Algorithms for Self-Organized Resource Allocation in Wireless Networks
Autor: | Olav Tirkkonen, Juha-Matti Koljonen, Mikko J. Alava, Matti Peltomäki |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2012 |
Předmět: |
Mathematical optimization
Theoretical computer science Computer Networks and Communications Computer science Frequency assignment ta221 Aerospace Engineering Local optimum Local search (optimization) Resource management Electrical and Electronic Engineering Global optimization ta113 ta214 Channel allocation schemes ta114 ta213 Wireless network business.industry ta111 Cognitive network Distributed algorithm Automotive Engineering Cellular network Resource allocation business Algorithm Communication channel |
Zdroj: | IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY. 61(1):346-359 |
ISSN: | 0018-9545 |
DOI: | 10.1109/TVT.2011.2151891 |
Popis: | We consider the problem of allocating orthogonal resources in a self-organized manner on conflict graphs that describe discretized interference couplings of wireless networks. The target is to find a global optimum, i.e., a conflict-free allocation. We consider both random planar and nonplanar graphs that result from a path-loss model in a wireless network. Two algorithms for the self-organized coloring of arbitrary graphs are devised, and their performances are compared with each other and a rule-based reasoning algorithm that is known from the literature. The semigreedy distributed local search (SDLS) algorithm, which is a particularly simple algorithm that is proposed here, is shown to outperform other algorithms in several cases. In a cellular system setting, we consider a negotiated worst coupling procedure to symmetrize the interference coupling between cells, targeting an improvement of the channel quality of cell-edge users. We compare this approach with an interference coupling based on the path loss that was experienced between base stations and see significant gains in cell-edge performance. In addition, we compare the channel quality that was experienced by users in a cellular network that employs SDLS with the channel quality in a network that employs an interference-reducing network algorithm which finds a local optimum in terms of real-valued interference couplings. In some cases, attempting global optimization based on a conflict graph interpretation outperforms local real-valued optimization. |
Databáze: | OpenAIRE |
Externí odkaz: |
načítá se...