Arora, S., & Singh, S [5] proposed a nature-inspired node localisation algorithm based on Butterfly Optimization algorithm in which the unknown node location is calculated in the presence of Gaussian noise in received signal [1].The network consists of 25 to 150 nodes. The authors investigated the performance of the algorithm in terms of localization error, no. of nodes localized and computation time. The study is done for small sized homogeneous network and all the beacon nodes are static and optimum no. of beacon nodes required for localization is not studied. Sun, Z. et al[6] proposed multi-objective particle swarm localization optimization (MOPSOLA) algorithm for sensor node localization in which the objective function consists of geometric topology constraint and space distance constraint. The algorithm calculates the optimal solution by adopting both the dynamic maintenance operator for archive and global optimal selection operator based on proportion selection. The algorithm considers the presence of Gaussian noise with zero mean. The performance of the proposed algorithm is checked by varying no. of beacon nodes, varying radius, varying network connectivity, varying node density. The case of a WSN that consists of heterogeneous nodes is not considered. Zhao, X et al[7] proposed a three-dimensional node localisation algorithm in which a triangular pyramid is constructed by using the neighbouring anchor nodes an unknown node. Ferment point model is used to divide this structure into smaller sub-triangular pyramids. The sub-triangular pyramids in which the unknown node is present is further divided into a set of irregularly shaped subspaces by the mid-perpendicular model and then the centroid of the subspace is computed to determine the unknown node location. Performance of the algorithm is calculated in terms of localization error and localization coverage. Minu, M. C et al[8] proposed a range based node localisation algorithm based on non-linear optimization property of artificial immune system inspired population based algorithm called Clonal Selection Algorithm. The algorithm focuses on homogeneous network and only focuses on the minimization of localization error. Namin, P. H., & Tinati, M. A [9] proposed a PSO based node localization model for homogeneous WSN in which first DVdistance method with multilateration is used to estimate coarse position of unknown node and then PSO algorithm is used to fine tune the result. The algorithm addresses the problems associated to the node localization such as flip ambiguity, collective translation and error propagation. Song, L et al[10] proposed a mobile node localization algorithm that uses a combination of Monte Carlo localization method based on Improved QUasi-Affine TRansformation Evolutionary (QUARTE) Optimization method. Firstly, high-quality common nodes in the range of one hop of unknown nodes as temporary anchor nodes are selected. These temporary anchor nodes and actual anchor nodes are used as reference nodes for positioning and a sampling area is identified. Then the improved QUARTE optimization algorithm is used to estimate the location of unknown node in the sampling area. Authors claim that the algorithm provides high accuracy and high coverage with relatively small amount of anchor nodes. Xue, D. [11] proposed a model a range-free node localization algorithm based on particle swarm optimization algorithm. This algorithm modifies the least square method of location estimation to minimize the localization error caused by initial value sensitivity by applying PSO. Tan, R., et al[12] proposed Distance Mapping Algorithm (DMA) for unknown node localization in WSN. The algorithm estimates location of unknown node by using the estimation matrix and distance matrix together with the optimized linear transforming function and then applying GA. Authors claim that the algorithm is better in comparison to other algorithm in terms of localization accuracy, energy efficiency and scalability. Node heterogeneity is not considered. Liu, X., & Liu, C. [13] proposed a dynamic mathematics model for node localization in WSN and compared the performance of the model with the traditional DV-HOP,GA,PSO based node localization algorithms and found improved localization accuracy and localization coverage for the proposed model. The algorithm considers the dynamic behaviour of sensor nodes. Energy consumption and sensor node heterogeneity are not considered. Mozamir, M. S et al[14] proposed an Improved Global best Local Neighbourhood Particle Swarm Optimization (IGbLN-PSO) algorithm for minimization of error in estimation of unknown node location. The algorithm consists of two phases:-Exploration phase and Exploitation phase. The neighbour particles population that scattered around the main particles, help in the searching process to estimate the node location more accurately with less computational time. Node heterogeneity and energy efficiency are not considered.
Shen, S et al[15] proposed an improved DV-Hop algorithm for unknown node location estimation in WSN. The proposed algorithm uses probabilistic information based selective strategy for beacon node selection. Then a two-dimensional hyperbolic function to estimate the node location. Tian, J et al.[16] proposed an improved centroid location algorithm based on finite element method for node localization problem. In the finite element analysis, a node is an unknown function and represents the position of any point in the coordinate system. Therefore, it is necessary to assume a displacement mode which is a polynomial approximation method which can approximate any smooth function. Pascal’s triangle is used to ensure the symmetry of polynomials or geometric orientation. The initial location of unknown node is determined using signal strength information and triangle area of finite element method. The final location of the unknown node is estimated by centroid location algorithm. Zhang, L et al.[17] proposed a node localization algorithm that uses combine information of received signal strength indicator( RSSI) and Time of Arrival(TOA) for unknown node location estimation. The algorithm uses single mobile anchor node. This mobile anchor node moves according to the Gauss-Markov three-dimensional mobility model. Then based on the RSSI ranging in the near end and TOA ranging in the far end, precise distance between anchor node and unknown node is determined. Finally, maximum-likelihood estimation method is used to estimate the position of the unknown node. Tuncer, T. [18] proposed an intelligent node localization method that combines centroid localization algorithm, fuzzy logic and genetic algorithm for minimization of error in location information estimation. RSSI value measured by anchor nodes are given as input to the fuzzy system. Anchor nodes are assigned with weight values to increase the high value RSSI anchor nodes to participate in location estimation process for unknown nodes. Then to minimize estimation error, fuzzy system’s output membership functions are adjusted using genetic algorithm. Assaf, E.A et al.[19] proposed a range-free node localization algorithm in which each regular or position unaware node estimates its distance only to reliable anchors or position-aware nodes. Suitable position-aware nodes are selected using a anchor selection strategy that ensures an accurate distance estimation. A power-saving mechanism is incorporated with the algorithm that ensures energy efficiency. The algorithm is designed for 2d and 3d applications. Kumar, V., & Kumar, A. [20] proposed a node localization algorithm based on DV-Hop, PSO algorithm in which traditional DV-Hop localization algorithm using PSO is modified to make it suitable for localization of nodes displaced due to undesirable event in AOI. Radio irregularity model is used with this to make it suitable for anisotropic sensor network. Zhang, Y., & Liu, Y. [21] proposed an intelligent computing method- Grey Wolf Optimization method for node localization in WSN. The localization performance of the algorithm is compared with traditional PSO based algorithm. Rajakumar, R et al.[22] proposed a Grey Wolf Optimization based node localization method for improved location accuracy. According to the author this meta-heuristic algorithm is better meta-heuristic algorithm in terms of exploration and exploitation ability in search space. But optimum result is not guaranteed when there is limited diversity in search space.