
Reliability of data transmission and energy efficiency are critical challenges in the design of wireless sensor networks (WSN) for remote environmental monitoring. This article details the design, implementation, and analysis of a reliable datagram protocol (ACK) based on a request-response (Polling) architecture to overcome the limitations of low-cost systems. The project successfully migrated from an initial prototype (MK1) that relied on software-based ACK management over unstable 315 MHz radio links to an operational solution (MK2) utilizing the NRF24L01 (2.4 GHz) transceiver. This strategic shift was a key finding, as the NRF24L01's hardware-implemented Auto-Acknowledge and automatic retransmission capabilities successfully delegated reliability management to the link layer, eliminating critical system instability (sendtoWait failed). Built on this robust foundation, a hierarchical tree topology was implemented with a synchronous Polling protocol that not only ensures data integrity but also maximizes autonomy by forcing sensor nodes to remain in a low-power (sleep) mode until explicitly requested by the master. The results conclusively validate this hybrid hardware-software approach, demonstrating a negligible packet loss rate and achieving optimal energy consumption essential for deploying an autonomous and scalable telemetry system.
The widespread use of machine learning to address a wide range of practical issues has resulted in the requirement to gather and analyze massive amounts of data, some of which are regarded as sensitive and personal, raising major data protection issues. The existing EU legislation on data protection and AI typically demand privacy-enhancing technologies (PETs) to reach a general trustworthiness and to secure personal data. In the increasingly digitally connected world of today, data privacy has become crucial. The growth of data in many different fields has made protecting sensitive information more crucial than ever. To improve data privacy in the digital era, this abstract examines state-of-the-art machine learning approaches. A workable solution to data privacy concerns is provided by the machine learning subset of artificial intelligence. This paper explores the potential of machine learning methods to improve confidentiality safeguards across a variety of applications. Machine learning models can provide proactive defense against cyber-attacks by analyzing massive amounts of data and potentially identifying vulnerabilities and breaches in real-time.
The complexity of a graph can be determined using a network-theoretic method, which is given. This strategy is based on the connection between graph theory and the theory of determinants in linear algebra. In this article, we provide a unique algebraic method that makes use of linear algebra to derive simple formulas for the complexity of diverse novel networks. We can derive the explicit formulas for the globe network and its subdivision using this method. In the final least, we also determine their spanning trees entropy and compare it to each other
In this paper, we consider the NP-hard problem of finding the minimum resolving set of graphs. A vertex set B of a connected graph G resolves G if every vertex of G is uniquely identified by its vector of distances to the vertices in B. The cardinality of the minimal resolving set is the metric dimension of G. The metric dimension appears in various fields such as network discovery and verification, robot navigation, combinatorial optimization and pharmaceutical chemistry, etc. In this study, we introduce a hybrid approach (WCA_WOA) for computing the metric dimension of graphs that combines the water cycle algorithm and a whale optimisation algorithm. The WOA algorithm hybridises the WCA in order to obtain the optimal result and manage the optimization process. The results of the experiments show that the WCA_WOA hybrid algorithm outperforms the WCA, WOA, and particle swarm optimization methods
In Smart Home (SH) the devices and appliances can be controlled automatically and remotely via the internet. Sensors in a smart home detect changes, which are processed and used to make intelligent decisions, such as automatically turning on the lights. To be effective, a smart home's sensors must be connected, establishing a wireless sensor network as the backbone for the smart home's functionalities. For many researchers, the performance of WSN faces several challenges. Power consumption is the most crucial factor since nodes require power to collect, process, and send data but have limited energy. As a result we propose an improved protocol for WSN in smart homes called EEP-LEACH (Energy Efficient Protocol - Low Energy Adaptive Clustering Hierarchy) based on the LEACH protocol. Matlab was used for the algorithm's performance analysis, and the simulation result showed that, compared to others protocols the death time of the nodes was longer, thus managing to balance the consumption of energy of the network nodes. Python and Arduino are used in this research to perform some simulations.