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

2019/10/21 by Karen Adam, Adam, Karen, Adam Scholefield +3 · 1 citation
Computer Science · Engineering · Neuroscience · #Advanced Memory and Neural Computing #FOS: Electrical engineering #Neural Networks and Applications #Neural dynamics and brain function #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.1910.09413

openalex publication_date 2019/10/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

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

Cited by

• Self-Calibrating Integrate-and-Fire Time Encoding Machine

Related

• A Time Encoding approach to training Spiking Neural Networks
• Signal Coding and Perfect Reconstruction using Spike Trains
• Robust online reconstruction of continuous-time signals from a lean spike train ensemble code
• Spiking Neural Network Phase Encoding for Cognitive Computing
• Timing-Based Backpropagation in Spiking Neural Networks Without Single-Spike Restrictions