Multiobjective planning of indoor Wireless Local Area Networks using subpermutation-based hybrid algorithms
Graphical abstract
Introduction
Over the past decade, people have become more dependent on wireless communications. The widespread adoption of Wireless Local Area Networks (WLANs) has been motivated by mobility, high data throughput, and low structural cost. Thus, the potential arising from the massive presence of access points (APs) has boosted the application of this technology, which goes beyond providing Internet connection for accessing Web pages, video streaming services, and social networks. In the 5G era, the IEEE 802.11 standard is becoming another access technology option to data offload, relieving the overhead in cellular networks. Such characteristics highlight the relevance of WLAN planning facing emerging demands.
WLAN planning is often driven to provide suitable signal coverage in some targeted areas. However, key aspects such as interference level, AP capacity, and distribution of customers in the covered area are frequently neglected, despite their importance. These factors become critical in applications that require stable and low-delay connections, such as videoconferencing, voice over IP, and multiplatform instant messaging. Nowadays, these requirements have become essential with the BYOD (Bring Your Own Device) trend, in which users demand high-performance wireless networks on their mobile devices practically all the time.
Overprovisioning APs, besides requiring more investment, also can lead to degradation of network performance, due to interference problems caused by the limited frequency spectrum available in WLANs [1]. Therefore, APs must be located to meet network coverage and client demands, while minimizing interference between the access points. In real situations, many users tend to agglomerate in the same area [2], [3]. Such a behavior, combined with the increase in demand for different services, creates an unbalanced load on the wireless LAN, which may compromise its overall performance and the quality of service. Again, a suitable AP placement, combined with an efficient client-AP association policy, is important to maximize the signal-to-noise ratio on the client devices and improve the load balance.
The aforementioned aspects make WLAN planning a hard optimization problem, in which multiple design criteria and technical constraints must be considered. The application of exact methods to solve this problem is not feasible, given their high computational cost [4], [5]. Thus, metaheuristics appear as alternatives to obtain reasonable solutions within acceptable computational time. Genetic Algorithms (GAs), for instance, have been used to solve WLAN related problems [6], [7], [8], [9], [10]. Furthermore, metaheuristics are scalable and can be adapted to deal with problems of different features and dimensions.
An innovative approach, based on hybrid algorithms, is proposed in this work for Wi-Fi network planning. Such an approach is scalable and combines two evolutionary algorithms and one greedy method to obtain efficient designs for large-scale WLANs, considering multi-floor buildings. The developed method uses a new efficient mechanism for representing and decoding solutions by breaking the original permutation into subpermutations. This structure guarantees solution feasibility and promotes a dimensional reduction by preventing the algorithm from performing unnecessary operations. Two design criteria are considered: maximizing load balance and signal-to-interference-and-noise (SINR) ratio. The generated solutions should satisfy constraints of minimum coverage, availability of channels, and bandwidth of the APs. Two test scenarios were considered, with one of them built with real data collected from a large ESS (Extended Service Set) WLAN already in operation. Finally, the solutions were submitted to a sensitivity analysis procedure to validate their robustness regarding user profile variations (locations and demands) and access point failures.
This paper is organized as follows. Section 2 discusses some related works. Section 3 overviews the problem of WLAN planning, and Section 4 addresses its mathematical model. Section 5 presents the proposed approach, Section 6 states test scenarios, and Section 7 evaluates the performance of the proposed method on those scenarios. Finally, Section 8 concludes the manuscript.
Section snippets
Brief review of WLAN planning-related works
Most WLAN planning works are limited to solve the wireless coverage problem and ignore key design aspects such as channel assignment and interference, traffic demand requirements, user density, and AP load balancing. In this way, Liu et al. [11] developed an approach to minimize the AP transmission power while guaranteeing network coverage through a hybrid swarm intelligence optimization algorithm. This work focus on energy-saving in a real multi-floor WLAN. However, this approach has some
Problem features
A thorough WLAN planning is a complex but necessary task. An efficient wireless LAN design reduces network structural costs and performance-related problems. However, the WLAN deployment goes beyond identifying user positioning and installing APs. Considerations such as interference, coverage, and user clusters must be considered. This last problem becomes complicated in large-scale WLANs, as the user mobility is increased through the handoff process. That mobility can generate load imbalances
Mathematical model
The goal of the problem addressed in this study is to find the best positioning and channel mapping for the APs, considering the coverage and demand requirements of users. The model input data are: the set of candidate points for installation of APs; the total number of clients to be served; the features of the network area to be covered (dimensions, losses, estimated noise, etc.), and technical features of APs (gain, reception threshold, bandwidth, etc.). Two conflicting
Proposed WLAN planning approach
This section describes the proposed WLAN planning tool. It receives as inputs the mapped service area (obstacles, boundaries, floors, etc.); the set of candidate points (CPs) for installation of APs; the estimated client positions/bandwidth consumption; and equipment specifications (protocol, antenna gain, transmission power, etc.).
The structure of the proposed WLAN planning approach is shown in Fig. 1. It is composed of five main steps, as follows:
- Pre-processing RSS Matrix:
in this step, the
Test scenarios and user profiles
This section describes the test scenarios and the experiment configuration parameters of this work.
Computational results
This section presents the results obtained by the proposed WLAN planning approach.
Conclusion
The mass adoption of WLANs by users has resulted in high-density scenarios, generating problems of load imbalance and interference in the network. Thus, efficient WLAN planning is key to reduce deployment costs and offer quality access to users. However, such a design is complex due to the characteristics inherent to the problem: stochastic, strongly constrained, and with costly and nonlinear objective functions. Factors like these make the IEEE 802.11 network planning an attractive topic for
CRediT authorship contribution statement
Marlon P. Lima: Conceptualization, Investigation, Experiment, Software, Writing – original draft. Ricardo H.C. Takahashi: Writing – review & editing, Supervision. Marcos A.M. Vieira: Writing – review & editing. Eduardo G. Carrano: Conceptualization, Methodology, Writing – review & editing, Supervision.
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
The authors would like to thank the Brazilian agencies CAPES, CNPq, and FAPEMIG for their financial support. We are also grateful to Ekahau, Inc. for providing a functional version of Ekahau Site Survey software for seven-day testing.
References (47)
- et al.
A multi-objective approach to indoor wireless heterogeneous networks planning based on biogeography-based optimization
Comput. Netw.
(2015) - et al.
Wireless LAN load balancing with genetic algorithms
- et al.
Integer linear programming formulations for the variable data rate and variable channel bandwidth scheduling problem in wireless networks
Comput. Netw.
(2019) - et al.
Ethanol: A software-defined wireless networking architecture for IEEE 802.11 networks
Comput. Commun.
(2020) - et al.
Evaluation of throughput estimation models and algorithms for WLAN frequency planning
Comput. Netw.
(2007) - et al.
Analysis of the increase and decrease algorithms for congestion avoidance in computer networks
Comput. Netw. ISDN Syst.
(1989) - et al.
Centralized channel allocation algorithm for IEEE 802.11 networks
- et al.
Characterizing user behavior in a European academic WiFi network
Int. J. Handheld Comput. Res.
(2013) - et al.
DAPA: Capacity optimization in wireless networks through a combined design of density of access points and partially overlapped channel allocation
IEEE Trans. Veh. Technol.
(2016) - et al.
A two-stage approach to WLAN planning: Detailed performance evaluation along the Pareto frontier
Hybrid multicriteria algorithms applied to structural design of wireless local area networks
Appl. Intell.
A WLAN planning proposal through computational intelligence and genetic algorithms hybrid approach
Optimizing access point allocation using genetic algorithmic approach for smart home enviroments
J. Comput.
Multiobjective planning of wireless local area networks (WLAN) using genetic algorithms
Design and implementation of decisioning system for wifi access point installation using genetic algorithm
Toward the energy-saving optimization of WLAN deployment in real 3-D environment: A hybrid swarm intelligent method
IEEE Syst. J.
Optimization model for an indoor WLAN-based positioning system
Wireless access point positioning optimization
Cell Breathing Techniques for Balancing the Access Point Load in Wireless LANsTechnical report
Design of Indoor WLANs: Combination of a ray-tracing tool with the BPSO method
IEEE Antennas Propag. Mag.
Survivable mechanism for IEEE 802.11 WLAN improvements
Lecture Notes in Comput. Sci.
Network planning for indoor joint LTE and WLAN networks
WLAN planning: Separate and joint optimization of both access point placement and channel assignment
Ann. Telecommun. - Ann. Télécommun.
Cited by (3)
Bio-inspired algorithm for decisioning wireless access point installation
2023, International Journal of Electrical and Computer EngineeringEnergy-efficient resource allocation model for device-to-device communication in 5G wireless personal area networks
2023, International Journal of Communication SystemsSmart Load-Based Resource Optimization Model to Enhance the Performance of Device-to-Device Communication in 5G-WPAN
2023, Electronics (Switzerland)