Performance Improvement of a Receiver on a Chip (ROC)

There is an ongoing USAF need for low cost, real-time, wide bandwidth Fourier transformers capable of providing precise real-time detection and parameter estimation of a broad range of radar waveforms. A current prototype, known as MONOBIT, quantizes the DFT kernel function, and it essentially performs an approximation of the DFT in real-time. However, the current configuration of MONOBIT can only process pulsed Continuous Wave Radio Frequency signals. In this project, the extendibility of the MONOBIT to a broader class of radar waveforms will be investigated. We propose systematic approach with specific algorithmic concepts and frameworks for processing general pulse-compression radar signals including Linear/Nonlinear FM, binary Phase coded and frequency staggered CW pulse sequences. We also present new results in the proposal that signify the viability of the proposed extension of the MONOBIT. These preliminary results demonstrate future directions we plan to pursue during this effort. In addition, we suggest a new approach for radar waveform classification using MONOBIT generated slices of the Ambiguity function. The proposed research will help identify algorithmic and architectural tradeoffs that may be required to broaden the scope of MONOBIT's applicability and suggest strategies to improve its parameter estimation and pulse detection capabilities.