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

被引:31
作者
Chen, Jiechen [1 ]
Skatchkovsky, Nicolas [1 ]
Simeone, Osvaldo [1 ]
机构
[1] Kings Coll London, Learning & Informat Proc Lab, Kings Commun, London WC2R2LS, England
基金
英国工程与自然科学研究理事会; 欧洲研究理事会;
关键词
Sensors; Neuromorphics; Wireless communication; Wireless sensor networks; Receivers; Encoding; Sensor systems; Neuromorphic computing; spiking neural networks; semantic communications; SPIKING NEURAL-NETWORKS; IMPULSE RADIO; LOIHI; POWER;
D O I
10.1109/TCCN.2023.3236940
中图分类号
TN [电子技术、通信技术];
学科分类号
0809 ;
摘要
Neuromorphic computing is an emerging computing paradigm that moves away from batched processing towards the online, event-driven, processing of streaming data. Neuromorphic chips, when coupled with spike-based sensors, can inherently adapt to the "semantics" of the data distribution by consuming energy only when relevant events are recorded in the timing of spikes and by proving a low-latency response to changing conditions in the environment. This paper proposes an end-to-end design for a neuromorphic wireless Internet-of-Things system that integrates spike-based sensing, processing, and communication. In the proposed NeuroComm system, each sensing device is equipped with a neuromorphic sensor, a spiking neural network (SNN), and an impulse radio (IR) transmitter with multiple antennas. Transmission takes place over a shared fading channel to a receiver equipped with a multi-antenna impulse radio receiver and with an SNN. In order to enable adaptation of the receiver to the fading channel conditions, we introduce a hypernetwork to control the weights of the decoding SNN using pilots. Pilots, encoding SNNs, decoding SNN, and hypernetwork are jointly trained across multiple channel realizations. The proposed 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.
引用
收藏
页码:252 / 265
页数:14
相关论文
共 44 条
[21]  
lava-nc.org, LAV SOFTW FRAM
[22]   A neuro-inspired artificial peripheral nervous system for scalable electronic skins [J].
Lee, Wang Wei ;
Tan, Yu Jun ;
Yao, Haicheng ;
Li, Si ;
See, Hian Hian ;
Hon, Matthew ;
Ng, Kian Ann ;
Xiong, Betty ;
Ho, John S. ;
Tee, Benjamin C. K. .
SCIENCE ROBOTICS, 2019, 4 (32)
[23]  
Lichtsteiner P., 2006, 2006 IEEE INT SOL ST, P2060
[24]  
Liu YS, 2021, IEEE INT WORK SIGN P, P31, DOI [10.1109/SPAWC51858.2021.9593238, 10.1145/3472307.3484167]
[25]   Brain-inspired computing needs a master plan [J].
Mehonic, A. ;
Kenyon, A. J. .
NATURE, 2022, 604 (7905) :255-260
[26]   Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-based optimization to spiking neural networks [J].
Neftci, Emre O. ;
Mostafa, Hesham ;
Zenke, Friedemann .
IEEE SIGNAL PROCESSING MAGAZINE, 2019, 36 (06) :51-63
[27]   Semantic-Effectiveness Filtering and Control for Post-5G Wireless Connectivity [J].
Popovski, Petar ;
Simeone, Osvaldo ;
Boccardi, Federico ;
Gunduz, Deniz ;
Sahin, Onur .
JOURNAL OF THE INDIAN INSTITUTE OF SCIENCE, 2020, 100 (02) :435-443
[28]  
Qin ZJ, 2022, Arxiv, DOI [arXiv:2201.01389, DOI 10.48550/ARXIV.2201.01389]
[29]   Low-Power Neuromorphic Hardware for Signal Processing Applications: A review of architectural and system-level design approaches [J].
Rajendran, Bipin ;
Sebastian, Abu ;
Schmuker, Michael ;
Srinivasa, Narayan ;
Eleftheriou, Evangelos .
IEEE SIGNAL PROCESSING MAGAZINE, 2019, 36 (06) :97-110
[30]   Events-to-Video: Bringing Modern Computer Vision to Event Cameras [J].
Rebecq, Henri ;
Ranftl, Rene ;
Koltun, Vladlen ;
Scaramuzza, Davide .
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, :3852-3861