Mobile Netw Appl DOI 10.1007/s11036-008-0125-6

A Multi-Scale Statistical Control Process for Mobility and Interference Identification in IEEE 802.11 Ricardo Rabelo Oliveira · Antônio Alfredo Loureiro · Alejandro C. Frery

© Springer Science + Business Media, LLC 2008

Abstract The user centric nature of mobile devices is an important issue for enhancing the user experience in a mobile wireless communication. A strategy for improving such experience consists of applying a resource management focused in providing both the user and the application valuable information about the network context. The communication quality in wireless systems heavily depends on several factors, and adjusting the application and/or alerting the user about the network conditions is an important service when managing this type of devices. The quality perceived by the user depends on the network context, which can be defined as the events that have significant impact on the wireless communication. The identification of the wireless channel behavior is necessary for obtaining a context description. Due to intrinsic features at the wireless signal propagation it is difficult to measure and assess the fundamental wireless network context, e.g., predicting when low quality signal is being caused by interference or fading effects. This work presents a new method for user centric management on

R. R. Oliveira (B) · A. A. Loureiro Departamento Ciência da Computação, Universidade Federal Minas Gerais, Av. Antônio Carlos, 6627 - Pampulha Belo Horizonte - MG Belo Horizonte, Brazil e-mail: [email protected] A. A. Loureiro e-mail: [email protected] A. C. Frery Instituto de Computação LCCV & CPMAT, Universidade Federal de Alagoas, BR 104 Norte km 97 57072-970 Maceió, AL - Brazil e-mail: [email protected]

IEEE 802.11b/g: a strategy for communication survival, focused in improving the user experience. Our proposal uses a multi-scale control chart approach, MCEWMA – Moving Centerline Exponential Weighted Moving Average, associated with multi-resolution analysis using a wavelet transform. Keywords IEEE 802.11 · statistical control process · wavelets · mobility · interference

1 Introduction Mobile devices allowing users to access WLANs in offices or at their homes are a commonplace. The new generation of mobiles devices, such as smartphones, increase the connectivity variety, data storage and portability but also suffers the same limitations that previous types of connection suffer from. Mobile devices should offer both access to wireless services and autonomy through optimized battery usage. Products of the technologycal revolution such as cordless phones, Bluetooth devices and other wireless networks are sources of the interference in indoor environments that, along with the unavoidable objects, alter the communication behavior. Constant changes at the wireless signal may produce disconnections but, before that, the variation in quality of communication affects both the application performance and the user experience. We propose a new resource management technique for improving the user experience. The user suffers different impacts from the wireless environment due to radio propagation characteristics, affected by attenuation, fading, and interference. Considering the communication frame as the basic unit for the data