Search

Recently, many companies are building systems to generate profits through various AI-based demand forecasting algorithms. Therefore, this study aims to develop an application that allows users to monitor data related to AI demand prediction in real time. When creating this application, the goal is to help you easily understand the information by adapting the color to the purpose, and to easily convey the information and recognize it at a glance by considering the layout. In addition, the design was constructed using the check box function and each text function in the ID and password verification and the unique code partial configuration. Confidential data security can be enhanced by allowing only the data of the corresponding company to be checked using the unique code set for each company. As a result, real-time monitoring can be viewed without security problems with mobile applications. [ABSTRACT FROM AUTHOR]

Commercial products, such as the National Instrument's USRP (universal software radio peripheral) products and the latest instru- ments, provide functions to receive and analyze signals. Because the characteristics of the target signal are diverse, these systems are designed to receive a wideband signal to a narrowband signal and provide functions for extracting a desired signal and analyzing it in detail if necessary. In the past, these functions were implemented in the form of hardware, such as FPGA (field programmable gate array) and DSP(digital signal processor), for real-time processes. However, owing to the development of software technology, it is possi- ble to replace the signal-processing part of the acquisition system with software. The existing GPU-based signal processing unit is capa- ble of processing a few channeld only in real-time owing to its excessive computation and memory usage. To overcome this limitation, frequency-domain signal processing is applied and the computation time and memory usage are considerably reduced. Thus, real-time processing of hundreds of channels is possible. In this study, a multi-channel wideband signal processing unit is designed and im- plemented using programmable equipment, such as CPU and GPU processors, rather than expensive hardware equipment. [ABSTRACT FROM AUTHOR]