DeepFPC: Deep Unfolding of a Fixed-Point Continuation Algorithm for Sparse Signal Recovery from Quantized Measurements
This work addresses sparse signal recovery from quantized data, a domain-specific problem in compressed sensing, with incremental improvements over existing algorithms.
The authors tackled the problem of recovering sparse signals from quantized measurements by proposing DeepFPC, a deep neural network based on unfolding a fixed-point continuation algorithm, which achieved faster and more accurate recovery, outperforming state-of-the-art methods like FPC-l1 and 1-bit MUSIC in direction-of-arrival estimation.
We present DeepFPC, a novel deep neural network designed by unfolding the iterations of the fixed-point continuation algorithm with one-sided l1-norm (FPC-l1), which has been proposed for solving the 1-bit compressed sensing problem. The network architecture resembles that of deep residual learning and incorporates prior knowledge about the signal structure (i.e., sparsity), thereby offering interpretability by design. Once DeepFPC is properly trained, a sparse signal can be recovered fast and accurately from quantized measurements. The proposed model is evaluated in the task of direction-of-arrival (DOA) estimation and is shown to outperform state-of-the-art algorithms, namely, the iterative FPC-l1 algorithm and the 1-bit MUSIC method.