Configurable Harris Hawks Optimisation for Application Placement in Space-Air-Ground Integrated Networks | IEEE Journals & Magazine | IEEE Xplore

Configurable Harris Hawks Optimisation for Application Placement in Space-Air-Ground Integrated Networks


Abstract:

Space-Air-Ground Integrated Network (SAGIN) has recently emerged as a viable solution for reliable transmission, high data rates, and seamless connectivity with extensive...Show More

Abstract:

Space-Air-Ground Integrated Network (SAGIN) has recently emerged as a viable solution for reliable transmission, high data rates, and seamless connectivity with extensive coverage. However, the characteristics of the computation and communication devices located at various levels of SAGIN make application placement within such environments a challenging task. Real-time service expectations and resource requirements of applications further intensify this issue, and push the domain to operate beyond its capacity, resulting in uneven delays and significant overhead. Taking these constraints into account, SAGIN’s application placement problem can be expressed as a multiobjective optimisation problem. This paper aims to solve such a problem using a Dynamic Weight-configurable Harris Hawks Optimisation (DW-HHO) algorithm, considering diverse application contexts such as deadlines, resource usage and the number of application activities. It simultaneously minimises application total service time and host resource overhead with a robust global search. The performance of the proposed solution is compared with benchmark metaheuristic solutions such as PSO, NSGA-II, Greedy and Random. Experimental results demonstrate that DW-HHO outperforms other benchmark metaheuristic solutions in optimising resource utilisation and service delivery time of applications in SAGIN environments. The proposed DW-HHO demonstrates notable improvements over existing methods. Specifically, when evaluating the total service time for PSO, NSGA-II, Greedy, and Random, DW-HHO outperforms these methods by 7.28%, 9.07%, 13.01%, and 14.97%, respectively.
Published in: IEEE Transactions on Network and Service Management ( Volume: 21, Issue: 2, April 2024)
Page(s): 1724 - 1736
Date of Publication: 01 February 2024

ISSN Information:

Funding Agency:


References

References is not available for this document.