Volume 26 Issue 2 2026

The deployment of fifth-generation (5G) networks, particularly in millimeterwave (mmWave) bands, is significantly challenged by rain-induced signal attenuation, especially in regions with convective rainfall patterns. This study presents a comprehensive six-year (2010–2015) analysis of rain attenuation characteristics for 5G Frequency Range 1 (FR1: 3.5 GHz) and Frequency Range 2 (FR2: 30 GHz) in Johannesburg, South Africa. Using high-resolution hourly precipitation data and ITU-R models, we quantify attenuation, analyze its statistical properties, and characterize its temporal variations. Results reveal attenuation at 30 GHz is approximately 1,600–1,750 times greater than at 3.5 GHz, with maximum values exceeding 37.5 dB for a 7 km path during summer. Statistical analysis shows highly right-skewed, leptokurtic distributions, dominated by dry periods but punctuated by intense, short-duration events. Pronounced seasonal peaks occur in summer (December to February: DJF) and spring (September to November: SON), with a distinct diurnal peak in the late afternoon to early evening. These findings underscore the severe vulnerability of mmWave links to rain fade and provide essential data for designing robust fade margins, adaptive network strategies, and informed spectrum allocation for reliable 5G and beyond-5G wireless services in subtropical urban environments.

Escherichia coli O157:H7, an emerging cause of food-borne disease, has now been reported from several countries worlwide. Cattle and sheep are known as a principle reservoir of this agent. It is also prevalent in gastrointestinal tract of other animals like pigs, dogs, cats and birds. This agent can be transmitted to humans by food and water and even person to person contact. The random amplified polymorphic DNA (RAPD) technique has been applied to evaluate zoonotic potency of Escherichia coli O157:H7 local isolates. Ten random decamer primers have been evaluated in this study to differentiate isolates originated from humans and animals. The OPA-02, OP-03, OPA-04, OPA- 07, OPA-08, OPA-09, OPA-10, OPA-13, OPA-19, and OPA-20 decamer primers show a good differentiation pattern of E. coli O157:H7 isolated from human, cattle and chicken feces, as well as from beef. The primers constantly produce 16, 16, 11, 14, 8, 24, 7, 11, 7, and 12 polymorphic bands, respectively. Various human isolates that were isolated from clinically ill and healthy patients share common genetic clusters with animal isolates from beef, as well as cattle and chicken feces with genetic similarity coefficients greater than 70%. According to this result, It has been concluded the transmission of E. coli O157:H7 local isolates from animals to humans is potential occur.

The present study was conducted at Cotton Research Institute, Ayub Agricultural Research Institute, and Faisalabad Pakistan during 2012. The boll size of Sbe-18 was large; tapering shape, good opening and leaf was of medium size as compared to all other varieties. It was concluded that higher heritability and genetic advance was found for monopodial and sympodial branches, nods to first flower days, number of bolls and plant height. Significant genotypic and phenotypic correlations were found for monopodial and sympodial branches, number of bolls, staple length and fibre fineness. Higher heritability, genetic advance and correlation indicated that selection may be helpful for the improvement of yield and quality of cotton.

In this paper we establish the feasibility of extracting clean water from a lake or a lagoon and then depositing it on adjacent arid land. The entire process is empowered using only the daytime/nighttime temperature variation. To obtain this water transformation and irrigation we use a large greenhouse type structure covering both a portion of the water and the adjacent shore land. The solar energy of the sun shining through the greenhouse heats the water producing a high temperature air/water-vapor mixture. Low evening temperature then produces a heavy dew upon the land. The paper provides the underlying thermodynamic analysis for the process, together with a brief review of the essential concepts including: mass; weight; pressure; partial pressure; volume; specific volume; temperature; molecular hypotheses; ideal gases; mixtures; saturation; evaporation; relative humidity; specific humidity; and dew point. We conclude with a series of illustrative computations validating the process. The results show that water condensate increases linearly with the greenhouse volume and with the humidity, but exponentially with the greenhouse temperature.

The diagrid structural system has emerged as a preferred solution for high-rise buildings because of its high lateral stiffness, reduced material usage, and aesthetic flexibility. This study compares the seismic performance of a thirty-storey steel building using two lateral load-resisting systems—(i) a conventional moment-resisting frame and (ii) a perimeter diagrid frame—both modelled in ETABS following IS 875 and IS 1893 (Part 1): 2016. The models’ dimensions are the same (30 m × 30 m plan, 3.2 m storey height). Comparative parameters include fundamental period, storey displacement, inter-storey drift, and base shear. Results indicate that diagrid systems exhibit superior lateral stiffness and drift control: fundamental period reduced by ≈ 41 %, top-storey deflection by > 50 %, and inter-storey drift by > 60 %. The 4-module diagrid had the best stiffness and deformation control of all the configurations that were examined. The findings show that the diagrid system can be used in seismically active regions and has the potential to be an affordable alternative to conventional frames.

Due to the rapid industrialization and urbanization, the water quality has been reduced significantly, which poses a threat to the environment and human health. Manual collection of water samples by hand and laboratory testing is a time and labor-consuming process and does not provide any source of real-time information. To solve such issues, an IoT-based Smart Water Quality Monitoring System (SWQMS) was designed on IoT with sensor analytics. Conductivity, dissolved oxygen, turbidity, and pH sensors are deployed to post real-time data and identify anomalies together with edge processing through microcontrollers. The IoT communication protocols are used to transfer the received data to a cloud platform, where this data may be stored, viewed, and analyzed to give predictions. SWQMS in freshwater sources had a high correlation with laboratory measurements, which demonstrates why it accurately monitors and detects the occurrence of pollution in time. The environment can be proactively controlled by constant monitoring, and it also gives early notice of contamination. The suggested SWQMS will constantly measure pH (7.037.14), turbidity (3.23.6⁻ 7.037.14⁻ 3.23.6NTU), dissolved oxygen (6.857.10 mg/L), and temperature (24.925.6 °C) with inbuilt multi-parameter sensors and edge analytics. IoT-based data collection and forecasting, real-time, and visualization on clouds allow the timely recognition of anomalies early and the delivery of warnings towards water quality management on ponds, industrial wastewater, and aquaculture facilities.

Floods are among the most devastating natural hazards that leave a lot of destruction in their path in terms of life, infrastructure, and environment. Sound and timely forecasting of floods is crucial to proper disaster warning and planning of resources by the government. The challenges exist in existing forecasting systems that do not offer scalability, early warning ability, delayed computation, and lack of heterogeneous real-time data integration. To address these shortcomings, the proposed paper presents a Hybrid Parallel Flood Prediction Framework (HPFPF) an integration of Parallel Hydrodynamic Simulation (PHS) and a Distributed Machine Learning-based Prediction Engine (DMLPE). The PHS module simulates concurrent river basin and rainfall-runoff, whereas DMLPE simulates historical and sensor data in parallel at various nodes, which allows predicting dynamically under different conditions. The data assimilation methods also increase the accuracy of the model by continually updating the forecasts with real time inputs. Experimental performances show HPFPF can compute as much as 70% faster and have prediction accuracy more than 93. The modularity of the framework enables to integrate in disaster alert systems of government to ensure proactive disaster risk mitigation through flood control.

Bakery production involves a myriad of decisions ranging from ingredient sourcing to product assortment, all aimed at meeting customer demands while maintaining profitability. Goal programming offers a powerful framework to navigate the complexities of bakery production by simultaneously addressing multiple objectives, such as maximizing profit, minimizing costs, and ensuring product quality. This abstract explores the application of goal programming techniques in the context of bakery production. By formulating the bakery's objectives into a mathematical model, goal programming enables decision-makers to find optimal solutions that balance conflicting goals. Key considerations include resource allocation, production scheduling, inventory management, and product diversification. Through a case study or simulation, this abstract demonstrates how goal programming can be used to enhance bakery operations. By considering constraints such as production capacity, ingredient availability, and market demand, the model identifies production plans that optimize the bakery's performance while satisfying various stakeholders' preferences. In addition, this abstract discusses the potential benefits and challenges of applying goal programming in small industries such as bakery production. Three goals are considered - maximizing daily sales profit, minimizing overtime and optimal utilization of the machines used in the production of bakery products. The Longo Optimizer solver indicated that SMEs may need to review their profit goals in line with their policy on overtime and time utilization of their machines.

The present work investigates heat and mass transfer characteristics in an inclined porous channel by incorporating the effects of double dispersion, Dufour diffusion, chemical reaction, and viscous dissipation under slip boundary conditions. Thermal and solutal dispersion mechanisms are considered to accurately model transport phenomena in porous media. A first-order homogeneous chemical reaction is assumed, while viscous dissipation is included to account for internal heat generation. The system of nonlinear equations governing momentum, energy, and concentration is transformed into a coupled system of ordinary differential equations using suitable similarity transformations and are solved numerically using the Successive Quadratic Linearization Method (SQLM). In addition, an entropy generation analysis is performed to assess thermodynamic irreversibility arising from heat transfer, fluid friction, and mass diffusion. The influences of key parameters such as slip parameter, dispersion coefficients, Dufour number, chemical reaction parameter, inclination angle, Brinkman number, and Eckert number on entropy generation rate, velocity, temperature, concentration, and Bejan number are examined. The results indicate that enhanced Dufour and viscous dissipation effects significantly increase temperature and entropy generation, whereas stronger chemical reactions suppress concentration and reduce solutal irreversibility. The present findings are relevant to the optimization of thermal systems involving porous channels and reactive transport processes.

The continuous evolution of the Internet of Things (IoT) and the increasing reliance on multi-cloud environments have highlighted the urgent need to secure interconnected devices. Malicious actors are now targeting Internet-connected systems and devices, necessitating more advanced measures to protect data and maintain system integrity. This growing dependence on interconnected systems has made the task of safeguarding data increasingly complex. With the exponential rise in data generation, there is a pressing need for innovative methods to enhance data security and analysis. Intrusion Detection Systems (IDS) serve as vital tools for monitoring and analyzing data to identify unauthorized access or anomalous behavior within networks and systems. This paper presents an innovative framework to address the issues of intrusion detection in MultiCloud IoT ecosystems, providing a thorough analytical methodology. The vast quantity, diversity, and speed of data generation in these networks pose significant challenges for traditional detection techniques. This study presents an intelligent and integrated solution to these issues, focusing on the unique complexities that arise from the convergence of IoT and multicloud technologies. The findings confirm that the Random Forest algorithm significantly improves detection accuracy, achieving an exceptional 99% accuracy rate, outperforming other conventional techniques.

As the automotive sector accelerates towards electric vehicles (EVs), predicting battery states accurately is vital for maximizing performance, safety, and lifespan. This project presents a novel approach that utilizes Explainable Data-Driven Digital Twins to forecast battery states in electric vehicles (EVs). It incorporates various advanced machine learning algorithms, including DNN, LSTM, CNN, SVR, SVM, FNN, RBF, RF, and XGBoost. The key objective is to enhance the accuracy of predicting critical battery metrics like SOC and SOH under diverse operating conditions. Additionally, the project applies explainable AI to uncover factors that impact battery performance. By harnessing the strengths of various algorithms, the digital twin model shows improved prediction accuracy and resilience compared to traditional methods. This research advances intelligent, adaptive battery management systems, paving the way for the future of electric transportation.