Skip to Main content Skip to Navigation

Channel Allocation in Mobile Wireless Networks

Abstract : The intensive use of mobile data services has increasingly increased resource consumption over wireless networks. The main resource used for wireless communication is the frequency spectrum. As data traffic rises sharply, new bands of the frequency spectrum are not available in the same proportion, making the frequency spectrum increasingly scarce and saturated. Several proposals have been presented to optimize frequency channel allocation in order to mitigate interference between nearby links that are transmitting data. Many of them use a single criterion and does not consider the user behavior to guide the channel allocation process. Users have routine cycles and social behavior. They routinary move to work, to school, use their mobile devices generating data traffic, and they meet with friends forming clusters. These characteristics can be explored to optimize the channel allocation process.This thesis presents a channel allocation strategy for wireless networks based on user behavior. Our main contribution is to consider some characteristics of the user behavior, such as mobility, traffic, and popularity in the channel allocation process. In this way, we prioritize the channel allocation for the nodes that will remain in the network in a future time window, with higher traffic in the network, and with more popularity. We adopt a distributed approach that allows limiting the number of messages exchanged in the network while quickly responding to changes in the network topology. In our performance evaluation, we consider scenarios in ad hoc and vehicular networks, and we use some synthetic mobility models, such as SLAW and Manhattan grid, and the traces dataset of Cologne city. In the scenarios, we compare our mechanism with different types of approaches: i.e., a centralized (named TABU), a random (named RANDOM), a with largest spectral distance (named LD), and a with single channel (named SC). We evaluate metrics such as aggregated throughput, packet delivery rate, and end-to-end delay. Simulations considering ad hoc scenario with unicast routing show that our strategy presents improvements in terms of throughput at the order of 14.81% than RANDOM and 16.28% than LD channel allocation. In vehicular scenario, our strategy shows gains of packet delivery rate at the order of 11.65% and 17.18% when compared to RANDOM and SC methods, respectively. In both scenarios, the performance of our strategy is close to the upper bound search of the TABU centralized approach, but with lower overhead.
Complete list of metadata

Cited literature [116 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, October 10, 2018 - 9:49:06 AM
Last modification on : Friday, February 4, 2022 - 3:09:34 AM
Long-term archiving on: : Friday, January 11, 2019 - 12:43:02 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01891855, version 1


Roni Shigueta. Channel Allocation in Mobile Wireless Networks. Library and information sciences. Université Paris-Saclay; Pontifícia universidade católica do Paraná, 2018. English. ⟨NNT : 2018SACLX037⟩. ⟨tel-01891855⟩



Record views


Files downloads