Affiliations: IBM India Research Laboratory, Block 1, Indian Institute of Technology, Hauz Khas, New Delhi 110016, India Tel.: +91 11 6861100; Fax: +91 11 6861555; E‐mail: {kapurva,rlakshmi,srajeev}@in.ibm.com
Note: [] Present address: Trillium Digital Systems India Pvt. Ltd., 2nd floor, Discoverer building, International Tech Park, Whitefield Road, Bangalore 560 066, India. E‐mail: l_ramachandran@trillium.com.
Abstract: Bluetooth is a promising new technology for short range, low power, low cost, pico‐cellular wireless connectivity between mobile devices. Efficient clustering or topology construction algorithms play a very important role in the fast connection establishment of such networks. The performance of these algorithms is mainly dependent on the device discovery time, i.e., the time taken by a node to discover and to connect to another node in its radio range, as well as the protocols used for organizing them into the required topology. We define metrics for device discovery performance and evaluate them by simulating the inquiry routine in Bluetooth for various ad‐hoc scenarios. We indicate the scenario where device discovery in Bluetooth can become a bottleneck. We then describe some recent results on topology construction algorithms for Bluetooth. Data applications running over Bluetooth such as http, ftp and real audio will need transport layer protocols such as TCP and UDP to send packets over the wireless links. We propose and compare a number of MAC scheduling algorithms with the aim of improving the performance of asynchronous data traffic over a Bluetooth piconet that supports multiple active slaves. We study how the presence of circuit‐switched voice impacts the performance of data traffic.
Keywords: Bluetooth technology, scatternet, piconet, device discovery, inquiry, paging, scheduling, cluster, topology, Media Access Control (MAC), Time Division Duplex (TDD), TCP, IP, UDP, utilization, throughput, end‐to‐end delays
