One Dimensional Detection for Radio Signals Based on SpecUNeSt-MLP

60,000 simulated broadband power spectra generated in MATLAB to address limited real-world data for training deep CNN models. The dataset was created by houweixu xu and last updated on figshare in April 2026. It includes samples with varied modulation schemes, signal durations, carrier frequencies, and bandwidths.

Use Cases

• Training deep CNN models for radio signal detection based on simulated power spectra.
• Benchmarking model performance on modulation scheme classification (e.g., 2ASK, BPSK, QPSK).
• Studying the effect of signal-to-noise ratio (SNR) levels on detection accuracy.
• Developing one-dimensional signal processing pipelines for satellite broadband data.

Strengths

• Contains 60,000 simulated samples, providing a substantial volume for model training.
• Simulation parameters are explicitly detailed, including five modulation schemes and a range of carrier frequencies and bandwidths.
• Dataset is split into 48,000 training and 12,000 validation samples, facilitating direct use in machine learning workflows.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• The data is simulated; performance on real-world signals may vary and requires separate testing.
• Row count for the actual data files is unknown, which may limit suitability assessment.

Provenance

Source

houweixu xu on figshare

Collection Method

Generated using a simulation framework implemented in MATLAB.

Time Range

Simulation timeframe not specified.

Freshness

Last updated 2026-04-15 11:01:03; freshness should be verified.

Geography

Spatial coverage not specified.