ICT Focus

Evaluation of Indicators for ICT Development Index using an Integrated Entropy Weighting Method

2023-06-17T11:01:14+08:00

When designing a data analysis using information entropy, metrics derived from this method must be observed, evaluated, and utilized. Information entropy, which reflects the level of uncertainty in a random variable, can be applied to a range of fields, including information and communication technology (ICT). The methodology proposed aims to assign weights to the indicators within the ICT development index and sub-indexes, enabling their global, regional, and country-wise ranking. To test the effectiveness of this methodology, we examined its potential applications for evaluating indexes. Our model integrates a novel approach that combines the entropy weight coefficient method, bootstrap method, correlation coefficient weighting method, and S-shaped diffusion stages of ICT. We present the evaluation results of the integrated calculating method.

Development of Underground Perimeter Intrusion Detection System

2023-10-17T05:16:47+08:00

An underground passive security system based on minimal electric charge detection was developed and implemented in the real application field. The fixed length of the sensor cable is buried underground named a security zone. It is generated an alarm when an intruder goes into the security zone. A shielded telecommunication cable can be used for sensing the generated minimal charges by the external intrusion. If two different types of materials are rubbed with each other, minimal electric charges are generated, called the triboelectric effect. Minimal electric charges are generated between the cable shield and conductors when forces and impacts are applied to the underground buried cable. The control device detects and amplifies generated minimal charges due to the intruders on the sensor cable. The proposed system can be shown a reliable intrusion detection rate of up to 96%-97% when we have kept and considered all installation and environmental conditions.

Predictive Neural Network Modeling for Diesel Engine Part Wear Assessment via Analysis of Wear Element Concentration in Used Oil

2023-12-25T07:13:32+08:00

Used oil serves as a repository of wear elements, reflective of the friction generated among various engine components. The accumulation of these elements offers vital insights for understanding wear and tear in diesel engine parts. Accurate prediction of such wear is imperative for strategic maintenance planning and cost efficiency.

This paper presents an innovative methodology that leverages artificial neural network modeling to forecast the degradation of locomotive diesel engine components based on the concentrations of wear elements in used oil and the corresponding vehicle mileage. The proposed supervised artificial neural network comprises two hierarchical architectures: one linking diesel locomotive operation and wear element concentration, and another associating diesel locomotive operation with diesel engine component wear. Through experimental validation, the results highlight the efficacy of the developed neural network model in precisely estimating the wear of diesel engine components. This predictive capability empowers informed maintenance decision-making to bolster engine reliability and performance.

Optimizing User-Level Packet Scheduling Performance through Optimal CQI-Based Resource Allocation in LTE

2023-12-29T11:30:05+08:00

Our research investigates the Dynamic Resource Allocation in the Downlink of OFDMA-based LTE-A networks. It addresses both user-level and system-level packet scheduling performance. At the user-level, a Traffic Differentiator stage segregates packet queues from active users into different service queues based on service types. Users are prioritized within each service queue based on their QoS requirements and wireless channel conditions, utilizing SPSSA. At the system-level, fairness among users is a key consideration. We propose the PITDSA in the TD Scheduler stage, which aims to allocate just enough radio resource to real-time and non-real time services and assign the remaining available resource to background service. In the FD Scheduler stage, we propose an optimal CQI selection algorithm for resource allocation to exploit frequency domain multi-user diversity. We present our simulation results and analyses of all novel algorithms, and the performance of the Optimal CQI selection algorithm is compared with other algorithms. Our proposed scheduling algorithm demonstrates improved QoS for real-time and non-real time services while maintaining a good traded user-level and system-level performance. Future work can focus on optimizing the throughput or the fairness or both, and more advanced and complex techniques can be designed with the same goal.