Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II

Research Article

Allocation Optimization Based on Multi-population Genetic Algorithm for D2D Communications in Multi-services Scenario

Download
152 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-73447-7_4,
        author={Xujie Li and Xing Chen and Ying Sun and Ziya Wang and Chenming Li and Siyang Hua},
        title={Allocation Optimization Based on Multi-population Genetic Algorithm for D2D Communications in Multi-services Scenario},
        proceedings={Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II},
        proceedings_a={MLICOM},
        year={2018},
        month={2},
        keywords={D2D communications Cellular network Multi-population genetic algorithm Resource allocation},
        doi={10.1007/978-3-319-73447-7_4}
    }
    
  • Xujie Li
    Xing Chen
    Ying Sun
    Ziya Wang
    Chenming Li
    Siyang Hua
    Year: 2018
    Allocation Optimization Based on Multi-population Genetic Algorithm for D2D Communications in Multi-services Scenario
    MLICOM
    Springer
    DOI: 10.1007/978-3-319-73447-7_4
Xujie Li,*, Xing Chen1, Ying Sun1, Ziya Wang1, Chenming Li1, Siyang Hua2
  • 1: Hohai University
  • 2: Talent Science and Technology Co., Ltd.
*Contact email: lixujie@hhu.edu.cn

Abstract

For D2D Communications in Multi-services scenario, fast resource allocation optimization is a crucial issue. In this paper, a resource allocation optimization method based on the multi-population genetic algorithm for D2D communications in Multi-services scenario is proposed. Due to the interference between the cellular user equipment (CUE) and D2D user equipment (DUE) which share the same frequency, the complexity of resource allocation increases. Firstly, the interference model of D2D communications is analyzed. Then the resource allocation problem is formulated and discussed. Next, a resource allocation scheme based on Multi-population genetic algorithm is presented. Finally, the analysis and simulation results show the Multi-population genetic algorithm can converge faster compared with standard genetic algorithm. Therefore, the Multi-population genetic algorithm is more suitable to the Multi-services scenario where the data rate demand varies quickly and frequently.