Novel Computational Kernels for Signal Processing and Digital Transformation
Mojtaba Mahdavi · 2024
Reciprocal calculation for numerical values and division calculations for arbitrary numbers constitute fundamental operations in various signal processing applications. This paper presents a flexible, low-complexity, and reconfigurable scheme for implementing reciprocal and division calculations. The key idea is to employ in memory computing (IMC) technique and perform the computations using a set of memristive devices. More specifically, the required multiplications and additions will be executed by leveraging the inherent properties of memristor devices, adhering to Ohm’s law and Kirchhoff’s current law. This innovative method not only enhances the flexibility achieved by leveraging the unique capabilities of IMC in programming the memristors but also underscores the processing efficiency. Moreover, by reducing computational complexity and latency, it outperforms traditional implementation approaches.