Implementation of Customized Greencall Algorithm for Energy-Efficient Of Wireless LANs
Emerging dual-mode phones incorporate a Wireless LAN (WLAN) interface along with the traditional cellular interface. The additional benefits of the WLAN interface are, however, likely to be outweighed by its greater rate of energy consumption. This is especially of concern when real-time applications, that result in continuous traffic, are involved. WLAN radios typically conserve energy by staying in sleep mode. With real-time applications like data transfer, this can be challenging since packets delayed above a threshold are lost. Moreover, the continuous nature of traffic makes it difficult for the radio to stay in the lower power sleep mode enough to reduce energy consumption significantly. In this work, I propose the GreenCall algorithm to derive sleep/ wake-up schedules for the WLAN radio to save energy during data transfer while ensuring that application quality is preserved within acceptable levels of users. I evaluate GreenCall on commodity hardware and study its performance over diverse network paths and describe my experiences in the process. I further extensively investigate the effect of different application parameters on possible energy savings through trace-based simulations. I show that, in spite of the interactive, real-time nature of data transfer, energy consumption during calls can be reduced by close to 80 percent in most instances.