Focuses on isolated nodes. Uses regional average energy and the distance between sensors to determine the data transmission for efficient data transmission.
The calculation of regional energy might increase the process time. Usage of regional energy might still exhaust the individual energy quickly.
—
Pseudocode or algorithm is not clearly discussed. Discussion on the mathematical computation only.
Balancing energy consumption Extend network lifetime.
A tight closed-form expression is proposed for the optimal number of cluster heads (CHs). Deriving a new optimal probability for a sensor node to serve as a CH for EDB-CHS-BOF protocol for the reason of achieving a balanced energy consumption. The clustering shape used is hexagonal as it is closer to reality.
The involvement of many operations might increase the overall complexity of the algorithm. Having adjacent CHs may cause long-distance communications, which lead to increased energy consumption.
—
Pseudocode or algorithm is not clearly discussed. Discussion on the mathematical computation only.
Balanced energy consumption. Improve network lifetime.
The SF-CHs algorithm can reduce the residual energy variance of nodes in the network. The optimization of the threshold function selects the optimal CHs.
Each sensor randomly generates a number in [0, 1] for CH selection in DEAL, where the random selection is still not the optimized way for CH selection.
Implement the method in a real environment for verification. Develop the method for application specific WSNs.
Pseudocode or algorithm is not clearly discussed. Discussion on the mathematical computation only.
Network division in the diagonal form to reduce the load of the network. Optimal values for the fuzzy inference system are analyzed. Data routing is discussed for optimal routing.
Setting the optimal parameters might be a difficult job and might also limit scalability.
A genetic algorithm can be used for CH selection and cluster formation to enhance the network lifetime.
Maximize the energy efficiency. Minimize time delay.
Usage of cyclic randomization improves the performance of the algorithm. Discusses the ability of FCR to handle multiple objectives.
Firefly algorithm might easily get stuck at local optima. The proposed algorithm suffers a high computational cost, which requires substantial minimization. High processing complexity.
Security constraints and other practical constraints can be considered in network modeling in the future.
Maximize the WSN’s lifetime. Maximize energy efficiency.
Focuses on controlling exploitation and exploration capabilities of GSA by using tournament selection. TDMA schedule is organized by the cluster head to avoid data collisions.
Solely focuses on energy efficiency and does not consider QoS. GSA might have problems with exploration and exploitation if treated separately.
To introduce a controller or a function that can determine the value of the algorithm parameters user and problem independently. The time and computational complexity of the algorithm can be analyzed.
High coverage ratio. Low redundancy ratio. Energy consumption balance.
Discusses multiobjective optimization model due to the uncertainty between coverage ratio and redundancy ratio. To achieve better optimization, this paper improves inertia weight to PSO.
PSO individually may fall into local optima. The number of cluster head influences the number of nodes alive using IAPSO, which might not be efficient.
Prolong network lifetime. Increase energy efficiency.
Artificial bee colony (ABC) algorithm to optimize its crucial parameters which are done offline. Harmony search (HS) algorithm is used for CH selection, and it is done online.
Single hop might cause the CH far away from BS to consume more energy. Tuning parameters might worsen the performance of the algorithm if the wrong tuning is made and needs more effort. CPMA has relatively high process time.
Deploy the method in a large-scale and mobile network environment considering the IoT applications that exist. Use different energy harvesting constraints in CPMA to improve the network’s throughput.
