Bluetooth energy-saving optimization: a queuing theory analysis

Bluetooth on mobile terminal can be used to scan the surroundings to discover other Bluetooth-enabled devices. However, with energy limitations, the scanning actions must be well arranged to reduce power consumption. In this paper we propose an energy-saving strategy that turns on and off the Bluetooth device at a configured time. The scenario of Bluetooth scanning is modeled as an M/M/1 vacation queue with special impatient customers and startup expenses. In the M/M/1 queue, the terminal with Bluetooth device is the only server which scans its nearby customers, and the on-off state of Bluetooth device is indicated as vacation behaviors. The aim of the paper is to find the best vacation strategy, or equivalently, the most energy-effective rules. Simulations demonstrate that the best strategy depends on the intended error rate, startup time and expenses, and many other factors. Generally, the performance of the multiple vacation strategy is satisfactory on most conditions.