TY - GEN

T1 - Data Collection for Classification of Radio Signals Using SDR Transceivers

AU - Stafeev, Daniil

AU - Ronkin, Mikhail

PY - 2023/5/15

Y1 - 2023/5/15

N2 - The paper deals with the problem of automatic classification of telecommunication signal modulation types using machine-learning methods. The task is actual in the field radio electronic, with the advent of integrated telecommunication systems, new modulation, and coding methods, it became necessary to search and develop simple, efficient, reliable methods for processing signals in channels with interference. The type of modulation is an important parameter on which the signal processing circuit of the receiving or transmitting device depends. It is impossible to extract useful information from the received signal without knowing the type of modulation. A database for the problem solution was collected through the experiments with software-defined radio. During the experiments the signals at a frequency of 900 MHz, with modulations: BPSK, GMSK, NBFM, OFDM were collected. The most popular of the contemporary methods for classification are evaluated and compared. As the result of this study, a method for collecting data using an SDR transceiver was considered, and a comparison was made of the accuracy of algorithms trained on the collected dataset.

AB - The paper deals with the problem of automatic classification of telecommunication signal modulation types using machine-learning methods. The task is actual in the field radio electronic, with the advent of integrated telecommunication systems, new modulation, and coding methods, it became necessary to search and develop simple, efficient, reliable methods for processing signals in channels with interference. The type of modulation is an important parameter on which the signal processing circuit of the receiving or transmitting device depends. It is impossible to extract useful information from the received signal without knowing the type of modulation. A database for the problem solution was collected through the experiments with software-defined radio. During the experiments the signals at a frequency of 900 MHz, with modulations: BPSK, GMSK, NBFM, OFDM were collected. The most popular of the contemporary methods for classification are evaluated and compared. As the result of this study, a method for collecting data using an SDR transceiver was considered, and a comparison was made of the accuracy of algorithms trained on the collected dataset.

UR - http://www.scopus.com/inward/record.url?partnerID=8YFLogxK&scp=85164921773

U2 - 10.1109/USBEREIT58508.2023.10158899

DO - 10.1109/USBEREIT58508.2023.10158899

M3 - Conference contribution

SP - 194

EP - 197

BT - Proceedings - 2023 IEEE Ural-Siberian Conference on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE Ural-Siberian Conference on Biomedical Engineering, Radioelectronics and Information Technology (USBEREIT)

Y2 - 15 May 2023 through 17 May 2023

ER -