RESUMEN
In order to realize the full potential of wireless edge artificial intelligence (AI), very large and diverse datasets will often be required for energy-demanding model training on resource-constrained edge devices. This paper proposes a lead federated neuromorphic learning (LFNL) technique, which is a decentralized energy-efficient brain-inspired computing method based on spiking neural networks. The proposed technique will enable edge devices to exploit brain-like biophysiological structure to collaboratively train a global model while helping preserve privacy. Experimental results show that, under the situation of uneven dataset distribution among edge devices, LFNL achieves a comparable recognition accuracy to existing edge AI techniques, while substantially reducing data traffic by >3.5× and computational latency by >2.0×. Furthermore, LFNL significantly reduces energy consumption by >4.5× compared to standard federated learning with a slight accuracy loss up to 1.5%. Therefore, the proposed LFNL can facilitate the development of brain-inspired computing and edge AI.
Asunto(s)
Inteligencia Artificial , Redes Neurales de la Computación , Encéfalo/fisiología , Reconocimiento en PsicologíaRESUMEN
Wireless communications technologies can support efficient healthcare services in medical and patient-care environments. However, using wireless communications in a healthcare environment raises two crucial issues. First, the RF transmission can cause electromagnetic interference (EMI) to biomedical devices, which could critically malfunction. Second, the different types of electronic health (e-Health) applications require different quality of service (QoS). In this paper, we introduce an innovative wireless access scheme, called EMI-aware prioritized wireless access, to address these issues. First, the system architecture for the proposed scheme is introduced. Then, an EMI-aware handshaking protocol is proposed for e-Health applications in a hospital environment. This protocol provides safety to the biomedical devices from harmful interference by adapting transmit power of wireless devices based on the EMI constraints. A prioritized wireless access scheme is proposed for channel access by two different types of applications with different priorities. A Markov chain model is presented to study the queuing behavior of the proposed system. Then, this queuing model is used to optimize the performance of the system given the QoS requirements. Finally, the performance of the proposed wireless access scheme is evaluated through extensive simulations.