Datasets
Standard Dataset
CPM signal modulation order identification dataset
- Citation Author(s):
- Submitted by:
- Kaiyuan Yang
- Last updated:
- Fri, 07/12/2024 - 12:47
- DOI:
- 10.21227/53yq-bq44
- Data Format:
- License:
- Categories:
- Keywords:
Abstract
This dataset is utilized for the research of blind identification of CPM signal modulation order. The signal parameters in the dataset range as follows: modulation index from 0.125 to 1, modulation order of 2, 4, 8, pulse types including REC, RC, SRC, TFM, and GMSK, and correlation lengths of 1 to 8. The signals are oversampled by a factor of 10, transmitted through an additive white Gaussian noise (AWGN) channel, and the signal-to-noise ratio (SNR) ranges from 0 to 30dB.
Use this dataset to conduct research on blind identification of the modulation order of CPM signals. The dataset files are in *.npy format, with each file storing a segment of CPM signal data structured as 1×1025×2, where the first dimension represents the number of channels input to the network, the second dimension represents the signal length, and the third dimension stores the IQ data separately. The parent folder name of the signal data files represents the modulation order label, and the parent folder name of the labels represents the signal-to-noise ratio (Eb/N0, measured in dB).