Modular Adaptive Recognition System (MARS)
MARS is a portable, integrated set of digital signal processing (DSP) software tools providing signal recognition, collection, analysis, and generation capabilities. The software can process “live” signals in the radio frequency (RF) environment, “streaming” data samples from client applications, and captured signals that have been stored to disk. It may execute on a single, general-purpose computer or in a distributed, networked computing environment, possibly employing embedded hardware. Both real-time and non real-time environments are supported.
The signal recognition capability of MARS supports the identification of signals of interest (SOIs) based on modulation characteristics and (optionally) signal content. Signal content may include transmitted bit sequences, character sequences, patterns, and linear recursive sequences (LRSs). A programmable SOI database is supported to dynamically define the characteristics of those signals that MARS is to recognize. When MARS recognizes a SOI, it measures signal parameters, records the signal to disk, and generates an “alarm” containing SOI information for further evaluation by the user.
MARS includes a set of signal analysis tools. The MARS Audio Tool supports real-time signal demodulation and audio playback. The MARS FFT Tool provides a streaming, real-time plot of a signal’s spectrum. The MARS Signal Tool is a graphical tool for viewing, analyzing, and manipulating static signal waveforms. It can process a multitude of signal file types, including those captured by its signal recognition process and those defined by the user. It can additionally create a variety of signal file types, supporting the conversion of one type to another.
MARS includes a set of signal generation capabilities, providing tools to create, view, and playback signal files based on user-provided definitions.