Over-the-air computation is the computation of mathematical functions by leveraging the signal superposition property of wireless multiple-access channels.
The computation is performed by harnessing interference in the propagation environment. All radios transmit at the same time, and the interference, naturally occurring in wireless networks, makes the intended computation, such as the arithmetic mean or sum.
Fast; with OAC, latency is reduced to a one-time cost because all radios transmit simultaneously. With OAC, interference becomes a new computational resource that has not been exploited before. It can be used for the underlying computations in data centers, sensor networks, and distributed control systems, such as swarms! Most importantly, it supports AI and ML over wireless networks; OAC can facilitate training neural networks at the edge while keeping the data private!
The project, funded by the National Science Foundation, aims to address fundamental research questions regarding the reliability, scalability, and feasibility of OAC to facilitate the development of computationally intensive applications. If you are interested in this project, please join us!