Evaluating a Dynamic Keep-Alive Messaging Strategy for Mobile Pervasive Systems
Björn A. Johnsson, Mattias Nordahl, Boris Magnusson
Abstract
Open-access reader
Björn A. Johnsson, Mattias Nordahl, Boris Magnusson
Abstract
Open-access reader
Timely loss-of-contact detection between devices in pervasive systems using mobile networks is an important aspect for both functionality and user experience. Traditional schemes where time-to-live is defined by the service provider are not adequate in mobile networks where communication failures are frequent. On the other hand, schemes using keep-alive messaging tend to increase the communication overhead, in particular if the time to detect a loss of contact needs to be short. We introduce a strategy where the time between keep-alive messages is adjusted dynamically according to the need in a particular application. Our solution was evaluated with usage data from a professional application built on the PalCom framework. The measurements show a 96 % decrease of the communication overhead while still maintaining the same system responsiveness.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Timely loss-of-contact detection between devices in pervasive systems using mobile networks is an important aspect for both functionality and user experience. Traditional schemes where time-to-live is defined by the service provider are not adequate in mobile networks where communication failures are frequent. On the other hand, schemes using keep-alive messaging tend to increase the communication overhead, in particular if the time to detect a loss of contact needs to be short. We introduce a strategy where the time between keep-alive messages is adjusted dynamically according to the need in a particular application. Our solution was evaluated with usage data from a professional application built on the PalCom framework. The measurements show a 96 % decrease of the communication overhead while still maintaining the same system responsiveness.
Key concepts: Computer science, Overhead (engineering), Service provider, Service (business), Computer network, Distributed computing, Mobile device, Mobile telephony