3.5 GHz Waveform Generation for Testing and Development of ESC Detectors
Raied M. Caromi, John Mink, Michael R. Souryal
Environmental Sensing Capability (ESC) sensors will be used in the 3.5 GHz Citizens Broadband Radio Service (CBRS) to detect and report the presence of federal incumbent radar signals in 100 MHz of spectrum. Unlike traditional radar detection schemes, ESC sensors will not have full knowledge of radar waveform parameters such as pulse repetition, pulse duration and center frequency of the incumbent radar. Furthermore, ESC sensors are expected to detect incumbent radar and identify its operational channel in the presence of interference from CBRS devices and adjacent-band emissions. This paper presents signal processing procedures and a software tool for generating ESC test waveforms. These waveforms cover multiple testing scenarios in which one or more radars operate in the presence of interference signals such as LTE TDD signals and adjacent-band radar emissions. We utilize field-measured radar waveforms acquired by NASCTN in the 3.5 GHz band with a 225 MHz sampling rate. Field-measured waveforms include channel propagation effects such as time-varying multipath fading and pulse dispersion, similar to what an actual ESC sensor will observe. We present the signal processing blocks for decimating the measured waveforms and mixing them with interference signals at specified frequency offsets. Gains are adjusted to achieve a desired SIR, defined as the ratio of the peak power of the measured radar waveform to the peak or average power of the interference. In addition, we provide an open-source software tool with a GUI to visualize the resulting waveforms and to automate the process of generating the waveforms. The tool can randomize signal parameters such as start time, frequency, and SIR. The generated waveforms are saved as 90 second, 25 MHz sampled IQ data files, and their parameters are saved JSON format. The waveforms and their parameters can be used by ESC applicants and developers for training and testing radar incumbent detection algorithms.