Journal of Sensor and Actuator Networks

This paper presents test results of the performance comparison of 5G standalone (SA) and non-standalone (NSA) networks in the context of gathering data of remote sensors and machines. The study evaluates key network characteristics such as latency, throughput, jitter and packet loss (for UDP protocol only) using standardized tests to gain insights into the impact of these factors on real-time and data-intensive communication. In addition, a range of communication protocols including OPC UA, Modbus, MQTT, AMQP, CoAP, EtherCAT and gRPC were tested to assess their efficiency, scalability and suitability with different send data sizes. By conducting experiments in a controlled hardware environment, we have analyzed the impact of the 5G architecture on protocol behavior and measured the transmission performance at different data sizes and connection configurations. Particular attention is paid to protocol overhead, data transfer rates and responsiveness, which are crucial for industrial automation and IoT deployments. The results show that SA networks consistently offer lower latency and more stable performance, where robust and low-latency data transfer is essential. In contrast, lightweight IoT protocols such as MQTT and CoAP demonstrate reliable operation in both SA and NSA environments due to their low overhead and adaptability. These insights are equally important for time-critical industrial protocols such as EtherCAT and OPC UA, where stability and responsiveness are crucial for automation and control. The study highlights current limitations of 5G networks in supporting both remote sensing and industrial use cases, while providing guidance for selecting the most suitable communication protocols depending on network infrastructure and application requirements. Moreover, the results indicate directions for configuring and optimizing future 5G networks to better meet the demands of remote sensing systems and Industry 4.0 environments.

Underwater wireless optical communication (UWOC) has emerged as a key enabler for Internet of Underwater Things (IoUT) and autonomous sensing networks, but its reliability is severely affected by salinity-induced attenuation, scattering, and turbulence. This work presents a high-speed and salinity-resilient UWOC architecture that jointly exploits Polarization Division Multiplexing (PDM) and Multiple-Input Multiple-Output (MIMO) diversity to enhance link capacity and robustness in realistic oceanic conditions. Two 1 Gbps NRZ data channels at 1550 nm were transmitted using continuous-wave lasers and evaluated using a hybrid OptiSystem–MATLAB simulation framework with full channel modeling of absorption, scattering, turbulence, and salinity (32–36 ppt). Results reveal that the proposed PDM-MIMO system achieves more than an order-of-magnitude bit-error-rate (BER) reduction compared with non-MIMO or single-polarization baselines, maintaining acceptable BER levels up to 20 m. Performance degradation with increasing salinity is quantified, and results confirm that combined PDM and spatial diversity effectively mitigate salinity-induced losses. The presented design demonstrates a viable and scalable solution for next-generation underwater sensing and communication networks in coastal and deep-sea ecosystems.

The high-quality droplet formation in continuous inkjet printing (CIJ) is crucial for precise character deposition on product surfaces. This process, where a piezoelectric transducer perturbs a high-speed ink stream to generate micro-droplets, is highly sensitive to parameters like ink pressure and transducer amplitude. Suboptimal conditions lead to satellite droplet formation and charge transfer issues, adversely affecting print quality and necessitating reliable monitoring. Replacing inefficient manual inspection, this study develops MBSim-YOLO, a deep learning-based method for automated droplet detection. The proposed model enhances the YOLOv8 architecture by integrating MobileNetv3 to reduce computational complexity, a Bidirectional Feature Pyramid Network (BiFPN) for effective multi-scale feature fusion, and a Simple Attention Module (SimAM) to enhance feature representation robustness. A dataset was constructed using images captured by a CCD camera during the droplet ejection process. Experimental results demonstrate that MBSim-YOLO reduces the parameter count by 78.81% compared to the original YOLOv8. At an Intersection over Union (IoU) threshold of 0.5, the model achieved a precision of 98.2%, a recall of 99.1%, and a mean average precision (mAP) of 98.9%. These findings confirm that MBSim-YOLO achieves an optimal balance between high detection accuracy and lightweight performance, offering a viable and efficient solution for real-time, automated quality monitoring in industrial continuous inkjet printing applications.