ECCS-EPSRC: NeuroComm: Brain-Inspired Wireless Communications -- From Theoretical Foundations to Implementation for 6G and Beyond
Lead Research Organisation:
KING'S COLLEGE LONDON
Department Name: Engineering
Abstract
Current wireless systems, from Wi-Fi to 5G, have been designed by following principles that have not changed over the last 70 years. This approach has given us dependable, universal wireless connectivity solutions that can deliver any type of digital information. As computing systems substitute universal digital processors with specialised circuits for artificial intelligence (AI), and as wireless connectivity becomes an integral part of the sensing-compute-actuation fabric powered by AI, it is essential to rethink the fundamental principles underpinning the design of wireless systems. The global telecom market is estimated at around USD 850 billion, with the UK telecom industry generating around GBP 30 billion in 2020. The countries that will lead in the creation of the new technological principles and capabilities underpinning 6G will have a significant international market edge, making fundamental research on the subject a critical national policy issue.
In this context, neuromorphic sensing and computing are emerging as alternative, brain-inspired, paradigms for efficient data collection and semantic signal processing that build on event-driven measurements, in-memory computing, spike-based information processing, reduced precision and increased stochasticity, and adaptability via learning in hardware. The neuromorphic sensing and computing market was valued at USD 22.5 million in 2020, and it is projected to be worth USD 333.6 million by 2026. Current commercial use cases of neuromorphic technologies range from drone monitoring to the development of fast and accurate COVID-19 antibody testing. NeuroComm views the emergence of neuromorphic technologies as a unique opportunity for the development of efficient, integrated wireless connectivity and semantic processing -- referred to broadly as wireless cognition. Specifically, NeuroComm aims systematically addressing the integration of neuromorphic principles within an end-to-end system encompassing sensing, computing, and wireless communications.
The informational currency of neuromorphic computing is not the bit, but the timing of spikes. Neuroscientists have long studied the efficiency and effectiveness of spike-based communications in biological neurons. In the context of wireless cognition, spike-based processing and communication raise novel fundamental questions regarding optimal joint signaling and computing strategies. NeuroComm will take the approach of starting from first, information-theoretic, principles, addressing the problem of what to implement before investigating how to best deploy neuromorphic based wireless cognition. To this end, the project aims at developing an information-theoretic framework for the analysis of wireless cognition systems with neuromorphic transceivers. The efficiency of neuromorphic computing hinges on the co-design of hardware and software. NeuroComm posits that a close integration of neuromorphic computing and communications at the design stage will be needed in order to fully leverage the benefits of brain-inspired wireless cognition.
NeuroComm is a collaboration between King's College London (KCL) as lead institution and Princeton University (PU) as academic partner, along with NVIDA, Intel Labs, AccelerComm, and IBM Zurich as industrial partners. The research will build on the PIs' expertise in information theory, machine learning, communications, and neuromorphic computing to explore theoretical foundations, algorithms, and hardware implementation.
In this context, neuromorphic sensing and computing are emerging as alternative, brain-inspired, paradigms for efficient data collection and semantic signal processing that build on event-driven measurements, in-memory computing, spike-based information processing, reduced precision and increased stochasticity, and adaptability via learning in hardware. The neuromorphic sensing and computing market was valued at USD 22.5 million in 2020, and it is projected to be worth USD 333.6 million by 2026. Current commercial use cases of neuromorphic technologies range from drone monitoring to the development of fast and accurate COVID-19 antibody testing. NeuroComm views the emergence of neuromorphic technologies as a unique opportunity for the development of efficient, integrated wireless connectivity and semantic processing -- referred to broadly as wireless cognition. Specifically, NeuroComm aims systematically addressing the integration of neuromorphic principles within an end-to-end system encompassing sensing, computing, and wireless communications.
The informational currency of neuromorphic computing is not the bit, but the timing of spikes. Neuroscientists have long studied the efficiency and effectiveness of spike-based communications in biological neurons. In the context of wireless cognition, spike-based processing and communication raise novel fundamental questions regarding optimal joint signaling and computing strategies. NeuroComm will take the approach of starting from first, information-theoretic, principles, addressing the problem of what to implement before investigating how to best deploy neuromorphic based wireless cognition. To this end, the project aims at developing an information-theoretic framework for the analysis of wireless cognition systems with neuromorphic transceivers. The efficiency of neuromorphic computing hinges on the co-design of hardware and software. NeuroComm posits that a close integration of neuromorphic computing and communications at the design stage will be needed in order to fully leverage the benefits of brain-inspired wireless cognition.
NeuroComm is a collaboration between King's College London (KCL) as lead institution and Princeton University (PU) as academic partner, along with NVIDA, Intel Labs, AccelerComm, and IBM Zurich as industrial partners. The research will build on the PIs' expertise in information theory, machine learning, communications, and neuromorphic computing to explore theoretical foundations, algorithms, and hardware implementation.
Publications
Chen J
(2025)
Knowing When to Stop: Delay-Adaptive Spiking Neural Network Classifiers With Reliability Guarantees
in IEEE Journal of Selected Topics in Signal Processing
Chen J
(2024)
Agreeing to Stop: Reliable Latency-Adaptive Decision Making via Ensembles of Spiking Neural Networks.
in Entropy (Basel, Switzerland)
Croisfelt V
(2024)
Batch Selection and Communication for Active Learning with Edge Labeling
Deng W
(2025)
CSI Transfer From Sub-6G to mmWave: Reduced-Overhead Multi-User Hybrid Beamforming
in IEEE Journal on Selected Areas in Communications
Farzaneh A
(2024)
Quantile Learn-Then-Test: Quantile-Based Risk Control for Hyperparameter Optimization
in IEEE Signal Processing Letters
Guo M
(2025)
Efficient Wireless Federated Learning via Low-Rank Gradient Factorization
in IEEE Transactions on Vehicular Technology
Hou Q
(2025)
Automatic AI Model Selection for Wireless Systems: Online Learning via Digital Twinning
in IEEE Transactions on Wireless Communications
Hou Q
(2025)
What If We Had Used a Different App? Reliable Counterfactual KPI Analysis in Wireless Systems
in IEEE Transactions on Cognitive Communications and Networking
| Description | This research has pioneered neuromorphic wireless split computing - a fundamentally new approach to energy-efficient artificial intelligence for mobile and edge devices. Unlike conventional AI that processes data in fixed batches, this system mimics how biological brains work, using sparse, event-driven "spikes" to communicate information wirelessly between devices. Achievement 1: Multi-Level Spike Communication Traditional neuromorphic systems use binary spikes (on/off), but we demonstrated that multi-level spikes - carrying additional information in their amplitude - can significantly improve accuracy in wireless systems. We developed both digital and analog transmission schemes for standard OFDM wireless interfaces, discovering an optimal "sweet spot" for spike payload size that balances information richness against wireless transmission reliability. Practical demonstrations using software-defined radios and neuromorphic cameras validated the approach for real-world applications like gesture recognition. Achievement 2: Frequency-Selective Neurons for Spectral Processing We introduced balanced resonate-and-fire (BRF) neurons that naturally extract frequency-domain features from time-series signals without expensive preprocessing for neuromorphic split computing. Like tuning forks that vibrate at specific frequencies, these neurons "resonate" at preferred frequencies, making them ideal for audio and radio signal processing. This innovation eliminates the need for computationally costly Fast Fourier Transforms (FFTs), dramatically reducing both energy consumption and spike rates. Achievement 4: Comprehensive System Design and Analysis We developed complete energy consumption models accounting for both neuronal (somatic) operations and synaptic computations. The research included detailed analysis of quantization effects and practical demonstrations on both audio recognition and wireless modulation classification tasks, outperforming conventional artificial neural networks while using fewer parameters and demonstrating significant reductions in energy consumption compared to conventional approaches. . |
| Exploitation Route | Academic Routes: The research opens pathways for wireless communications researchers to integrate neuromorphic principles into 6G system design, particularly for edge intelligence and ultra-low-latency applications. Signal processing communities can build on the frequency-selective neuron models for always-on sensing applications. Machine learning researchers can explore hybrid architectures combining resonant dynamics with transformer or state-space models. Industry and Non-Academic Routes: Telecommunications companies developing 6G standards could adopt neuromorphic split computing for massive IoT deployments. Neuromorphic chip manufacturers can implement RF neuron models in next-generation hardware. IoT device manufacturers and edge AI platform providers could integrate these energy-efficient architectures for battery-powered applications in wearables, smart sensors, and autonomous systems. The work is particularly relevant for emerging standards bodies defining intelligent radio access networks (O-RAN Alliance) where distributed, event-driven processing aligns with network softwarization trends. |
| Sectors | Digital/Communication/Information Technologies (including Software) |
| Title | GitHub repo for all the papers reported |
| Description | The repository contains code for all the papers published within my group. |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2023 |
| Provided To Others? | Yes |
| Impact | The code is being used by researchers worldwide. |
| URL | https://github.com/orgs/kclip/repositories |