z-logo
open-access-imgOpen Access
Desempenho da estratégia do aperto em leilões recursivos para descarregamento de tráfego via comunicações dispositivo-a-dispositivo
Author(s) -
Lucas Soares de Brito
Publication year - 2016
Language(s) - English
Resource type - Dissertations/theses
DOI - 10.26512/2016.04.d.21553
Subject(s) - computer science , computer network , network packet , bidding , payment , packet forwarding , business , marketing , world wide web
The explosive growth of mobile data traffic in the last few years has lead mobile network operators (MNO) to seek efficient ways to offload their core infrastructure. In particular, deviceto-device communications has emerged as a key technology to accomplish that. In order to work, the MNO’s clients need to be properly incentivized to share their devices’ resources to the benefit of others. One solution to promote user collaboration is the deployment of recursive auctions, i.e., hop-by-hop bidding contests for forwarding packets to their destinations. In this scenario, each client can implement its own auction strategy, so it can share the incentives provided by the MNO (payments, etc.) to accomplish its job. The operator sets a maximum budget for each packet, and the clients pay a fine if the packet is not delivered within a given deadline. In this dissertation, the Tightness strategy for such recursive auctions is evaluated, which is based on the idea of how “tight” a node is to forward a packet to its destination within the associated deadline. Different preference functions (for auction winner decision) are investigated, and the performance of the Tightness strategy is studied in homogeneous networks, i.e., when all devices implement the same strategy. This study is carried out based on discrete-event simulations under static and mobile scenarios. For performance comparison, two baseline strategies are also investigated: one that prioritizes packet delivery over budget gains, and a greedy one, that always pick the lowest bid regardless of packet delivery within the deadline. All strategies are evaluated on discrete-event simulations based on the ns-3 simulator, and compared according to packet delivery ratio, average budget per node, budget fairness, and average number of hops to destination. The presented results show that the Tightness strategy is more effective than simply using shortest-path routing without taking into account the nodes’ bids. This happens because the nodes who perceive a “tight” condition to deliver a packet within the announced deadline discourage the auctioneer from choosing them by bidding high values. The only metric that the Tightness strategy is slightly lower is the fairness, despite not presenting higher variations as the mobility increases, i.e., it is more robust.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here
Accelerating Research

Address

John Eccles House
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom