The modern stage of development of communication networks is the concept of the Internet of Things. The main task is to create a single network, which includes objects of information (virtual) and physical (real) worlds and will ensure the interaction of objects with each other.　　Clustering has proved to be one of the most important methods of creating wireless sensor network (WSN). The functioning of a clustered WSN depends largely on the head node selection algorithm, and the main requirements are to ensure the maximum life cycle of the network and the maximum coverage.　　This dissertation aims to study cluster based algorithms for wireless sensor networks and improves threshold-sensitive energy efficient network algorithm (TEEN) by implementing a fallback mechanism (error detection and recovery phase) for clustering in WSN for increasing the number of successfully transmitted packets both from members of the cluster to cluster head (CH) and from CH to the base station. In addition, to provide a tolerance to the value of the error probability, this dissertation analyzes the performances of DT, LEACH, TEEN, SEP and DEEC routing and self-organization algorithms for homogeneous and heterogeneous WSNs, using the following indicators:Life cycle, Stability, Instability, Number of living nodes and number of dead nodes.　　Firstly, the analysis of DT, LEACH, TEEN, SEP and DEEC shows that the LEACH algorithm is an extremely effective protocol for homogeneous WSN. It helps to reduce energy consumption by seven times compared to the direct interaction of sensor nodes, and the TEEN algorithm is superior in terms of the duration of the period of stability and residual energy of all the algorithms considered and it is recommended for use in both homogeneous and heterogeneous WSNs.　　Then, on the basis of the hierarchical cluster based TEEN algorithm an improved FT-TEEN fault-tolerant clustering algorithm for WSN isdeveloped. It provides an increase in the number of successfully transmitted packets from both cluster members to CH and from CH to the base station compared to the TEEN algorithm, as well as tolerance to the error probability value, while for the basic TEEN algorithm the number of successfully transmitted packets is significantly reduced with increasing error probability within the considered limits.　　In addition, this paper examines characteristics for WSN to determine a technique for locating sensor nodes so that at least 90% coverage can be provided for the two-dimensional (2D) and three-dimensional (3D) WSN.