Abstract:Compressed sensing is a signal processing scheme that reconstructs high-dimensional sparse signals from a limited number of observations. In recent years, various problems involving signals with a finite number of discrete values have been attracting attention in the field of compressed sensing. In particular, binary compressed sensing, which restricts signal elements to binary values $\{0, 1\}$, is the most fundamental and straightforward analysis subject in such problem settings. We evaluate the typical performance of noiseless binary compressed sensing based on $L_{1}$-norm minimization using the replica method, a statistical mechanical approach. We analyze a general setting where the elements of the observation matrix follow a Gaussian distribution, including a non-zero mean. We demonstrate that the biased observation matrix indicates more reconstruction success conditions in binary compressed sensing. Our results are consistent with the outcomes of several prior studies.

What problem does this paper attempt to address?

The paper is primarily dedicated to addressing the signal reconstruction problem in binary compressed sensing. Specifically, it evaluates the typical performance of binary compressed sensing based on the L1 norm minimization method and investigates the impact of non-zero mean elements in the observation matrix on signal reconstruction. The core contributions of the paper are: 1. **Theoretical Analysis**: Using the replica method from statistical mechanics, the typical performance of noisy binary compressed sensing based on L1 norm minimization is evaluated. 2. **Non-zero Mean Observation Matrix**: The observation matrix is analyzed under general settings, including cases with non-zero mean, and it is demonstrated that a biased observation matrix can indicate more successful reconstruction conditions in binary compressed sensing. 3. **Numerical Validation**: The theoretical results are validated through numerical simulations, and the L1 norm minimization is solved using the Gurobi optimizer. The results show that the theoretical predictions are largely consistent with the numerical simulation results. In summary, this paper aims to explore, through theoretical and numerical means, how to improve the success rate of signal reconstruction in binary compressed sensing, especially in cases where the observation matrix has a non-zero mean.

Phase transition in binary compressed sensing based on $L_{1}$-norm minimization