Efficient modulation mode recognition based on joint communication parameter estimation in non-cooperative scenarios
Year: 2024
Type: article
Abstract: Due to the neglect of the retrieval of communication parameters (including the symbol rate, the symbol timing offset, and the carrier frequency), the existing non-cooperative communication mode recognizers suffer from the generality ability degradation and severe difficulty in distinguishing a large number of modulation modes, etc. To overcome these drawbacks, this paper proposes an efficient communication mode recognizer consisting of communication parameter estimation, the constellation diagram retrieval, and a classification network. In particular, we define a 2-D symbol synchronization metric to retrieve both the symbol rate and the symbol timing offset, whereas a constellation dispersity annealing procedure is devised to correct the carrier frequency accurately. Owing to the accurate estimation of these crucial parameters, high-regularity constellation maps can be retrieved and thus simplify the subsequent classification work. Numerical results show that the proposed communication mode recognizer acquires higher classification accuracy, stronger anti-noise robustness, and higher applicability of distinguishing multiple types, which presents the proposed scheme with vast applicable potentials in non-cooperative scenarios.
Authors: Xiangdong Huang, Yimin Wang, Yanping Li, Xiaolei Wang
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $800
OpenAlex ID: W4404170448