Neuromorphic Integrated Sensing and Communications

@article{Chen2022NeuromorphicIS,
  title={Neuromorphic Integrated Sensing and Communications},
  author={Jiechen Chen and Nicolas Skatchkovsky and Osvaldo Simeone},
  journal={ArXiv},
  year={2022},
  volume={abs/2209.11891}
}
—Neuromorphic computing is an emerging technology that support event-driven data processing for applications requir-ing efficient online inference and/or control. Recent work has in- troduced the concept of neuromorphic communications, whereby neuromorphic computing is integrated with impulse radio (IR) transmission to implement low-energy and low-latency remote inference in wireless IoT networks. In this paper, we introduce neuromorphic integrated sensing and communications (N-ISAC), a novel… 
[PDF] Semantic Reader
1 Citations

Figures from this paper

One Citation

Neuromorphic Wireless Cognition: Event-Driven Semantic Communications for Remote Inference

The proposed NeuroComm system is shown to significantly improve over conventional frame-based digital solutions, as well as over alternative non-adaptive training methods, in terms of time-to-accuracy and energy consumption metrics.

References

SHOWING 1-10 OF 20 REFERENCES

Neuromorphic Wireless Cognition: Event-Driven Semantic Communications for Remote Inference

The proposed NeuroComm system is shown to significantly improve over conventional frame-based digital solutions, as well as over alternative non-adaptive training methods, in terms of time-to-accuracy and energy consumption metrics.

End-to-End Learning of Neuromorphic Wireless Systems for Low-Power Edge Artificial Intelligence

An end-to-end training procedure is introduced that treats the cascade of encoder, channel, and decoder as a probabilistic SNN-based autoencoder that implements Joint Source-Channel Coding (JSCC).

Spike-Based Sensing and Communication for Highly Energy-Efficient Sensor Edge Nodes

This work presents a sensory system with analog spike-based signal processing for sensing and communication, encoding the sensory information in the pulse repetition frequency (PRF), getting rid of energy hungry A/D and D/A conversion.

1.2 nW Neuromorphic Enhanced Wake-Up Radio

A wake-up radio with a neuromorphic pre-processing system both biased in weak inversion region and achieves an energy efficiency of 1.2 pJ/bit with a minimum detectable signal of -27 dBm is proposed.

Impulse Radio Address Event Interconnects for body area networks and neural prostheses

An analysis of the capacity of a network of UWB AER nodes, which demonstrates its ability to support several thousand neurons across multiple transmitters, even at relatively high sustained firing rates, suitable for body area networks (BANs) and wireless networked neural protheses.

Integrated Sensing and Communication with Delay Alignment Modulation

This paper first derives the output signal-to-noise ratios (SNRs) for ISIfree communication and radar sensing, respectively, and proposes an efficient beamforming design for DAM-ISAC to maximize the communication SNR while guaranteeing the sensing performance.

An all-digital spike-based ultra-low-power IR-UWB dynamic average threshold crossing scheme for muscle force wireless transmission

D-ATC with regard to a fixed threshold system and an Average Threshold Crossing (ATC) demonstrating improved robustness, and introduces the thresholding algorithm verified on a dataset of 190 sEMG recorded signals.

An Overview of Signal Processing Techniques for Joint Communication and Radar Sensing

This paper provides a comprehensive overview of the state-of-the-art on JCR systems from the signal processing perspective, and a balanced coverage on both transmitter and receiver is provided for three types of J CR systems, namely, communication-centric, radar-centric and joint design and optimization.

Spiking Neural Networks—Part II: Detecting Spatio-Temporal Patterns

This letter reviews models and training algorithms for the dominant approach that considers SNNs as a Recurrent Neural Network and describes an alternative approach that relies on probabilistic models for spiking neurons, allowing the derivation of local learning rules via stochastic estimates of the gradient.

Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-based optimization to spiking neural networks

This article elucidates step-by-step the problems typically encountered when training SNNs and guides the reader through the key concepts of synaptic plasticity and data-driven learning in the spiking setting as well as introducing surrogate gradient methods, specifically, as a particularly flexible and efficient method to overcome the aforementioned challenges.