Manchester Metropolitan University's Research Repository

    IADE: an Improved Differential Evolution Algorithm to preserve sustainability in a 6G network

    Zhou, Zhou, Shojafar, Mohammad, Abawajy, Jemal and Bashir, Ali Kashif ORCID logoORCID: https://orcid.org/0000-0001-7595-2522 (2021) IADE: an Improved Differential Evolution Algorithm to preserve sustainability in a 6G network. IEEE Transactions on Green Communications and Networking, 5 (4). pp. 1747-1760. ISSN 2473-2400

    Accepted Version
    Download (2MB) | Preview


    Differential evolution (DE) algorithm is utilized to find an optimized solution in multidimensional real applications like 5G/6G networked devices and support unlimited connectivity for terrestrial networks due to high efficiency, robustness, and easy achievements. With the development of new emerging networks and the rise of big data, the DE algorithm encounters a series of challenges, such as the slow convergence rate in late iteration, strong parameter dependence, and easiness to fall into local optimum. These issues exponentially increase the energy and power consumption of communications and computing technologies in 5G/6G network like a networked data center. To address this and leverage a practical solution, this paper introduces IADE, an improved adaptive DE algorithm, to solve the problems mentioned earlier. IADE improves the scaling factor, crossover probability, variation, and selection strategy of the DE algorithm. In IADE, the parameters adaptively adjusted with the population's iterative evolution to meet the parameter's different requirements values of network steering traffic in each period. Numerous experiments are carried out through the benchmark function to evaluate the performance of IADE, and the results obtained from the experiment illustrate that IADE surpasses the benchmark algorithms in terms of solution accuracy and convergence speed for large tasks around 10%, respectively.

    Impact and Reach


    Activity Overview
    6 month trend
    6 month trend

    Additional statistics for this dataset are available via IRStats2.


    Repository staff only

    Edit record Edit record