Elsevier

Computer Communications

Volume 168, 15 February 2021, Pages 170-177
Computer Communications

5G heterogeneous network selection and resource allocation optimization based on cuckoo search algorithm

https://doi.org/10.1016/j.comcom.2020.12.026Get rights and content

Abstract

In order to solve the problem of spectrum resource shortage and high-speed access in 5G network, the multi-agent system is embedded into the standard cuckoo algorithm, and the multi-agent cuckoo algorithm is proposed. Firstly, the spectrum information of network channel sharing available to users is obtained, and the cuckoo search algorithm is used to optimize under the condition of satisfying the quality of service (QoS) guarantee of users, and the optimal allocation scheme is obtained by iterating for many times. The use steps are illustrated by examples. Compared with the traditional genetic algorithm, the calculation complexity can be reduced, and it can also be extended to more users and networks. In this algorithm, each cuckoo represents an agent, and all the agents constitute a von-Neumann structure. Through the neighborhood competition cooperation operator, mutation operator, self-learning operator and the evolution mechanism of cuckoo algorithm, they can continuously enhance the energy and improve the adaptability, and can quickly and accurately find the optimal solution of the problem.

Introduction

Resource balance optimization is a discrete optimization problem with high dimension, non-linear and multi constraints. Scholars at home and abroad have proposed many methods to solve this problem, which can be roughly divided into three categories: accurate algorithm, heuristic algorithm based on priority rules, and intelligent algorithm [1], [2], [3]. The results of precise algorithm are accurate but very time-consuming, which is difficult to apply to large and complex problems; heuristic algorithm based on optimization rules is fast but the quality of optimization results depends on the merits of the rules and the universality of the algorithm is poor. Intelligent algorithm is a kind of meta heuristic algorithm which simulates the physical phenomena or biological evolution process in nature [4], [5]. It has the characteristics of global, parallel, efficient and universal, and is a common algorithm to solve complex engineering optimization problems. 5G standard stipulates that 5G network peak transmission rate reaches 10 Gbit/s, traffic density reaches 10 TBPs / km2, user experience rate reaches 0.1–1 Gbit/s, and connection density reaches 1 million sets/km2. The end-to-end delay reaches ms level, which can guarantee the user experience at the speed of 500 km/h, and greatly improves the energy efficiency, cost efficiency and spectrum efficiency [6]. At present, a large number of relevant workers are committed to research and develop the application of 5G network in Unmanned Aerial Vehicle (UAV), logistics, vehicle network, etc., and there will be more 5G network application scenarios in the future.

Cloud computing is a new computing mode based on parallel computing, distributed computing, network computing and other technologies. It is the development direction of computer computing. Resource allocation is an important topic in cloud computing, and its efficiency has a great impact on the performance of the whole cloud computing. Because the resource allocation problem is a typical NP complete problem, the traditional optimization method is difficult to solve. Intelligent optimization algorithm has unique advantages in solving this kind of problem. 5G heterogeneous cellular network has the advantages of high capacity, deep coverage, low cost, high energy efficiency and load balancing, which is considered to be the most critical technology to achieve 1000 times capacity improvement of wireless communication network. In addition to the traditional spectrum resources (channel, bandwidth and power), the resources of 5G heterogeneous cellular network include base station access point (small cell base station and relay, etc.), buffer and RF antenna [7], [8], [9]. The resource management of 5G heterogeneous cellular network is to reasonably adjust and allocate the above resources according to the needs and distribution of user services, to achieve the best match between user service needs and wireless communication services, to improve the utilization efficiency of various network resources, and to reduce the consumption costs (power and cost costs, etc.) generated by heterogeneous networks in the service process [10], [11], [12]. 5G heterogeneous cellular network resource management is facing many problems and challenges, including performance index, spectrum and interference management, network topology, base station load management, link management, active cache management, multi antenna and precoding management. According to the decision structure, 5G heterogeneous cellular network resource management decision can be divided into centralized decision, cluster decision (semi centralized decision) and distributed decision.

The residual of the paper is organized as follows: Section 2 devote to discuss related works of 5G heterogeneous network selection and resource allocation optimization. 5G heterogeneous network and system model was expressed in Section 3. Section 4 described the implementation of multi-agent cuckoo algorithm for resource balance optimization. Experimental results were discussed and analyzed in detail in Section 5. Finally, Section 6 concluded the work and proposed the outlook.

Section snippets

Related work

In recent years, the issues of user association in heterogeneous networks, massive multiple-input multiple-output networks, millimeter wave networks, and energy harvesting networks has become a hot topic in recent years. Many scholars have made great achievements in the field of 5G heterogeneous network selection and resource allocation optimization. Fadel et al. formulate an optimization problem for Heterogeneous Networks multi-user selection in a multi-input-multi-output and orthogonal

5G network topology

An obvious difference between 5G heterogeneous cellular network and traditional macro cellular network is the network topology and cell coverage distribution. In the traditional macro cellular network, the base stations are evenly distributed, usually on the hexagon grid, and the service area corresponds to the hexagon in the grid [24]. In 5G heterogeneous cellular network, the small cell base station is irregularly distributed, and its service area is also uneven. As shown in Fig. 1, in

Standard cuckoo algorithm

In the algorithm, the feasible region is regarded as the nest for laying eggs, and the global optimal solution is regarded as the best host. The whole iterative optimization process simulates the process of finding the most ideal nest of cuckoo, and establishes the corresponding relationship between the problem solution set and cuckoo. Cuckoo algorithm is a natural heuristic algorithm developed by Xin She Yang and suash DEB in 2009. CS is based on the parasitic brooding behavior of cuckoo. In

Experiment and analysis

The hardware environment of this experiment is Intel Core i5, 2.3 GHz CPU, memory is 4 GB DDR3, the program is realized by MATLAB 2019. There are three sub groups with a total scale of 10 iterations 1000 times, generating 30000 scheduling times. Therefore, this paper sets a population scale of 30, the maximum number of iterations 1000, non-uniform variation Pa = 0.86, cross probability CR = 0.86, scaling factor F = 0.4, and compares the operation results of each algorithm under the same

Conclusion

In order to solve the problem of spectrum resource shortage and high-speed access in 5G network, this paper proposes a cuckoo search optimization method. Firstly, the spectrum information of network channel sharing available to users is obtained, and the cuckoo search algorithm is used to optimize under the condition of satisfying the QoS guarantee of users, and the optimal allocation scheme is obtained by iterating for many times. The use steps are illustrated by examples. Compared with the

CRediT authorship contribution statement

Ning Ai: Data analysis. Bin Wu: Formal analysis. Boyu Li: Validation. Zhipeng Zhao: Wrote the manuscript.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Research is supported by the National Key R\&D Program, China, supporting construction of national high performance computing environment operation management support platform for various scale operations (Grant No. 2016YFB0201403), the Natural Science Fund of China, new architecture of fat-tree data center batch scheduling in optical packet switching (NSFC project No. 61372085). It is also supported by Tianjin Key Laboratory of Advanced Networking (TANK), School of Computer Science and

References (28)

  • NiuJ. et al.

    Joint 3D beamforming and resource allocation for small cell wireless backhaul in hetnets

    IEEE Commun. Lett.

    (2017)
  • ZhangH.L. et al.

    Synthesis of heat exchanger networks based on modified cuckoo search algorithm

    Gao Xiao Hua Xue Gong Cheng Xue Bao/J. Chem. Eng. Chinese Univ.

    (2017)
  • NiknamS. et al.

    A multiband OFDMA heterogeneous network for millimeter wave 5G wireless applications

    IEEE Access

    (2016)
  • AzizM.A.E.

    Source localization using TDOA and FDOA measurements based on modified cuckoo search algorithm

    Wirel. Netw.

    (2017)
  • Cited by (22)

    • Artificial rabbits optimization: A new bio-inspired meta-heuristic algorithm for solving engineering optimization problems

      2022, Engineering Applications of Artificial Intelligence
      Citation Excerpt :

      As an illustration, one thorny question we are facing is the problem of spectrum resource shortage and high-speed access with the prosperity of 5G mobile communication network (Rasheed, 2022; Girjashankar and Upadhyaya, 2021; Israr et al., 2022). Through utilizing the spectrum information of network channel sharing to satisfy the quality of service (QoS) guarantee of users, the multi-agent system is embedded into an optimization method to determine the optimal allocation scheme by iteration (Ai et al., 2021). Another challenging optimization problem is how to manage training process and reduce communication cost in federated learning with the development of artificial intelligence (Hu et al., 2021; Rodríguez-Barroso et al., 2020; Chen et al., 2021).

    View all citing articles on Scopus
    View full text