Neural-Inspired Multi-Agent Molecular Communication Networks for Collective Intelligence

TL;DR

The paper addresses energy and processing constraints in IoBNT by moving from super-capable single agents to a network of simple, diffusion-based transceivers inspired by cortical networks. It introduces a Greenberg-Hastings excitable automaton with a threshold-based firing rule and derives a mean-field fixed-point equation for the steady-state excitation vector $\mathbf{e}^*$, incorporating Gaussian molecular arrivals with mean $\mu_i[k]$ and variance $\sigma_i^2[k]$ under a memory length $L$. A second-order phase transition at the activation threshold $Q_{th}$ (about $500$) emerges, with both pairwise and collective mutual information peaking at criticality, indicating maximal information propagation at the edge of chaos. The approach is validated by stochastic simulations of $M=100$ agents and the analytical fixed-point framework, suggesting a scalable reservoir-like mechanism for molecular computing in energy-constrained IoBNT systems.

Abstract

Molecular Communication (MC) is a pivotal enabler for the Internet of Bio-Nano Things (IoBNT). However, current research often relies on super-capable individual agents with complex transceiver architectures that defy the energy and processing constraints of realistic nanomachines. This paper proposes a paradigm shift towards collective intelligence, inspired by the cortical networks of the biological brain. We introduce a decentralized network architecture where simple nanomachines interact via a diffusive medium using a threshold-based firing mechanism modeled by Greenberg-Hastings (GH) cellular automata. We derive fixed-point equations for steady-state populations via mean-field analysis and validate them against stochastic simulations. We demonstrate that the network undergoes a second-order phase transition at a specific activation threshold. Crucially, we show that both pairwise and collective mutual information peak exactly at this critical transition point, confirming that the system maximizes information propagation and processing capacity at the edge of chaos.

Figures (4)

• Figure 1: Proposed system model contrasting traditional point-to-point MC with the proposed bio-inspired network of excitable agents.
• Figure 2: Average mean and standard deviation of activity in the system with $T_s=1 \ \mathrm{s}$, $D=79.4 \ \mathrm{\mu m}$, $Q_0=100$, $r=4$, $L=5$, $\tilde{r}_1=0$, $\tilde{r}_2=0.5$
• Figure 3: Pairwise and collective mutual information in the network. Same simulation parameters as Fig. \ref{['fig:mean_std_activity']} are used.
• Figure 4: Theoretical and simulated excitation probabilities. Simulation parameters are same as Fig. \ref{['fig:mean_std_activity']} except $r_2=1$