# Encoding and Decoding Mixed Bandlimited Signals using Spiking Integrate-and-Fire Neurons

arXiv: https://arxiv.org/abs/1910.09413 · HTML: https://mlsanity.com/paper/1910.09413/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Oct 2019.

## Built on by (most cited first, 2 of 2)

- [A Time Encoding approach to training Spiking Neural Networks](https://mlsanity.com/paper/2110.06735/) (2021)
- [How Asynchronous Events Encode Video](https://mlsanity.com/paper/2206.04341/) (2022)

## Abstract (arXiv)

Conventional sampling focuses on encoding and decoding bandlimited signals by recording signal amplitudes at known time points. Alternately, sampling can be approached using biologically-inspired schemes. Among these are integrate-and-fire time encoding machines (IF-TEMs). They behave like simplified versions of spiking neurons and encode their input using spike times rather than amplitudes. Moreover, when multiple of these neurons jointly process a set of mixed signals, they form one layer in a feedforward spiking neural network. In this paper, we investigate the encoding and decoding potential of such a layer. We propose a setup to sample a set of bandlimited signals, by mixing them and sampling the result using different IF-TEMs. We provide conditions for perfect recovery of the set of signals from the samples in the noiseless case, and suggest an algorithm to perform the reconstruction.