]A. Capone, M. Gerla; R. Kapoor, ”Efficient Polling Scheme for Bluetooth picocells,” IEEE ICC 2001, Page(s):1990 - 1994.

[5]C. Cordeiro, S. Abhyankar, and D.P. Agrawal, “Design and Implementation of QoS-driven Dynamic Slot Assignment and Piconet Partitioning Algorithms over Bluetooth WPANs,” IEEE INFOCOM March 2004, Page(s):1252 - 1263.

[6]H.S. Liron, K. Ronen, S. Adrian, and Z. Gil, “Load-Adaptive Inter-Piconet Scheduling in Small-Scale Bluetooth Scatternets,” IEEE Communications Magazine July 2004, Page(s):136 - 142.

[7]V.B. Misic; J. Misic, “Queueing Analysis of Delay In Bluetooth Slave/Slave Bridge,” IEEE MASCOTS 2002, Page(s):155 – 158.

[8]V.B. Misic; J. Misic, K.L. Chan, “Improving The Performance of Bluetooth Piconets with Synchronous and Asynchronous Traffic,” IEEE GLOBECOM 2003, Page(s):205 - 209.

[9]P. Johansson, M. Kazantzidis, R. Kapoor and M. Gerla, “Bluetooth: An Enabler for Personal Area Networking,” IEEE Network 2001, Page(s): 28-37.

[10]P. Johansson, R. Kapoor, M. Kazantzidis and M. Gerla, “Rendezvous Scheduling in Bluetooth Scatternets,” IEEE ICC 2002, Page(s):318 – 324.

[11]R. Kapoor, A. Zanella, M. Gerla, “A Fair and Traffic Dependent Scheduling Algorithm for Bluetooth Scatternet,” Mobile Networks and Applications 2004, Page(s): 9 – 20.

[12]S. Baatz, M. Frank, C. Kühl, P. Martini and C. Scholz, “Bluetooth Scatternets: An Enhanced Adaptive Scheduling Scheme,” IEEE INFOCOM, June 2002, Page(s):782 - 790.

[13]Sim++ Version 1.0 a software toolkit for event scheduling and queuing model simulations developed at the CISE Department University of Florida Gainesville FL USA.

[14]T.Y. Lin and Y.C. Tseng,” An Adaptive Sniff Scheduling Scheme For Power Saving In Bluetooth,” IEEE.Wireless Communications, December 2002 Page(s):92 - 103.

[15]W. Zhang and G. Cao, “A Flexible Scatternet-wide Scheduling Algorithm for Bluetooth Networks,” IEEE International Performance, Computing, and Communications Conference 2002, Page(s):.291 - 298.