InsideODU: News for Faculty and Staff
Thursday, February 14, 2013

 Top Story


Uddin (left) and Nadeem
Audio Scheme Saves Power for Wi-Fi Users

Users of smartphones and tablets jump at the chance to use Wi-Fi radio-wave interfaces to connect to the Internet. Whether at home or a coffee shop, the users find convenience and cost savings in the interfaces. But they also find a vexing problem - the interface's drain on a device's battery.

Now, researchers at Old Dominion believe they have found a solution, counterintuitive though it may be.

Tamer Nadeem, an assistant professor of computer science, and a graduate student he mentors, Mostafa Uddin, have come up with a solution that might seem at first blush to be another drain on the battery. Theirs is an audio solution that requires a boost from a smart device's microphone and speaker.

Continuous use of a Wi-Fi interface is a major drain on the power of smart devices. Therefore, a device depends upon a power-saving mechanism (PSM) to minimize the wakeup time of the Wi-Fi interface. When devices are connected to an access point (AP), typically the Wi-Fi interface of the smart device has to periodically wake up, say, every 0.1 seconds, to check on the Wi-Fi beacon frame from the AP. The device, based on the information in the beacon, decides whether there are any data held at the AP that need to be fetched.

"This periodic wakeup of the Wi-Fi interface still consumes a significant amount of energy," Nadeem explained. "From experiments, we found that the Wi-Fi interface consumes three times more energy then the audio interface in smart devices. Therefore, in our scheme, we utilize the audio channel to communicate periodically between the device and the AP instead of waking up the Wi-Fi interface."

Added Uddin, "Statistics show that users prefer the Wi-Fi interface on their smart devices for Internet access. The typical connection mode in homes, workplaces and even for outdoors, such as in coffee shops, is this infrastructure mode. Hence our scheme, which we call A2PSM, is applicable in most of our daily activities."

Uddin said A2PSM also permits direct communication between the Wi-Fi peer devices, such as file sharing, multiplayer games and media streaming, all of which are gaining in popularity.

In A2PSM, the audio interface uses an acoustic signal as an audio beacon, which carries specific information similar to the Wi-Fi beacon frame. "In the proposed scheme, this audio beacon is generated in a periodic fashion similar to the Wi-Fi beacon," Uddin said. "This is an audio signal that contains one or more high frequency sinusoidal signal components - greater than 18 kilohertz, which is too high for the human ear to detect. We use these frequencies to carry PSM information similar to the one in the Wi-Fi beacon."

Nadeem and Uddin say they believe that smart hand-held devices will come to be the dominant computational devices and that the need to overcome the inefficiencies of Wi-Fi interfaces will increase.

"Given the several inefficiencies of Wi-Fi interfaces today, we envision introducing this novel communication framework that utilizes the audio interface, that is, the microphone/speaker, on smart devices in developing more efficient Wi-Fi networks," Nadeem said. "Unlike other interfaces in smart devices such as Bluetooth interface, audio interface, hardware and software, in current commodity smart devices are open and flexible, which enables us to integrate with the Wi-Fi interface at lower layers of the network stack (MAC and PHY layers) to realize our vision."

The researchers said their experimental evaluation has shown that their audio solution works at the 3-meter range between two smartphones and that they believe the A2PSM scheme should be able to support up to a 30-meter range. "As a future work, we want to evaluate our scheme at higher distances for more powerful speakers," Nadeem said.

And what about the noise inside a coffee shop? Will the audio system tolerate that? "We evaluated the scheme under different indoor/outdoor noisy environments," Uddin explained. "Given that human voices and most of the background noises are far below 18KHz, typically below 6kHz, we find the frequency range we use for communication is quite distinctive from surrounding noises."

Even though their scheme does have the advantages of not interfering with Wi-Fi networks and saving power consumption, it also presents challenges because of the slow propagation and low data rate of the acoustic channel, the researchers said.

This research of Nadeem and Uddin has already been the subject of an article in V3, an online computer topics magazine based in the United Kingdom. They will present the work later in February at the HotMobile '13 Conference in Jekyll Island, Ga.

Return to InsideODU