Narasimha Vasishta Kidambi

Title: Narasimha Vasishta Kidambi and Mixed-Signal Computing-in-Memory Innovation.

Introduction

Narasimha Vasishta Kidambi is an inventor associated with Atlanta, Georgia, and Georgia Tech Research Corporation. His work includes a patent for a mixed-signal computing-in-memory system designed to solve integer-constrained least-square optimization problems.

Latest Patents

Kidambi’s patent, “Mixed Signal Computing-in-Memory Solver for Integer Constrained Least Square Problems,” describes a system that transforms integer-constrained least-square problems into energy-minimization problems addressed through Hopfield neural network dynamics. Its computing-in-memory components perform iterative optimization and assess solution quality, using analog computation within memory arrays to reduce data movement.

The system represents integer variables in binary form and generates neuron states through current comparisons. Programmable noise injection helps the optimization process escape local minima. It also supports constant-time operation for predictable latency and dynamic-time operation for faster convergence. The patent presents this approach as a way to improve speed and energy efficiency compared with conventional digital processing for these computationally difficult problems. Kidambi is credited with 1 patent.

Career Highlights

Kidambi’s listed affiliation is Georgia Tech Research Corporation. The patent reflects work at the intersection of analog computing, memory architecture, and optimization algorithms.

Collaborations

Saibal Mukhopadhyay is identified as Kidambi’s coworker. The patent’s subject matter connects their research environment with efforts to develop computing architectures for demanding optimization tasks.

Conclusion

Narasimha Vasishta Kidambi’s patented computing-in-memory solver applies analog processing and neural network dynamics to integer-constrained optimization, aiming to reduce data movement while improving computational efficiency.