기본 콘텐츠로 건너뛰기

K-Means Based Bee Colony Optimization for Clustering in Heterogeneous Sensor Network

 

K-Means Based Bee Colony Optimization for Clustering in Heterogeneous Sensor Network

Article

K-Means Based Bee Colony Optimization for Clustering in Heterogeneous Sensor Network

by  1,2,*, 3, 2,*, 2, 2, 4 and 5
1
Department of Computer Science, Ho Technical University, Ho VH-0044, Ghana
2
Department of Computer Science, Ghana Communication Technology University, Accra PMB 100, Ghana
3
Department of Computer Science, Kwame Nkrumah University of Science and Technology, Kumasi K384, Ghana
4
Faculty of Accounting Durban, University of Technology, P.O. Box 1334, Durban 4000, South Africa
5
ICT and Society Research Group, Department of Information Technology, Durban University of Technology, P.O. Box 1334, Durban 4000, South Africa
*
Authors to whom correspondence should be addressed.
Sensors 202424(23), 7603; https://doi.org/10.3390/s24237603
Submission received: 16 October 2024 / Revised: 22 November 2024 / Accepted: 26 November 2024 / Published: 28 November 2024
(This article belongs to the Special Issue Advanced Applications of WSNs and the IoT—2nd Edition)

Abstract

In Wireless Sensor Networks (WSNs), an efficient clustering technique is critical in optimizing the energy level of networked sensors and prolonging the network lifetime. While the traditional bee colony optimization technique has been widely used as a clustering technique in WSN, it mostly suffers from energy efficiency and network performance. This study proposes a Bee Colony Optimization that synergistically combines K-mean algorithms (referred to as K-BCO) for efficient clustering in heterogeneous sensor networks. This is to develop a robust and efficient clustering algorithm that addresses the challenges of energy consumption and network performance in WSNs. The K-BCO algorithm outperformed comparative clustering algorithms such as H-LEACH, DBCP, and ABC-ACO in average error rate (AER), average data delivery rate (ADDR), and average energy consumption (AEC) for transmitting data packets from sensors to cluster heads. The K-BCO outperformed other algorithms in terms of ADDR at 95.00% against H-LEACH (75.86%), DBCP (72.07%) and ABC-ACO (90.08%). The findings indicate that the K-BCO not only optimizes energy consumption but also guarantees more stable and robust solutions, thereby extending the network lifetime of WSNs. Thus, K-BCO is recommended to practitioners in wireless sensor networks as it paves the way for more efficient and sustainable wireless communication.

1. Introduction

Wireless Sensor Network (WSN) comprises independent sensors deployed across a geographical area to collect and share data or information through wireless communication channels [1]. These sensors are deployed to observe physical or environmental properties like vibration, pressure, temperature, sound, motion, or pollutants. The sensors can collaborate to transmit data to a central point within the network known as the sink node (SN) or base station (BS) [2]. Wireless Sensor Networks (WSNs) provide many benefits in various applications, such as environmental monitoring, disaster management, healthcare, agriculture, and smart city infrastructure. This makes them valuable technology in different fields, including real-time data collection, enhanced monitoring and control, increased safety and security, and innovative applications [1,3]. Although WSNs are valuable in these areas, their successful development and deployment are faced with many challenges, including power consumption and packet loss. Sensors are typically battery-powered, and energy is a critical resource. Once a sensor node’s battery is depleted, it can no longer function, leading to gaps in data collection and potentially compromising the entire network. In WSN, sensors may have different energy requirements. As the number of nodes increases, the volume of data transmitted can lead to network congestion and increased energy consumption. Efficient data routing and transmission strategies are essential to mitigate these issues. The lifetime of a WSN is often limited by the energy of its nodes. Therefore, strategies that prolong the operational time of the network are crucial for maintaining continuous monitoring and data collection. However, the challenge is selecting a sensor that has enough energy to be the Cluster Head (CH).
Clustering in WSNs provides a structured technique for WSN organization, resource management, and communication optimization, leading to improved scalability, efficiency, reliability, and performance of wireless sensors [2,4]. Clustering is a method used to group sensor nodes into clusters, where each cluster has a designated leader known as the cluster head (CH). This approach reduces the amount of data that needs to be transmitted to the sink node, as data from individual nodes can be aggregated at the cluster head before being sent. While clustering can enhance energy efficiency and data management, traditional methods may not effectively adapt to the dynamic nature of WSNs, especially when considering the varying energy levels and capabilities of different sensor nodes. Again, it helps WSNs to optimize energy consumption through the Cluster Head, Duty Rotation, Data Aggregation, Efficient Routing, Sleep Scheduling, and Localized Processing [2].
Nature-inspired optimization methods are one of the approaches to solving wireless sensor network problems in terms of finding the optimal approach to cluster formation. Nature-inspired approaches help avoid unpromising search paths in any optimization problem, thus helping reduce the time spent searching for the best or near-optimal solutions. Examples of such methods include Bee Colony Optimization (BCO), which is a nature-inspired optimization algorithm based on honeybees’ foraging behavior [5,6]. BCO is a metaheuristic optimization algorithm inspired by the foraging behavior of honeybees. It mimics how bees search for food and communicate the location of food sources to one another. BCO can be used to optimize routing and clustering by efficiently selecting cluster heads based on factors like energy levels and node proximity. The foraging behavior of honeybees could be applied to find the most efficient routes for data transmission in a WSN, thereby reducing energy consumption, improving network lifetime, and enhancing the overall network performance [3].
Unfortunately, the traditional Bee Colony Optimization (BCO) faces challenges such as premature convergence, where the algorithm settles on suboptimal solutions too quickly, limiting its effectiveness in complex scenarios [5]. This also limits its performance in finding the optimal routing paths in heterogeneous sensor networks. This is because there is diversity and variability among the sensor nodes within the network. This study focuses on developing an efficient clustering algorithm by combining the k-mean clustering technique with the Bee Colony Optimization algorithm (K-BCO) to address the challenge associated with the traditional BCO technique. K-means is a popular clustering algorithm that partitions data into K clusters based on similarity [7]. It is straightforward and efficient but can struggle with dynamic environments and may not account for the varying energy levels of sensor nodes in WSNs.
In a heterogeneous WSN, sensors may have varying hardware specifications, memory capacity, processing power, and energy resources [8]. For instance, some sensor nodes might have strong batteries or an efficient sensing capability, while others may be more limited. This variation affects how sensor nodes participate in data collection, processing, and communication. Different sensor nodes may generate data at varying rates depending on their sensing capabilities and the environmental conditions they monitor. This variability can lead to uneven data traffic within the network, complicating the clustering and data aggregation processes. The proposed K-BCO algorithm aims to account for these differences to enhance data collection efficiency. The study also considers the importance of energy efficiency in WSNs. Heterogeneity in energy levels among sensor nodes means that some nodes may have more energy available for data transmission and processing than others. The K-BCO algorithm is designed to select cluster heads based on their energy status, ensuring that nodes with sufficient energy are chosen to lead data transmission. This approach helps to balance energy consumption across the network, prolonging the overall network lifetime. The physical arrangement of sensors can change over time, especially in situations where mobile sensors are involved. The K-BCO algorithm’s ability to dynamically adjust cluster formations helps maintain connectivity and communication efficiency in such environments. WSNs have unique requirements regarding data accuracy, latency, and reliability. The heterogeneity of application needs means that a one-size-fits-all clustering approach may be ineffective. The K-BCO algorithm is designed to be adaptable, allowing it to cater to the specific demands of various applications while optimizing energy consumption and network performance.
The sections of this paper are organized as follows: Section 2 (related works), Section 3 (theoretical framework), Section 4 (proposed clustering technique), Section 5 (experimental settings of WSN), Section 6 (simulation results), Section 7 (discussion of results), Section 8 (conclusions).

2. Related Works

Jacob I. and Darney P. (2021) [9] developed the Artificial Bee Colony (ABC) Optimization Algorithm for enhancing routing in wireless networks. The Artificial Bee Colony (ABC) Optimization aimed at improving the performance and efficiency of wireless communication systems by applying nature-inspired optimization techniques. Comparative analysis of network performance was performed using performance metrics such as packet delivery ratio, latency, and energy efficiency between the ABC algorithm and existing models. The study underscores the significance of nature-inspired optimization algorithms, particularly the ABC algorithm, in improving routing efficiency, reducing interference, and enhancing overall network performance in wireless communication systems. It stated potential applications of the developed algorithm in addressing complex challenges and optimizing wireless networks for various technological advancements. Although the study enhanced the performance of the entire network, the proposed scheme was unable to select optimal paths based on dynamic network conditions and traffic patterns.
Dawood et al. (2021) [10] also proposed an algorithm for energy-efficient clustering in heterogeneous wireless sensor networks. The algorithm was developed to improve energy efficiency in terms of intra-cluster communication by considering distance and energy factors. Their study aimed at reducing energy consumption within clusters by minimizing communication distances between nodes and cluster heads. The key components of the proposed algorithm include cluster formation, cluster head selection based on proximity and energy levels, and optimizing intra-cluster communication to reduce energy consumption. They presented simulation results, comparing the performance of the proposed protocol with the PEGASIS protocol in terms of PDR, average end-to-end delay, and energy consumption. The proposed algorithm outperformed the PEGASIS protocol in terms of lifetime, energy efficiency, and average end-to-end delay, thereby improving the lifetime of the wireless sensor network. Also, using proximity to the center of the zone and maximum residual energy as criteria for CH selection may affect the overall performance of the network since a faulty sensor may have more energy. This may disrupt effective data collection from some segments of the field. The proposed algorithm did not consider mobile sensors within the network, hence raising the scalability issue.
Another study by Agbehadji et al. (2021) [8] developed a clustering algorithm based on a nature-inspired approach for energy optimization in a heterogeneous wireless sensor network. Their approach focused on the heterogeneity of devices, energy requirements of nodes, and network characteristics. The performance of the proposed scheme was measured based on the number of data packets successfully received at the base station; data transmission and reception were also efficient in the network. In [8], performance was compared with existing models such as Distributed Energy-Efficient Clustering (DEEC), Developed Distributed Energy-Efficient Clustering (DDEEC), and Extended version of Distributed Energy-Efficient Clustering (EDEEC).
Jacob I. and Darney P. (2021) [9] proposed the Artificial Bee Colony (ABC) Optimization Algorithm as a method for enhancing routing in wireless networks. This algorithm is inspired by the foraging behavior of honeybees and is designed to optimize various performance metrics in wireless communication systems. The study aimed to provide a novel approach to routing that enhances efficiency and performance through the use of bio-inspired algorithms. Their algorithm initializes a population of artificial bees, which represent potential solutions (routing paths) in the network. Each bee is assigned a specific route to explore based on the network model. The behavior of the bees was simulated to mimic the natural foraging process. Each bee evaluates the quality of its assigned route based on performance metrics such as throughput, delay, and energy consumption. The bees communicate information about the quality of their routes to other bees, similar to how real bees share information about food sources. The bees assess the performance of their routes using predefined performance indicators. This evaluation helps determine which routes are more efficient and should be prioritized for data transmission. Based on the evaluations, the bees select the best routes to follow. The algorithm employs a mechanism to exploit the best-known routes while also exploring new paths to avoid local optima. If a bee finds a better route than its current one, it updates its position to this new route. They demonstrated that once the optimal paths are determined, data packets are transmitted along these routes. The algorithm ensures that packets are spread across multiple paths to enhance reliability and reduce the risk of packet loss due to interference or congestion. The proposed scheme is validated through simulations (e.g., using MATLAB), where various scenarios are tested to evaluate the performance improvements achieved by the ABC algorithm compared to existing routing protocols. While the proposed ABC optimization scheme presents a promising approach to enhancing routing in wireless networks, it lacks some capabilities, such as fast convergence rates, local optima, network scalability, and adaptability to changing network conditions. Addressing these limitations through further research and refinement could improve the algorithm’s robustness and applicability in diverse network scenarios.
Zhu et al. (2024) [11], in their study, proposed a Random Dual Strategy Artificial Bee Colony (RDABC) algorithm to address the challenges of low coverage accuracy and slow convergence speed in wireless sensor networks (WSNs). The RDABC focuses on enhancing the exploration and exploitation capabilities of the optimization process, thereby improving overall coverage and efficiency in WSNs. The RDABC randomly deploys sensor nodes within the designated field or geographical area for the coverage optimization task to begin. RDABC uses two search strategies: Classical Search Strategy and Elite Search Strategy. The update equation for the classical search, which was based on the original Artificial Bee Colony (ABC) algorithm, is computed in Equation (1) as:
Vt=Xt+(XtXk)
where Vt is the new position of the bee, Xt is the current position, Xk is a randomly selected individual from the population, and  is a random parameter that influences the search direction.
This strategy focuses on the best solutions found so far. The update equation for the elite search is computed in Equation (2) as:
Vt=XEt+(XbestXk)
where XE is the central position of all elite solutions (top 10% of the population) and Xbest is the best individual in the population.
The RDABC incorporated a random selection method to switch between the classical and elite search strategies, allowing for flexible transitions and balancing exploration and exploitation. RDABC introduced a cross-mutation strategy derived from genetic algorithms. This strategy allows for multi-dimensional updates, which helps in exploring the solution space more effectively. The specific implementation details of the cross-mutation are not explicitly defined in the provided text, but it generally involves combining features from multiple solutions to create new candidate solutions. The RDABC iterates through the steps until a stopping criterion is met, such as a maximum number of iterations or a satisfactory level of coverage. The final output includes the optimized positions of the sensor nodes and the achieved coverage area. Although RDABC offers significant improvements over traditional methods, it may be faced with challenges like implementation complexity due to the dual search strategies and cross-mutation and being trapped in local optima, particularly in highly complex or multimodal optimization landscapes.
The Wolf Search Algorithm (WSA) operates based on the natural behaviors of wolves during hunting, incorporating principles of individual searching, memory utilization, and threat response [12]. WSA simulates the hunting behavior of wolves, where each wolf acts as an independent searching agent. The algorithm balances exploration (searching new areas) and exploitation (refining known good areas) to find optimal solutions.
Each wolf evaluates its current position and the positions of its peers within a defined visual range. The movement of a wolf is influenced by the quality of its position and the best position found by its peers; this is computed in Equation (3) as:
xnew=xold+·xbestxold+β·escape()
where xnew is the new position of the wolf, xold is the current position of the wolf.
xbest is the best position found by the wolf or its peers,  is a coefficient that controls the attraction towards the best position, β is a coefficient that controls the randomness of the escape movement, escape() is a function that generates a random position to jump to, simulating the wolf’s response to threats.
When a wolf encounters a threat (simulated as a random event), it jumps away from its current position to avoid being trapped in local optima. This is akin to a mutation in genetic algorithms. The escape movement is defined in Equation (4) as:
xscape=xold+random(d,d)
where d is a predefined distance that determines how far the wolf can jump away from its current position, random (−dd) generates a random value within the range of −d to d.
WSA incorporates a memory mechanism where wolves remember previously visited positions. This memory helps prevent revisiting the same positions and enhances search efficiency. Each wolf evaluates the quality of its current position based on a fitness function, which is problem-specific. The fitness function determines how close the current position is to the optimal solution. The fitness function f(x) is defined in Equation (5) as:
fx=Objective Function(x)
The WSA follows the general steps outlined below:
Step 1: Initialize a population of wolves with random positions.
Step 2: Evaluate the fitness of each wolf’s position.
Step 3: For each wolf, determine the best position found by itself and its peers.
Step 4: Update the position of each wolf using the movement formula.
Step 5: If a threat is encountered, apply the escape movement.
Step 6: Repeat the evaluation and updating process for a set number of iterations or until convergence criteria are met.
Rami Reddy et al. (2023) [13] proposed a scheme based on the improved Grey Wolf Optimization (EECHIGWO) algorithm to enhance energy efficiency, average throughput, network stability, and overall network lifetime in wireless sensor networks (WSNs) through optimal selection of cluster heads (CHs). The EECHIGWO specifically focused on addressing the challenges associated with traditional clustering methods, such as premature convergence, lack of population diversity, and the imbalance between exploration and exploitation in the optimization process. The EECHIGWO incorporated elements of the wolf search algorithm (WSA) to improve the efficiency and effectiveness of cluster head selection in WSNs, leading to enhanced network performance and longevity. The proposed EECHIGWO defines a fitness function that evaluates the potential CHs based on criteria such as residual energy, distance to the base station, and intra-cluster distance. The WSA uses this fitness function to guide the selection process, ensuring that the chosen CHs are optimal for energy efficiency and network stability. The EECHIGWO algorithm demonstrated a substantial improvement in network lifetime, with reported enhancements of up to 333.51% compared to several existing protocols, such as SSMOECHS, FGWSTERP, and LEACH-PRO. Although the EECHIGWO aims to balance exploration and exploitation, there may still be scenarios where it could get trapped in local optima, especially in highly complex or non-linear search spaces. The EECHIGWO may also introduce additional complexity in terms of implementation and parameter tuning.
Jiang et al. (2016) [14] developed an underwater sensor network redeployment algorithm based on the wolf search algorithm (RAWS) to optimize how sensors are deployed in underwater environments. The focus of RAWS was to address challenges like inefficient network coverage, high computational complexity, energy inefficiency, obstacles, and environmental changes affecting underwater sensor networks’ performance. The RAWS allows each node to perform an independent search for monitoring purposes. This does not enable sensors to exchange information with one another, which reduces communication overhead and computational complexity. Sensors move towards targets when detected or explore freely when no targets are present. The proposed RAWS included an escape mechanism that allows sensors to change their direction randomly when they encounter an obstacle to avoid it. This feature helps prevent sensor node failure and allows the algorithm to escape local optima during the search process. The node coverage for each node si, which was computed in Equation (6) as:
Neventsi=1nfpj,si0
where, fpj,si is a function that determines if the target pj is within the sensing range of the node si.
The performance of the proposed RAWS was evaluated through simulations. The RAWS results demonstrated improvements in network coverage, energy conservation, and obstacle avoidance compared to other existing algorithms, such as the artificial fish swarm algorithm. Although the proposed RAWS has demonstrated some improvement in network performance, it may be sensitive to the initial random deployment of nodes. If nodes are poorly positioned initially, it may take longer for the algorithm to achieve optimal coverage, or it may converge to a suboptimal solution. Also, the escape mechanism employed by RAWS relies on the ability of nodes to detect obstacles. If the detection range is limited or if obstacles are not accurately represented in the model, nodes may fail to avoid them effectively, leading to potential collisions or failures.

3. Theoretical Framework

3.1. Bee Colony Optimization Algorithm

The Bee Colony Optimization (BCO) algorithm is a nature-inspired optimization scheme based on the foraging behavior of honeybee colonies [15]. BCO imitates the food-foraging behavior of honeybees to find solutions to complex optimization problems. Honeybee colonies show incredible efficiency in searching for food sources in their immediate environment and embark on complex foraging behavior to locate, collect, and bring food (nectar) to the hive [13].
The honeybee communicates via dance movements and performs a distinctive “waggle dance”, signaling or sending information about the location of food sources to other bees in the colony [5]. Honeybees share details like the distance, direction, and quality of food sources through these dance movements, letting other bees optimize their foraging routes and efficiently gather nectar. Honeybee colonies collectively work to gather nectar from multiple sources of food [6]. The collaborative activities of individual honeybees contribute to the overall success of the colony in getting enough food for survival and growth.
Like honeybee colonies, the BCO algorithm focuses on an efficient search for high-quality solutions by coordinating the search behavior of artificial bees [6]. The BCO algorithm’s iterative nature and adaptive search technique make it a strong optimization strategy for solving complex optimization problems. The BCO algorithm comprises two main phases, the forward pass and the backward pass, which can be described as follows:

3.1.1. Forward Pass

At this stage, every honeybee finds a solution by exploring the search space around its colony. The quality of the solution found by a honeybee affects the probability of that honeybee’s continuous independent search or following another honeybee [5]. The forward pass can be expressed in Equation (7) as:
1.
Probability of Selecting a Neighbor Solution:
Pij=f(Sij)k=1nf(Sik)
where, Pij is the probability that a honeybee ‘i’ selects a neighbor solution Sij during the exploration phase. f() evaluates the quality of a solution. Thus, a lower fitness value shows a better solution to minimize a problem. f(Sij) calculates the fitness of the neighbor solution under consideration. k=1nf(Sik) is the sum of fitness values of all neighbor solutions available to honeybee i. Dividing the fitness of the current neighbor solution by the sum of the fitness values of all neighbors determines the probability of a honeybee i choosing a solution Sij for further exploration.
From Equation (4), each honeybee examines the quality of neighbor solutions based on a fitness function fSij is the probability of selecting a neighbor solution, which is computed as the ratio of the fitness of the neighbor solution to the sum of the fitness values of all neighbor solutions. This probability determines the likelihood of a honeybee choosing a particular neighbor solution during a search [15].
2.
Local Search Equation
After selecting a neighbor solution, the honeybee performs a local search around that solution, which can be expressed in Equation (8) as follows:
Sij=Sij+ij×SijSkj
where Sij is the new solution, updated by moving toward the neighbor solution Skj which is based on a random factor ij. This local search method allows honeybees to scout the search environment more effectively and potentially discover better solutions [15].
This study introduces the Wolves Search Algorithm (WSA) into the local search operation of honeybees as the coefficient vector. The coefficient vector represents a set of parameters that influences the movement and decision-making processes of the wolves during the search for optimal solutions [16]. Again, it allows each wolf to balance their search in new solution space and then refine the search around known good solutions. This ensures the adaptability of search in non-explored areas. The coefficient vector A is calculated based on factor a, which decreases linearly from 2 to 0 throughout iterations [17]. This implies that A changes its values at each time step, influencing the wolves’ movement towards the prey. Hence, given Equation (9):
A=2·a·r1a
where a is a linearly decreasing parameter from 2 to 0 over the iterations, r1 is a random vector in the range [0, 1]. Therefore, the honeybees’ local search formula is now computed in Equation (10) as:
Sij(new)=Sij(current)+Aij×Sij(best)Skj(current)
where, Sij(current) is the current value of j in the solution SiSij(best) is the corresponding j value from the best solution Sk in the population. Aij is the coefficient vector derived from the WSA, which replaces the random factor ij. The ability of WSA to evaluate positions based on the quality of solutions found by peers makes it suitable for optimizing routing paths and clustering in WSNs, where finding optimal solutions is crucial for performance and energy efficiency. The combination of the Bee Colony Optimization (BCO) algorithm with the Wave Search Algorithm (WSA) can enhance the performance of optimization tasks in Wireless Sensor Networks (WSNs) as follows:
i.
The BCO algorithm is known for its effective exploration capabilities due to its nature-inspired approach based on the foraging behavior of honeybees. However, it can sometimes suffer from premature convergence to suboptimal solutions. The WSA, on the other hand, is designed to explore the search space more thoroughly by simulating wave propagation. By combining these two algorithms, the strengths of both can be leveraged, leading to improved exploration and exploitation of the solution space.
ii.
The integration of WSA can help accelerate the convergence of the optimization process. While BCO may take longer to find optimal solutions due to its reliance on the collective behavior of bees, the wave propagation mechanism in WSA can guide the search process more efficiently, allowing for faster identification of high-quality solutions.
iii.
One of the challenges in optimization problems is the tendency of algorithms to get trapped in local optima. The combination of BCO and WSA can enhance robustness against this issue. The wave search mechanism can introduce diversity in the search process, helping to escape local optima and explore new areas of the solution space.
iv.
In WSNs, the environment can be dynamic, with changing conditions and requirements. The hybrid approach can provide greater adaptability, as the wave search can quickly adjust to changes in the search landscape, while BCO can maintain a balance between exploration and exploitation.
v.
In WSNs application, energy efficiency is critical; combining BCO with WSA can lead to more effective resource management. The hybrid algorithm can optimize the selection of cluster heads and data transmission paths, ensuring that energy consumption is minimized while maintaining high levels of data delivery and network performance.
vi.
The combination of BCO and WSA can be applied to a wide range of optimization problems beyond WSNs, including routing, scheduling, and resource allocation in various fields. This versatility makes the hybrid approach a valuable tool for researchers and practitioners looking to solve complex optimization challenges.

3.1.2. Backward Pass

Bees return to the hive after exploring the search space and comparing their solutions to further adjust their search strategies to avoid getting trapped in local optima. Thus, the Backward Pass Equations are described as:
1.
Comparison of Solutions
The comparison solution can be expressed in Equation (11) as:
CompareSij,Skj=1, if fSij<fSjk0, otherwise
The backward pass compares the current solution Sij found by honeybees with the neighbor solution Sjk concerning their fitness values. If the fitness of the neighbor solution is good (lower in the case of minimization problems), the honeybee updates its solution to move towards a better solution. The comparison process helps honeybees adjust their search direction based on the quality of solutions found by other bees.
2.
Update Solution
The update strategy is defined in Equation (12) as:
Sij=Sij+Sij×CompareSij,Skj
where the Honeybee updates its solution Sij by integrating a change Sij based on the comparison result. If the neighbor’s solution Skj is better, the honeybee adjusts its solution toward the neighbor’s solution. This update mechanism allows bees to learn from the collective search behavior and improve their solutions iteratively.

3.2. BCO Algorithm Workflow

Step 1: Bees are initialized with random solutions in the search space.
Step 2: Bees probabilistically explore the search space, select neighbor solutions, and perform a local search. This is known as the Forward Pass.
Step 3: Bees perform a Backward Pass. Thus, bees compare their solutions with neighbors, update their solutions based on comparison results, and share information.
Step 4: The forward and backward passes are repeated a specified number of times or until a stopping criterion is met.
Step 5: To determine their quality, solutions are evaluated based on an objective function.
Step 6: The algorithm aims at converging towards optimal solutions by repeatedly improving the solutions through exploration and exploitation.

3.3. K-Means Algorithm

The K-means algorithm is a clustering method in the field of data mining and machine learning [18]. The K-means algorithm was introduced by Stuart Lloyd in 1957 as a technique for pulse-code modulation (PCM) quantization in signal processing [18,19]. It was later rediscovered by MacQueen, who, in 1967, introduced the algorithm as a clustering technique in the context of pattern recognition. The K-means algorithm was initially applied in signal processing and image compression, which have proven to be very effective. It is also known for its simplicity, efficiency, and effectiveness in partitioning data into clusters based on similarity [20].
Integrating the K-means algorithm into the Bee Colony Optimization (BCO)) helped to solve complex optimization problems that required clustering [21,22]. The K-means clustering method used an iterative procedure to assign data points to clusters based on centroid proximity and update centroids until convergence. Although the K-mean is effective for many applications, it has some limitations relating to initialization, cluster shape assumptions, and scalability [2].
The K-Means clustering algorithm or scheme is a well-known partitioning technique that aims to divide n data points into K clusters. The scheme repeatedly assigns data points to the nearest cluster centroid (cluster heads) and updates the centroids based on the mean of the data points in each cluster.
The following steps constitute the process involved in K-Means clustering:
Step 1: Initialization
  • Select the number of clusters, K.
  • Randomly initialize K cluster centroids, C=c1,c2,,cK.
Step 2: Assignment
  • For every data point xi, calculate the distance to each centroid cj using a distance metric (e.g., Euclidean distance).
  • Assign the data point xi to the cluster with the nearest centroid: argminj=xicj2.
Step 3: Update
  • Update the cluster centroids by calculating the mean of all data points assigned to each cluster: cj=1SjxiSjxi where Sj is the set of data points assigned to cluster j.
Step 4: Convergence
  • Iterate the Assignment and Update steps until convergence criteria are met. Convergence can be based on the change in cluster assignments or centroid positions.
Step 5: Final Result
  • The algorithm converges when the cluster assignments stabilize and the centroids no longer change significantly. The final result is a set of K clusters with their respective centroids.
K-Means clustering technique aims to minimize the within-cluster sum of squares, where the objective function is: J=i=1KxSixci2 where Si is the set of data points assigned to cluster i and ci is the centroid of cluster i. Updating cluster assignments and centroids repeatedly, K-Means seeks to find a partition of the data that minimizes the total within-cluster variance. The iterative nature of the K-mean algorithm allows for efficient assignment and updating of clusters, making it effective for organizing nodes in WSNs. However, K-means has limitations, such as sensitivity to initialization and assumptions about cluster shapes. Therefore, by integrating K-means with BCO, the study will leverage its strengths while addressing its weaknesses, particularly in terms of energy consumption and network lifetime.

4. Proposed Clustering Technique in Wireless Sensor Networks (WSNs)

The proposed clustering technique combines the forward and backward passes repeatedly to generate the optimal solutions. In this regard, the sensor nodes (SN) perform the forward and backward passes repeatedly for a specified number of times to locate a cluster head (CH) with the optimal distance. One advantage of the proposed clustering technique is its ability to search all available nodes and then optimize energy in transmitting data packets between sensor nodes (SN) and cluster heads (CH) and among CHs. Optimizing the energy takes into consideration the distance to communicate in routing data packets to the sink node or base station through another CH.
The proposed scheme combines WSA, K-means, and Bee Colony Optimization to optimize the performance of WSNs. The integration of K-means with BCO (K-BCO) aims to enhance the clustering process, improve energy efficiency, and extend the network’s operational lifetime. The combination of WSA, K-means, and BCO is because there is a need to develop a robust and efficient clustering algorithm that addresses the challenges of energy consumption and network performance in WSNs. This integrated approach aims to provide more stable and optimal solutions, ultimately leading to improved network lifetime and efficiency.

4.1. Cluster Formation and Cluster Head Selection

Clustering enhances routing in WSNs by improving energy efficiency, facilitating data aggregation, and supporting the development of efficient routing protocols tailored for hierarchical network structures [1,8]. This routing technique models the food search method by honey bees, where SNs find the optimal distance to a CH. CHs also use optimal distance to communicate and route data packets to the sink node or base station through another CH. In this regard, the forward and backward passes are performed iteratively to locate a CH with the optimal distance. The following subsections describe the cluster formation and bit error determination stages.
During this process, the energy model and Bit Error Rate (BER) determination are considered in cluster formation and CH selection at every given time. These are described as follows:
  • Energy model
The study adopts the energy model by [10] to determine the energy requirement for a sensor or CH to send n-bit of a data packet over a distance d. Hence, the transmission energy is defined in Equation (13) as:
Etxn,d=Extelecm+Etxampt(n,d)
where Extelecm represents the energy consumed for electronic transmission and Etxampt(n,d) is the energy consumed for amplification during transmission. Hence, the energy consumption f(x) is expressed in Equation (14) as:
fx=n×Eelec+(n×Efs×d2), d<d0n×Eelec+(n×Eamp×d4), dd0
where fx is the energy consumption for sending a message over a certain distance in the network. Where Eelec represents the energy consumed for the electronic transmission, Efs represents the energy required for signal amplification, d is the distance over which the message is being transmitted and Eamp represents the energy consumed for signal amplification. Furthermore, the energy consumed for receiving the message from a sender is computed in Equation (15) as:
Erxn=n×Eelec
The residual energy (Eres) of a sensor node is calculated using Equation (16):
Eres=EinEtxn,d+Erx(n)
where, Ein is the initial energy of the sensor node, Etxn,d  is the transmission energy calculated earlier, and Erx(n) is the energy consumed for receiving the message.
The study utilized the Euclidean distance formula to compute the distance d between SN and CH and between CHs and the sink node or the base station (BS). The Euclidean distance formula is a fundamental concept in mathematics that calculates the straight-line distance between two points in Euclidean space [23,24]. In a two-dimensional space (such as a Cartesian plane), the Euclidean distance d between two points x1,y1 and x2,y2 is expressed in Equation (17):
d=x2x12+y2y12
1.
Bit Error Rate (BER) Determination
The Bit Error Rate (BER) is a measure of the percentage or fraction of bits in a data stream that has been corrupted during transmission [25,26]. It represents the ratio of the number of erroneous bits received to the total number of bits transmitted over a specific period, which is expressed in Equation (18).
BER=BtnerBtnt
where Btnt represents the total number of bits transmitted and Btner is the total number of erroneous bits received.
A lower BER value indicates a higher level of data accuracy and reliability in the communication system. For instance, a sensor of BER of 0.01 means that 1 out of every 100 bits transmitted is received in error. BER is crucial in assessing the quality of communication channels, especially in scenarios where noise, interference, or channel impairments can introduce errors in the transmitted data.
The optimal BER for wireless communication systems is typically set based on the required quality of service (QoS) for the application [25]. In high-reliability applications such as medical devices or critical infrastructure monitoring, a very low BER of 109 may be considered as optimal. The BER requirements of 1012 to 1015 may be considered optimal in optical systems to ensure error-free transmission over long distances.

댓글

이 블로그의 인기 게시물

Non-contact exposure to dinotefuran disrupts honey bee homing by altering MagR and Cry2 gene expression

  Non-contact exposure to dinotefuran disrupts honey bee homing by altering  MagR  and  Cry2  gene expression Dinotefuran is known to negatively affect honeybee ( Apis mellifera ) behavior, but the underlying mechanism remains unclear. The magnetoreceptor ( MagR , which responds to magnetic fields) and cryptochrome ( Cry2 , which is sensitive to light) genes are considered to play important roles in honey bees’ homing and localization behaviors. Our study found that dinotefuran, even without direct contact, can act like a magnet, significantly altering  MagR  expression in honeybees. This non-contact exposure reduced the bees’ homing rate. In further experiments, we exposed foragers to light and magnetic fields, the  MagR  gene responded to magnetic fields only in the presence of light, with  Cry 2 playing a key switching role in the magnetic field receptor mechanism ( MagR–Cry2 ). Yeast two-hybrid and BiFc assays confirmed an interactio...

New Report – Interlocked: Midwives and the Climate Crisis

New Report – Interlocked: Midwives and the Climate Crisis Earlier this year, midwives from 41 countries shared their experiences of working in communities affected by climate change through our survey, Midwives’ Experiences and Perspectives on Climate Change. Their voices shaped our new report, Interlocked: Midwives and the Climate Crisis , which highlights how midwives are already responding to the health impacts of climate disasters like floods, wildfires, and extreme heat—and why they must be included in climate action plans. What did we learn?Climate change is damaging community health: 75% of midwives reported that climate change is harming the communities they serve, with rising rates of preterm births, food insecurity, and restricted access to care during disasters like floods. Midwives are critical first responders: Midwives are often the first and only healthcare providers on the ground in crises, delivering care during wildfires, floods, and extreme heat. Midwives face signi...

Bee attack claims life of newspaper distributor

  Bee attack claims life of newspaper distributor Newspaper distributor Pushparaja Shetty (45), who sustained severe injuries in a bee attack, succumbed to his injuries on Thursday at a hospital in Mangaluru. Pushparaja was attacked by a swarm of bees on Wednesday morning while walking at Kenjaru Taangadi under Bajpe town panchayat limits. He was immediately admitted to a hospital for treatment but could not survive the ordeal. Fondly known as ‘Boggu’ in the Porkodi area, Pushparaja was well-known for his dedication to delivering newspapers on foot to every household. He was admired for his generosity, as he often distributed sweets to schoolchildren on Independence Day using his own earnings and contributed part of his income to the betterment of society. Pushparaja was unmarried and is survived by three brothers and one sister.

“Global honey crisis”: Testing technology and local sourcing soars amid fraud and tampering concerns

  “Global honey crisis”: Testing technology and local sourcing soars amid fraud and tampering concerns The World Beekeeping Awards will not grant a prize for honey next year due to the “inability” to thoroughly test honey for adulteration. The announcement comes amid the rise of honey fraud in the EU, where a 2023 investigation found that 46% of 147 honey samples tested were likely contaminated with low-cost plant syrups.  Apimondia, the International Federation of Beekeepers’ Associations, organizes the event at its Congress, whose 49th edition will be held in Copenhagen, Denmark, in September 2025. The conference brings together beekeepers, scientists and other stakeholders. “We will celebrate honey in many ways at the Congress, but honey will no longer be a category, and thus, there will be no honey judging in the World Beekeeping Awards. The lessons learned from Canada 2019 and Chile 2023 were that adequate testing was impossible if we are to award winning honey at the Con...

Unveiling the Canopy's Secrets: New Bee Species Discovered in the Pacific

  Unveiling the Canopy's Secrets: New Bee Species Discovered in the Pacific In an exciting development for environmentalists and beekeeping experts, researchers have discovered eight new species of masked bees in the Pacific Islands, shining a light on the rich biodiversity hidden within the forest canopy. This discovery underscores the critical role bees play in our ecosystems and the pressing need for conservation efforts to protect these vital pollinators. A New Frontier in Bee Research By exploring the forest canopy, scientists have opened a new frontier in bee research, revealing species that have adapted to life high above the ground. These discoveries are crucial for understanding the complex relationships between bees, flora, and the broader ecosystem. The new species of masked bees, characterized by their striking black bodies with yellow or white highlights, particularly on their faces, rely exclusively on the forest canopy for survival. The Importance of Bee Conservation...

Start the New Year Humming Like a Bee

  Start the New Year Humming Like a Bee There are lots of opportunities to be as busy as a bee during these winter holidays. As we hustle toward the dawn of the New Year, it can be hard to notice that the natural world is actually suggesting something different for us right now. We’re past the solstice, but the winter still stretches ahead, the days are still short and the nights long. We’re being invited into a quieter, more inner-focused time. The ancient yogis were all about this inner focus. In India, for example, the Upanishads, the Sanskrit writings that accompanied the development of Hinduism — and alongside it, yoga — beginning around 800 B.C.E., went deeper than earlier texts had into philosophy and questions of being. With the goals of increased inner awareness and higher consciousness, yoga was at that time not yet as focused on the body or on asanas, as it now can tend to be. But the yogis did develop many practices to try to open the way to those goals. They discovered...

The Essential Role of Bees in Our Ecosystem and the Challenges They Face

The Essential Role of Bees in Our Ecosystem and the Challenges They Face Bees have been an integral part of our ecosystem and human agriculture for over four centuries, primarily utilized for their honey in the early days. Now, they are recognized more for their critical role in pollination, which directly impacts a vast majority of the food we consume. Mark Lilly, a prominent beekeeper, emphasizes that without bees' pollination efforts, a significant portion of our diet would be at risk. However, bees face numerous threats, including varroa mites and, in specific regions like West Virginia, black bears. These challenges necessitate concerted efforts for preservation and protection. Frequently Asked Questions (FAQs) Q1: Why are bees so important to the ecosystem? A1: Bees are crucial for pollinating plants, which is necessary for the production of fruits, vegetables, and nuts. Without bees, a large portion of the food we eat would not be available. Q2: What are the main threats to...

The Growing Dilemma of Pet Burials in South Korea: Environmental and Legal Hurdles

The Growing Dilemma of Pet Burials in South Korea: Environmental and Legal Hurdles As the number of pets in South Korea steadily increases, pet owners face a growing dilemma due to the limited number of pet cemeteries across the country. Under the current legal framework, pet remains are classified as household waste, requiring specific disposal methods that are neither convenient nor easily accessible for all citizens. The imbalance between the rising pet population—over 5.52 million households with pets by the end of 2022—and the mere 70 registered pet cemeteries has led to significant challenges. This shortage is particularly evident in regions like Jeju Island, where pet owners must undergo arduous journeys to the mainland to ensure their pets receive proper burials. The root cause of this shortage lies in the strict regulations governing the establishment of pet cemeteries. These facilities, which must include crematoriums, face numerous restrictions, such as proximity to resident...

Climate Crisis Claims Glacier's Vital Climate Data Archive

  Climate Crisis Claims Glacier's Vital Climate Data Archive A recent study published in Nature Geoscience reveals a distressing consequence of global warming: the irreversible loss of valuable climate data stored in alpine glaciers. The research, conducted by a team led by Margit Schwikowski from the Paul Scherrer Institute (PSI), underscores the alarming rate at which glaciers are melting and highlights the implications for climate research. The study focuses on the Corbassière glacier at Grand Combin in Switzerland, where ice cores drilled in 2018 and 2020 were intended to serve as vital climate archives. However, comparing the two sets of ice cores reveals a grim reality—global warming has rendered the glacier unsuitable as a reliable climate archive. Glaciers, renowned as climate archives, encapsulate valuable information about past climatic conditions and atmospheric compositions. The fluctuating concentrations of trace substances in ice layers provide insights into historica...

German election: Climate and environment take a back seat

  German election: Climate and environment take a back seat When the coalition government comprising the center-left  Social Democratic Party (SPD) ,  Greens  and neoliberal  Free Democratic Party (FDP)  emerged after the last German federal election in the fall of 2021, then-incoming Chancellor  Olaf Scholz  (SPD) did not object to being called the "climate chancellor." That was no surprise: the climate crisis had been a top issue during the election campaign. The new government made the fight against climate change a task for the Economy Ministry and appointed Vice Chancellor  Robert Habeck  from the Greens as its head. Three and a half years later, campaign speeches barely mention climate protection. The dominant issues are how to curb irregular immigration and how to boost Germany's sluggish economy. Skeptical view of renewable energy The head of the center-right  Christian Democratic Union (CDU) ,  Friedrich Merz ...