QuanBioTrust: a quantum-enhanced bio-inspired trust framework for next-generation WBANs

Abstract

Wireless Body Area Networks (WBANs) face critical challenges in achieving energy efficiency, security, and real-time reliability for continuous biomedical monitoring, as conventional protocols are hindered by energy constraints, vulnerability to cyberattacks, and high latency. To address these limitations, this paper proposes QuanBioTrust, a novel, integrated framework that fuses quantum-enhanced optimization, bio-inspired intelligence, and a dynamic zero-trust security model to create a scalable, secure, and self-sustaining WBAN architecture. QuanBioTrust leverages Quantum Particle Swarm Optimization (QPSO) with quantum entanglement to optimize cluster head (CH) selection by concurrently evaluating residual energy, sink proximity, trust score, and bio-energy harvesting potential (20–50 µW), reducing computational complexity to ~ 500 operations per round, a 28–45% improvement over HCEL (~ 700 ops) and EGWO (~ 1000 ops). Inspired by natural systems, the framework employs flocking algorithms for mobility-adaptive clustering, firefly synchronization for efficient multi-hop fractal routing (3–5 paths), and slime mold-inspired data fusion to achieve 50–70% compression of ECG packets (200-bit → 80-bit), drastically reducing transmission energy. The framework maintains a Packet Delivery Ratio (PDR) of 97% under normal conditions (90% under attack) and an end-to-end latency of 48 ms (185 ms under attack), well below the 250 ms threshold for real-time monitoring, outperforming EAFST (1000/1200 ms) and rivaling EDC-ER (50/180 ms). This quantum-biological fusion establishes a new paradigm for next-generation WBANs, offering a robust, energy-efficient, and resilient solution for secure, long-term, real-time healthcare monitoring.

Introduction

Wireless Body Area Networks (WBANs) offer a revolutionary approach to wireless communication, connecting miniature sensor nodes, deployed on, within, or around the human body, to external systems like medical servers or wearable devices¹. These compact, low-power devices, typically a few centimeters in size, integrate sensors, microcontrollers, and transceivers operating in the Industrial, Scientific, and Medical (ISM) or ultra-wideband (UWB) spectra². They monitor physiological parameters such as heart rate, blood oxygen saturation (SpO2), body temperature, glucose levels, and electromyography (EMG) signals, transmitting data to a central sink, often a smartphone, smartwatch, or gateway, for real-time analysis³. Employing short-range communication (1–2 m) via protocols like IEEE 802.15.6, WBANs prioritize minimal power use and reliable transmission despite body-induced challenges, including tissue absorption and movement artifacts⁴. In healthcare, they enable remote monitoring of chronic conditions, early anomaly detection, and personalized treatments through closed-loop systems like insulin pumps, improving quality of life and reducing costs by decreasing hospital visits⁵.

Despite their potential, WBANs face a persistent challenge: energy efficiency, which limits their operational lifespan and adoption⁶. Sensor nodes, powered by small batteries (10–100 mAh) or emerging harvesting techniques, operate under tight energy constraints due to their size and the impracticality of replacing batteries, especially in implants. Energy demands arise from sensing, processing, and, most critically, wireless communication, where transceivers consume 10–100 mW over 0.1–2 m, amplified by path loss (exponent 2–4) and tissue interference causing attenuation and multipath fading⁷. Inefficient energy use shortens network lifetimes to hours or days, far below the weeks or months needed, risking packet loss that delays critical alerts, such as in cardiac monitoring, and compromising safety. Dynamic topology from body motion further increases energy needs by 30–50% due to retransmissions, while security and low-latency requirements add overhead, creating a complex trade-off that traditional protocols, designed for static networks, fail to address effectively⁸.

Efforts to tackle WBAN energy challenges have progressed from basic to advanced strategies, yet they fall short of meeting body-centric demands⁹. Early approaches like LEACH¹⁰ used static clustering and rotated cluster heads (CHs) to balance energy load, but their fixed structure and simplistic CH selection overlook mobility-induced shifts (up to 50% distance variation) and tissue interference (20–40 dB path loss), limiting longevity[¹¹]. Bio-inspired methods like Particle Swarm Optimization (PSO) optimize CH selection, cutting transmission energy by 15–25%, while Ant Colony Optimization (ACO) reduces costs by 10–20% through multi-hop routing¹². However, their computational complexity and poor adaptability to rapid topology changes hinder real-time performance¹³. Trust-based models detect malicious nodes with 85–95% accuracy, enhancing security, but their overhead negates energy gains¹⁴. These incremental improvements, extending lifetimes by hours or boosting delivery by single-digit percentages, lack the holistic integration needed to address WBAN unique energy, security, and scalability demands comprehensively.

To address the critical challenges of efficiency, security, and scalability in next-generation Wireless Body Area Networks, we propose QuanBioTrust: A Quantum-Enhanced Bio-Inspired Trust Framework for Next-Generation WBANs. Leveraging Quantum Particle Swarm Optimization with quantum entanglement, QuanBioTrust optimizes cluster head (CH) selection by evaluating residual energy, sink proximity, connectivity, trust, and bio-energy harvesting potential (20–50 µW/cm²), achieving 20–30% lower computational complexity compared to traditional methods. Bio-inspired mechanisms, including adaptive clustering and synchronized multi-hop routing, draw from natural systems to enable 40% energy savings and 15–20% higher packet delivery rates. A predator–prey trust model, integrated with quantum key distribution (QKD), ensures robust security by isolating malicious nodes and enabling zero-trust encryption under threats, while fractal redundancy enhances path resilience. By combining quantum-enhanced optimization, bio-energy harvesting, and dynamic trust management, QuanBioTrust establishes a scalable and secure standard for next-generation WBANs, meeting the evolving demands of real-time biomedical monitoring.

The major contributions of this paper are as follows:

• Quantum-enhanced optimization framework We propose a Quantum Particle Swarm Optimization method, enriched with entanglement, to optimize cluster head selection by evaluating residual energy, sink distance, connectivity, trust levels, and harvesting potential concurrently. This approach slashes computational overhead by 20–30% compared to classical techniques, establishing a new benchmark for energy-efficient, dynamic clustering in WBANs.

• Bio-inspired adaptive mechanisms QuanBioTrust leverages nature-inspired strategies, flocking for mobility-responsive clustering, slime mold-driven data fusion to reduce redundancy, and firefly synchronization for efficient routing, delivering up to 40% energy savings and 15–20% higher data delivery rates. This robust, adaptive framework addresses WBAN-specific challenges like topology shifts and interference with precision.

• Integrated trust and security system We introduce a predator–prey trust model paired with quantum encryption in a zero-trust architecture, dynamically isolating malicious nodes while securing data with latency increases below 50 ms. This scalable solution balances energy efficiency and resilience, offering a comprehensive defense against sophisticated threats in WBANs.

The remainder of this paper is organized as follows: “Related works” Section reviews the related works, while “System Model” Section outlines the system model and design goals. “QuanBioTrust” Section introduces the proposed QuanBioTrust framework in detail. “Security Analysis of QuanBioTrust” Section presents the security analysis of QuanBioTrust, and “Experimental Setup” Section evaluates its performance using relevant metrics. “Result and discussion” Section discusses the results and key observations, followed by “Discussion and Limitations of QuanBioTrust” Section, which highlights the limitations and provides further discussion. Finally, “Conclusion and Future Works” Section concludes the paper and outlines potential directions for future research. Table 1 resume all Key Terms used in this paper.

Table 1.

Glossary of Key Terms.

Acronym	Full form	Description
ACO	Ant Colony Optimization	A bio-inspired optimization algorithm used in some WBAN protocols for multi-hop routing, reducing transmission energy by 10–20% but limited by computational complexity in dynamic topologies
CH	Cluster Head	A sensor node dynamically selected via QPSO in QuanBioTrust to manage intra-cluster data fusion (50–70% compression, 10 nJ/bit) and inter-cluster routing to the sink, based on residual energy, proximity (0.5–2 m), connectivity (3–5 neighbors), trust (0.5–0.8), and harvesting potential (20–50 µW)
ECG	Electrocardiogram	A physiological signal (360 Hz, 11-bit resolution) monitored by WBAN sensor nodes to detect cardiac conditions (e.g., arrhythmias), processed in QuanBioTrust simulations using the MIT-BIH dataset, generating 200-bit packets at 10 Hz
IEEE 802.15.6	Institute of Electrical and Electronics Engineers Standard 802.15.6	A communication standard for WBANs, operating in the 2.4 GHz ISM band or UWB (3–10 MHz), used by QuanBioTrust for short-range (0.1–2 m) low-power transmission with 20–40 dB tissue attenuation
ISM	Industrial Scientific and Medical	Radio frequency bands (e.g., 2.4 GHz) used by WBAN nodes for low-power, short-range communication, as specified in IEEE 802.15.6, subject to tissue interference and path loss (exponent 2–4)
PDR	Packet Delivery Ratio	The percentage of data packets successfully delivered to the sink, with QuanBioTrust achieving 97% under normal conditions and 90% under attack, surpassing HCEL (94%/85%) and EDC-ER (92%/80%) in 50-node simulations
PSO	Particle Swarm Optimization	A bio-inspired optimization algorithm used in some WBAN protocols for cluster head selection, reducing transmission energy by 15–25% but less efficient than QuanBioTrust QPSO due to higher computational complexity
QKD	Quantum Key Distribution	A quantum-based encryption method in QuanBioTrust using 32-bit entangled keys to secure data transmission, achieving 99.9% eavesdropping protection with 20–25 nJ/bit overhead, critical for zero-trust security
QNN	Quantum Neural Network	A quantum-inspired neural model in QuanBioTrust for trust reasoning, enhancing detection of variable data patterns (e.g., ECG spikes) compared to EDC-ER LSTM, contributing to 95.3% threat reduction
QPSO	Quantum Particle Swarm Optimization	A quantum-inspired optimization algorithm in QuanBioTrust for cluster head selection, evaluating residual energy, sink proximity, connectivity, trust, and harvesting potential (~ 500 ops/round, 15 nJ/bit), reducing complexity by 20–30% compared to PSO
SNR	Signal-to-Noise Ratio	A measure of signal quality in WBAN communications, typically 10–20 dB in QuanBioTrust, accounting for tissue attenuation (20–40 dB) and mobility-induced variations (± 50% distance)
SpO2	Blood Oxygen Saturation	A physiological parameter monitored by WBAN sensor nodes, challenging for protocols like EDC-ER due to variability, but effectively handled by QuanBioTrust quantum compression for real-time monitoring
UWB	Ultra-Wideband	A low-power, high-bandwidth (3–10 MHz) communication protocol used by QuanBioTrust nodes (IEEE 802.15.6) for short-range (0.1–2 m) data transmission, resilient to body-induced interference
WBAN	Wireless Body Area Network	A network of 10–50 sensor nodes deployed on or within the human body to monitor physiological parameters (e.g., ECG, glucose), coordinated by a sink node, with QuanBioTrust achieving 12,800-round lifetime and 97% PDR
ZTT	Zero-Trust Topology	A security architecture in QuanBioTrust assuming no node is inherently trusted, using a predator–prey trust model and QKD to achieve 95.3% threat reduction, isolating malicious nodes in 1–4 rounds

Related works

This section provides a detailed analysis of previous papers, highlighting their contributions, results, and limitations. Table 2 offers a quick comparison of all works, making it easier to position QuanBioTrust as a novel and comprehensive solution.

Table 2.

Comparative Analysis of WBAN Protocols.

Ref	Proposed method	Strengths	Weaknesses
15	HCEL (Hybrid Clustering Approach for Extending WBAN Lifetime)	Hybrid clustering enhances energy efficiency and stability. Significant delay reduction	Complex parent node selection may increase computational overhead. Limited scalability discussion
16	EEMR (Energy Enrichment Multi-Hop Routing)	High throughput and energy efficiency. Extended network lifetime	Complexity of EFBOA and DLH-LD may increase computational overhead. Limited scalability analysis
17	Enhanced Gray Wolf Optimization (EGWO)	High throughput and energy efficiency. Reduced computation time. Effective trust management	Computational complexity of EGWO may limit scalability. Lacks real-world testing
18	Fuzzy Logic and Nash Equilibrium Clustering	High throughput and packet delivery ratio. Energy efficiency. Secure data transmission	Complexity of fuzzy logic and Nash equilibrium may increase overhead. Limited discussion on extreme mobility
19	LF2ZO with DSAC and MSS	High packet delivery ratio and energy efficiency. Reduced overhead and cluster building time	Complexity of LF2ZO and MSS may increase computational demands. Limited scalability analysis
20	EAFST (Energy Aware Forwarder Selection Technique)	High stability, throughput, and energy efficiency. Low path loss and delay	Higher propagation delay due to multi-hop (0.001 s vs. 0.0005 s for direct). Lacks mobility consideration
21	EDC-ER (Efficient Data Compression for Energy Reduction)	98% energy saving. Low computational complexity. High compression ratio for repetitive data	Poor prediction accuracy for variable signals (e.g., oxygen). Needs improvement for non-repetitive data

The study in¹⁵ presents HCEL (Hybrid Clustering Approach for Extending WBAN Lifetime), an energy-efficient routing protocol for Wireless Body Area Networks that blends static and dynamic clustering. HCEL optimizes parent node selection using a utility function based on residual energy, sink proximity, and received signal strength indicator (RSSI), constrained by energy thresholds and serving node limits. Operating in three phases, network deployment, parent node selection, and data transmission, it balances energy use, extends network lifetime, and enhances stability. Simulations in NetworkX reveal HCEL edge over SIMPLE and ERRS, reducing end-to-end delay by 56% and 40%, respectively, while improving stability period (23% vs. 20% for SIMPLE and 21% for ERRS), residual energy, packet delivery ratio, and throughput. HCEL excels in energy efficiency and delay reduction, making it reliable for WBANs. However, its complex parent node selection may increase computational overhead, and its scalability in larger networks remains unaddressed. SIMPLE offers a straightforward multihop approach with decent stability (20%) but suffers from higher delays (0.8 s) and less effective energy balancing. ERRS provides moderate stability (21%) yet lags with higher delays (0.5 s) and lower packet delivery ratios, reducing its reliability.

The research in¹⁶ introduces the Energy Enrichment Multi-Hop Routing (EEMR) protocol, a metaheuristic approach to boost WBAN network lifetime and routing efficiency by tackling path loss, energy consumption, and transmission barriers. EEMR operates in two phases: the Enhanced Flower Bee Optimisation Algorithm (EFBOA) minimizes energy use in clustering and cluster head selection, while Dynamic Local Hunting and Location Discarding (DLH-LD) optimizes routes. MATLAB and Network Simulator tests against M-ATTEMPT, EECR, and EMPR show EEMR achieving 95% throughput at 10,000 iterations, 30% energy efficiency (vs. 50% for M-ATTEMPT, 65% for EECR, 80% for EMPR), 20% path loss (vs. 45%, 55%, 60%), and 90% network lifetime (vs. 60%, 45%, 15%). EEMR high throughput and energy efficiency suit healthcare applications like heart rate monitoring, though EFBOA and DLH-LD complexity may raise overhead, and scalability is underexplored. M-ATTEMPT maintains stable 30% throughput but falters with 45% path loss and 50% energy use. EECR offers 75% throughput yet struggles with 55% path loss and 65% energy consumption. EMPR simplicity is overshadowed by 60% path loss, 80% energy use, and a mere 15% lifetime.

The paper in¹⁷ proposes an Enhanced Gray Wolf Optimization (EGWO) method for WBANs in healthcare, optimizing cluster head selection, energy efficiency, and trustworthiness. It features network initialization, comprehensive trust (CT) calculations (direct, indirect, energy-based), and distance-driven energy models, with EGWO balancing trust and efficiency via a fitness function. MATLAB R2022b simulations outperform RRP, MOMHR, AODV using Graphs, and Traditional AODV, yielding 18,430 kbps throughput at 100 rounds (vs. 5000, 2400, 14,400, 1500 kbps), 48.14-s computation time (vs. 100, 3500, 400 secs), higher residual energy, and improved stability (70 live nodes at 1000 rounds vs. 30 dead). An adaptive penalty coefficient (theta) adjusts to abnormal behavior, enhancing trust management. EGWO excels in throughput and efficiency, ideal for healthcare monitoring, but its complexity may limit scalability, and real-world validation is absent. RRP offers moderate throughput (5000 kbps) but higher computation time (100 secs). MOMHR simplicity yields low throughput (2400 kbps). AODV using Graphs achieves 14,400 kbps but with moderate efficiency, while Traditional AODV lags with 1500 kbps and 3500-s computation.

The study in¹⁸ presents a clustering control scheme for WBANs using fuzzy logic and Nash equilibrium to enhance scalability, stability, and resource management. It optimizes cluster head (CH) selection, relay and channel choices, and secure transmission via lightweight Diffie–Hellman key exchange and reinforcement learning, outperforming TAFLR, EQRSRL, and SEBA. It achieves 3200 kbps throughput at 1800 s (vs. 2800, 2600, 2900 kbps), 93% packet delivery ratio (vs. 85%, 80%, 88%), 0.43 J energy use at 350 nodes (vs. 0.5, 0.48, 0.45 J), 95% cluster efficiency, and 0.31-s delay. This model shines in throughput, delivery, and security, fitting medical needs, though integrating fuzzy logic and Nash equilibrium may raise overhead, and extreme mobility is undiscussed. TAFLR offers 2,800 kbps throughput and 85% delivery but higher energy use (0.5 J). EQRSRL balances metrics but lags with 2600 kbps and 80% delivery. SEBA 2900 kbps and 88% delivery are offset by scalability challenges.

The authors in ¹⁹ propose an energy-efficient WBAN routing protocol using the Levy Flight and Frilled Lizard Optimization (LF2ZO) algorithm, Dynamic Self-executing Active Cross-propagated (DSAC) method for clustering, and Momentum Sparrow Search (MSS) for cluster head (CH) selection, targeting reliability and longevity in IoT healthcare. Two-hop neighbor clustering optimizes transmission distances, while LF2ZO prioritizes high-energy nodes. NS-2 simulations surpass PMQoSR, WMCP, HBCM, and HBCMHBO, with 97% PDR (vs. 86%, 89%, 88%, 90%), 6.2 J energy use at 500 nodes (vs. 7.4, 7.2, 7.0, 6.8 J), 15-s cluster building time (vs. 75, 65, 30, 25 secs), 80-s cluster lifetime (vs. 30, 35, 60, 65 secs), and 4,000-byte overhead (vs. 20,000, 18,000, 10,000, 8000 bytes). It excels in PDR and efficiency, ideal for healthcare, though LF2ZO and MSS complexity may strain computation, and scalability analysis is limited. PMQoSR 86% PDR is offset by 7.4 J energy and 75-s build time. WMCP 89% PDR comes with high overhead (18,000 bytes). HBCM and HBCMHBO offer moderate gains but higher energy and overhead.

The work in²⁰ introduces the Energy Aware Forwarder Selection Technique (EAFST) for WBANs, enhancing stability and lifetime by arranging sensors in concentric rings around a coordinator. High-priority nodes transmit directly, while medium- and low-priority nodes use forwarders chosen by residual energy, queue size, and path loss, with roles rotated to save energy. Simulations outperform DFST, SIMPLE, and PCRP, with 56% better stability (0 dead nodes at 2000 rounds vs. 2, 4, 6), 51% higher throughput (6000 kbps at 14,000 rounds vs. 4000, 3000, 2000 kbps), 68% longer lifetime (12,000 rounds vs. 7000, 6000, 5000), and 38% less energy use (10 J vs. 20, 30, 40 J). EAFST also reduces path loss (50 dB vs. 100, 150, 200 dB) and delay (0.001 s vs. 0.003, 0.004, 0.005 s). It shines in stability and efficiency for healthcare, though multi-hop delays and lack of mobility focus are drawbacks. DFST, SIMPLE, and PCRP offer moderate performance but higher energy use and lower stability.

The Efficient Data Compression for Energy Reduction (EDC-ER) method in²¹ for WBANs combines Long Short-Term Memory (LSTM) networks and Run Length Encoding (RLE) to achieve 98% energy savings in biosensors tracking blood pressure, respiration, oxygen (SPO2), and heart rate. It operates in four steps: training an LSTM network on time-series data, generating accurate predictions, computing prediction errors to reduce data size, and applying RLE to compress errors, outperforming Huffman (97%), Arithmetic (80–85%), and LZW (80–85%). EDC-ER excels in predicting repetitive signals like BPDias and respiration with low errors, though it falters with variable oxygen data, showing larger test errors. Its strengths, low complexity, high compression, and efficiency for repetitive data, suit WBAN sensors, but its weakness in handling non-repetitive signals suggests a need for refinement.

Related work on Wireless Body Area Network protocols, spanning references^15–21, lays bare a series of critical gaps that threaten their real-world healthcare viability. Mobility remains a blind spot, with protocols like EAFST²⁰ offering no countermeasures for node shifts due to postural changes, think arm swings or torso twists, leaving communication hanging by a thread. Computational complexity and scalability falter as HCEL¹⁵ wrestles with intricate parent node selection, EEMR¹⁶ juggles the elaborate EFBOA and DLH-LD algorithms, LF2ZO¹⁹ layers on DSAC and MSS, and fuzzy logic with Nash equilibrium¹⁸ piles up overhead, all while dodging serious scalability scrutiny for sprawling networks. Variable data prediction stumbles in EDC-ER²¹, where a fixation on repetitive patterns, fine for steady heartbeats, trips over erratic signals like oxygen (SPO2), exposing weak accuracy under pressure. Trust and security languish in most protocols^{15,16,19–21}, save for EGWO¹⁷ with its thorough trust calculations and¹⁸ with its nimble Diffie–Hellman key exchange, leaving sensitive patient data dangling vulnerably. Energy harvesting is a glaring omission across the board, squandering the potential to stretch lifetimes beyond EEMR 90%¹⁶ or EAFST 12,000 rounds²⁰. Real-world testing is nowhere to be found, with all protocols clinging to simulations (e.g., MATLAB, NS-2, NetworkX) as EGWO¹⁷ exemplifies, sidestepping practical proof. Multi-hop designs in EAFST²⁰ (0.001 s vs. 0.0005 s direct) and HCEL¹⁵ drag down speed with propagation delays, a liability for urgent healthcare needs. QuanBioTrust steps into these breaches with quantum-inspired flair: quantum annealing flexes routing to match mobility dance; quantum neural networks (QNNs) tackle variable data twists, outpacing EDC-ER; quantum key distribution (QKD) fortify security and trust beyond^17,18, piezoelectric and thermoelectric harvesting powers longevity, and entanglement-inspired routing cuts delays to a whisper, shows promise as a robust framework for secure WBANs healthcare monitoring.

System model

In this section, we outline the foundational system of the QuanBioTrust protocol by presenting the network architecture and subsequently formalizing the energy optimization problem that underpins its design.

Network model

The QuanBioTrust network model is designed as a heterogeneous Wireless Body Area Network comprising a set of 10–50 miniature sensor nodes strategically deployed on or within the human body, complemented by a single central sink node for data aggregation and external communication, as illustrated in Fig. 1 of the manuscript. Each sensor node is a compact, low-power device (dimensions ~ 1–2 cm², weight < 5 g) equipped with advanced sensing, processing, and wireless transmission capabilities tailored for monitoring physiological parameters such as electrocardiogram (ECG) signals (360 Hz, 11-bit resolution), blood glucose levels, body temperature, or oxygen saturation (SpO2). These nodes operate under stringent energy constraints, with initial battery capacities of 0.05–0.15 J (equivalent to ~ 10–100 mAh at 1.5 V) and power budgets of 100–500 µW, necessitating ultra-low-power microcontrollers (e.g., ARM Cortex-M0, ~ 10 mW) and transceivers compliant with the IEEE 802.15.6 standard (2.4 GHz ISM band or ultra-wideband (UWB), 3–10 MHz bandwidth)[²²]. Communication occurs over short distances (0.1–2 m), contending with significant body-induced challenges, including tissue attenuation (20–40 dB due to muscle, fat, and skin layers), path loss (exponent n = 2–4), and mobility-driven topology changes (± 50% inter-node distance variation due to walking, breathing, or posture shifts). To supplement battery power, each node integrates bio-energy harvesting mechanisms, such as piezoelectric generators (e.g., PVDF films, ~ 20 µW/cm² from limb motion) and thermoelectric generators (e.g., Bi₂Te₃-based, ~ 30 µW/cm² from body heat at ~ 310 K), yielding 20–50 µW depending on node placement (e.g., wrist, torso) and user activity level, offsetting ~ 30–40% of operational energy costs (e.g., 3–4 × 10⁶ nJ over 10,000 rounds)²³.

Fig. 1.

The network architecture.

The sink node, typically a more capable device like a smartphone or wearable hub (e.g., positioned at the waist), serves as the central data collection and processing hub with a higher energy capacity (0.5–2 J, equivalent to ~ 100–500 mAh) and enhanced computational resources (e.g., ARM Cortex-A series, > 1 GHz). It interfaces with external systems, such as medical servers or cloud platforms, via Wi-Fi or cellular networks, ensuring secure data transmission for real-time healthcare applications (e.g., arrhythmia detection, insulin dosing). The sink coordinates the WBAN by receiving aggregated data from dynamically selected cluster heads (CHs), which are sensor nodes chosen via Quantum Particle Swarm Optimization based on residual energy, proximity to the sink (0.5–2 m), connectivity degree (3–5 neighbors), trust score (0.5–0.8), and harvesting potential. CHs perform intra-cluster data fusion using slime mold-inspired algorithms (50–70% compression, reducing 200-bit ECG packets to 80–100 bits, ~ 10 nJ/bit) and relay data to the sink through multi-hop paths using firefly-synchronized fractal routing (3–5 paths, ~ 10 nJ/bit), ensuring resilience against link failures due to body mobility. The network topology is highly dynamic, with nodes forming adaptive clusters (90% coverage within 0.5 m of a CH) that reconfigure every 100–500 rounds to accommodate topology changes caused by body movements (e.g., ± 50% distance variation during walking at 1–2 m/s). Intra-cluster communication uses single-hop transmissions (0.1–0.5 m, 100 nJ/bit), while inter-cluster routing to the sink employs multi-hop paths (up to 3 hops, 100 pJ/bit/m²), optimized by QPSO to minimize energy-intensive transmissions and maintain low latency (< 50 ms normal, < 200 ms under attack) and high PDR (97% normal, 90% under attack).

Energy‑efficient problem

Consider a WBAN with m sensor nodes and a single sink node . Each node has an initial energy , a bio-energy harvesting rate , and a trust score . The network operates over a time horizon T, divided into discrete rounds , during which nodes sense data, form clusters, and relay aggregated information to the sink via cluster heads (CHs). The energy consumption of node at time t, denoted , comprises several components: sensing energy , processing energy (e.g., for data fusion), transmission energy (to CH or sink), and maintenance energy (e.g., for re-clustering or encryption). The energy update for each node is given by:

where represents the duration of each time slot, and is bounded by , reflecting battery capacity limits, accounting for the finite capacity of the node energy storage.

The network lifetime is defined as the duration until either the first node exhausts its energy or a critical proportion of nodes fail, whichever occurs first, as either event disrupts sufficient data coverage and network reliability. The cumulative energy expenditure across all nodes over the time horizon T is:

Transmission energy constitutes the primary drain on node resources, influenced by both the distance to the receiving node and the volume of data transmitted. When node sends bits to a CH or sink at distance , the transmission energy is calculated as:

In this expression,  = 50 nJ/bit denotes the energy required by the transceiver electronics,  = 100 pJ/bit/m² represents the amplifier energy per bit per square meter, and , ranging between 2 and 4, adjusts dynamically to reflect free-space propagation or multi-path fading, depending on a distance threshold.

Optimization objective

The energy-efficient problem within the QuanBioTrust protocol is framed as a constrained optimization task designed to simultaneously pursue two interconnected objectives: maximizing the network lifetime and minimizing the total energy consumption across all nodes throughout the operational period. Network lifetime represents the duration the WBAN remains functional, specifically defined as the earliest time at which any node exhausts its energy, or a significant fraction of nodes fail, whichever occurs first, as this marks the point where data coverage and reliability begin to degrade critically. Simultaneously, the total energy consumption , calculated as the sum of all nodes’ energy expenditures over T, must be minimized to ensure efficient resource utilization. This dual-objective optimization reflects the core challenge of sustaining WBAN operation under stringent energy constraints while meeting performance requirements. Formally, the optimization problem is expressed as:

subject to minimizing .

This formulation reflects a multi-objective optimization where and are inherently linked, extending lifetime requires judicious energy management, yet excessive conservation might compromise data delivery or security. To make this tractable, QuanBioTrust balances these goals through a set of practical constraints that ensure operational viability:

• Energy availability Every node must maintain for all i and , preventing premature node failure that could fragment the network, especially critical in sparse deployments where each node contribution to coverage is significant.

• Data reliability The packet delivery ratio (PDR), measuring successful data transmissions, must remain above 95% at each round, ensuring that vital health data (e.g., heart rate or glucose levels) reaches the sink consistently, as even a 5% loss could miss critical anomalies.

• Latency bound End-to-end delay must stay below 250 ms, a stringent requirement for real-time applications where delays beyond this threshold could hinder timely medical responses, such as in arrhythmia detection.

• Trust requirement Active nodes must uphold a trust score above a dynamic threshold (e.g., 0.5–0.8), safeguarding against malicious behavior like data tampering, which is vital in healthcare contexts handling sensitive patient information.

• Cluster coverage At least 90% of nodes must stay within 0.5 m of a CH at all times, accounting for mobility and ensuring robust intra-cluster communication despite posture changes.

QuanBioTrust

QuanBioTrust is an advanced, quantum-enhanced, bio-inspired trust framework designed to address the critical challenges of energy efficiency, security, and adaptability in next-generation Wireless Body Area Networks through a seven-step, event-driven architecture that seamlessly integrates quantum computing principles, biological intelligence, and zero-trust security. The process begins with network initialization, where each sensor node is provisioned with an initial energy reserve (0.05–0.15 J), bio-energy harvesting capability (20–50 µW from body heat and motion), and a trust score of 0.9, establishing a dynamic foundation for role assignment. This is followed by quantum-enhanced clustering using Quantum Particle Swarm Optimization (QPSO) with entanglement [²⁴], where nodes are modeled as qubits in superposition to evaluate residual energy, proximity to the sink, connectivity, trust, and harvesting potential, enabling energy-efficient and mobility-adaptive cluster head selection. A bio-inspired predator–prey trust model is then employed to continuously monitor node behavior, identifying and isolating malicious “predator” nodes based on packet delivery anomalies and dynamic trust thresholds, ensuring a secure network backbone with minimal overhead. Within clusters, a slime mold-inspired fusion algorithm reduces data redundancy, followed by quantum compression that leverages multi-qubit encoding to shrink ECG packets by 50–70%, with the compressed data entangled with the sink’s quantum state for secure transmission. Data is then routed to the sink via fractal multi-hop routing, which creates 3–5 self-similar, redundant paths, synchronized using firefly-inspired algorithms to minimize collisions and enhance fault tolerance against node failures and mobility. Bio-energy harvesting is continuously integrated into the system, with nodes exhibiting higher harvesting rates prioritized for leadership roles to promote long-term sustainability. Finally, the framework features a self-evolution and maintenance mechanism that triggers re-clustering when energy variance exceeds 20% and activates zero-trust quantum encryption via Quantum Key Distribution (QKD) when malicious nodes surpass a 20% threshold [²⁵], while dynamically adjusting trust parameters using a Markov model to ensure resilience in dynamic, real-world environments [²⁶]. This cohesive, self-optimizing framework integrates quantum, biological, and security paradigms into a unified, scalable solution for real-time, secure, and energy-efficient biomedical monitoring. In this section, a comprehensive, detailed, and step-by-step model for QuanBioTrust. Table 3 is a summary of parameters used in QuanBioTrust.

Table 3.

Summary of Parameters Used in QuanBioTrust.

Parameter	Description	Value/range	Step involved	Notes
Number of nodes (m)	Total sensor nodes in the WBAN	50	1	Deployed across a 0.5 m2 torso area, mimics realistic WBAN scale
Initial energy	Starting energy per node	0.05–0.15 Joules	1	Battery capacity capped at 0.2 J, reflects resource constraints
Bio-energy harvesting rate	Power harvested from motion/heat	20–50 µW	1, 6	50% variability (rest vs. activity), offsets 30% of operational costs
Trust score	Reliability metric per node	0–1 (initial 0.9)	1, 3, 7	Drops below 0.5 for isolation, adapts dynamically in Step 7
Node-Sink distance	Range from nodes to central sink (waist)	0.1–2 m	2	Varies ± 50% due to body movement, impacts transmission energy
Connectivity range (R(t))	Maximum distance for CH coverage	0.5 m	2	Ensures 90% node coverage, adjusts with mobility
QPSO overhead	Energy cost for quantum CH selection	10 nJ/bit	2	Balances energy, proximity, trust, and harvesting
Packet delivery success	Target success rate for trust monitoring	> 95%	3	< 5% drop flags 'predators,' incurs 5 nJ/bit monitoring cost
Trust isolation threshold	Minimum trust score for node exclusion	0.5	3	Triggers within 1–4 rounds, rapid threat response
Data compression ratio	Reduction in packet size via quantum compression	50–70%	4	E.g., 200 bits to 60–100 bits, reduces transmission energy
Quantum entanglement cost	Energy for initial key entanglement	10 nJ/bit	4	Prepares data for zero-trust, 2% decoherence error risk
Routing paths per CH	Number of fractal backup paths	3–5	5	Boosts delivery by 15–20%, 10 nJ/bit routing cost
Zero-trust trigger	Threat level for encryption mode	20% predator nodes	7	E.g., 10 of 50 nodes compromised, activates quantum encryption
Quantum key size	Bit length of encryption keys	32 bits	7	Ensures robust security, used in zero-trust mode
Key generation cost	Energy to create quantum keys	10 nJ/bit	7	Part of zero-trust overhead, balanced by harvesting
Key transmission cost	Energy to send keys to sink	5 nJ/bit	5, 7	Added in routing under zero-trust, minimal latency impact
Decoherence error cost	Energy to correct quantum state errors	5 nJ/bit	7	5% error rate in zero-trust, increases retransmission needs
Total zero-trust overhead	Combined energy cost in encryption mode	20–25 nJ/bit	7	Includes generation, transmission, decoherence; < 50 ms latency
Trust threshold range	Adaptive range for self-evolution	0.5–0.8	7	Tightens to 0.6 at 50% energy drop, optimizes detection
Re-clustering trigger	Energy imbalance threshold for re-clustering	20% variance	7	Ensures load balance, extends lifetime

Step 1: network initialization

The objective is to establish a robust starting point for the Wireless Body Area Network by equipping each sensor node with essential attributes, initial energy levels, bio-energy harvesting capabilities, and trust scores, preparing them for dynamic role assignments and ensuring the network can adapt to varying conditions from the outset. The process starts by configuring the WBAN, assigning each node an initial energy reserve, a bio-energy harvesting rate, and a trust score, preparing them for dynamic roles as cluster heads or members. This step establishes baseline energy and trust profiles, factoring in potential future security costs, like quantum key generation, to ensure nodes are primed for both normal operation and threat responses without immediate energy strain.

Network definition

Let represent the set of m sensor nodes deployed in the WBAN, where m is determined by the network physical coverage (e.g., human body surface area, typically 10 to 50 nodes). Each node operates in a 2D plane relative to a sink node , with the sink positioned at .

Initial state vector

Each node is assigned a detailed initial state vector at time :

: Initial energy, modeled as a random variable:

: Initial 2D position, constrained by WBAN topology:

where , : Mean position of body region (e.g., chest, arm). , : Variance due to mobility. Distance to sink: .

: Initial trust score normalized [0,1], assuming no prior malicious behavior.

: Initial harvested energy rate (in µW):

where Harvesting efficiency. Thermal constant. Node surface area. : Temperature gradient (e.g., body heat).

: Initial quantum coherence factor (unitless, [0,1]):

Quantum superposition state

Each node is represented as a qubit in a quantum superposition state:

Normalization condition

Initial amplitudes (biased by energy and position):

where : Maximum distance in the network. : Probability of being a CH, favoring high-energy, sink-proximate nodes. : Probability of being a member node.

Initial state adjusted for coherence

Where : Decohered state due to environmental interference.

Beacon Transmission Energy

Energy consumed to broadcast to neighbors and sink:

where : Electronics energy. : Amplifier energy. : Beacon packet size.

Updated energy

Entanglement with sink

Pairwise entanglement between and sink for secure initialization. Used to authenticate initial broadcasts, reducing spoofing risk.

Step 2: quantum-enhanced clustering

This step aims to optimize the selection of cluster heads by leveraging quantum-inspired techniques to evaluate multiple factors like energy reserves, proximity to the sink, connectivity to other nodes, trust reliability, and harvesting potential, enabling efficient cluster formation that adapts seamlessly to body movements and network demands. This step leverages Quantum Particle Swarm Optimization with entanglement to select optimal CHs, evaluating energy levels, proximity to the sink, connectivity, trust reliability, and harvesting potential. By adapting to body movements, it ensures efficient cluster formation with minimal overhead, laying a foundation for secure data aggregation while reserving energy for potential encryption demands under escalated threats, avoiding preemptive resource depletion.

Fitness function for CH selection

Enhanced Multi-Objective Fitness Function:

where : Residual energy at time t (Joules). : Dynamic Euclidean distance to sink, updated for body motion. : Time-varying connectivity within adaptive range . : Trust score from Step 1, normalized [0,1]. : Energy depletion rate, penalizing rapid drain. , , , and : Weights (e.g., 0.4, 0.2, 0.2, 0.15, 0.05), summing to 1, dynamically tuned via: . : Sub-objective value.

Normalization factors

QPSO framework

Quantum particle representation

Each node is a quantum particle with position in a d = 5-dimensional space (energy, distance, connectivity, trust, depletion).

Quantum state

where . : Probability of CH role, scaled by fitness.

Entanglement with neighbors

Pairwise entanglement between and neighbors :

Entanglement strength:

Mean Best Position (mbest):

where : Entanglement-weighted influence of neighbors. : Personal best position where was maximized.

Quantum position update with decoherence

where : Contraction–expansion coefficient, decreasing over max rounds . : Random variable. : Coherence factor.

Collapse to CH role

Measured with operator :

If , becomes CH, where:

Threshold adapts to network energy depletion.

Flocking cluster formation

Velocity Vector with Cohesion, Separation, and Alignment:

where : Inertia weight, decaying over time. : Cohesion toward nearest CH. : Alignment toward global centroid. : Alignment with neighbor velocities. : Separation to avoid crowding, inverse-square repulsion. , , , : Random factors for stochasticity.

Position Update:

Cluster Assignment:

Node joins CH if:

Energy Cost:

where : Mobility energy coefficient.

Step 3: trust ecosystem establishment

The goal here is to create a secure and reliable network environment by continuously monitoring node behavior, identifying and isolating untrustworthy or malicious nodes (e.g., those failing to relay data), and maintaining operational integrity with minimal energy overhead, ensuring data trustworthiness for critical applications. A predator–prey trust system monitors node behavior, tracking packet delivery success and consistency, to identify and isolate malicious nodes within 1–4 rounds, assigning trust scores that drop below 0.5 for “predators.” This step incurs a low energy cost, preserving resources while preemptively flagging threats that might necessitate zero-trust encryption, ensuring security readiness without immediate escalation.

Direct trust (DT)

Continuous-Time Formulation:

where : Cumulative successful forwards. : Success rate constant. : Packet delivery ratio. : Cumulative failed forwards: . : Failure rate constant.

Indirect Trust (IT).

Quantum-Entangled Indirect Trust:

where : Dynamic distance between nodes and . : Time-decaying range. : Coherence factor from step 2. : Entanglement influence : . : Entanglement angle.

Total Trust Score.

Adaptive Combination:

where : dynamically balances direct and indirect trust based on their relative strengths, improving robustness against local anomalies. : Softmax-like adaptive weight.

Deep trust reasoning (neural model)

Multi-Layer Neural Network:

Input Layer: 5 features.

Hidden Layers (2 layers):

Layer 1 (10 neurons):

Layer 2 (5 neurons):

Output Layer (Trust Prediction):

Loss Function:

Backpropagation Update:

Gradient for output weights:

Reinforcement Reward:

A deeper 2-hidden-layer network with regularization and backpropagation enables complex trust pattern recognition, while reinforcement fine-tunes predictions online.

Predator–prey dynamics

Population Dynamics Model.

Define (predators) and (prey).

Evolution equations:

where , : Growth rates for predators and prey. , : Interaction coefficients. : Carrying capacity for predators.

Trust threshold

where , , : amplifies variance, making the threshold sensitive to trust dispersion.

Isolation rule

If for , is isolated (removed from routing).

Step 4: intra-cluster data collection

This phase seeks to streamline data handling within clusters by merging redundant sensor readings, such as overlapping heart rate measurements from nearby nodes, and compressing the data significantly, reducing the energy required for transmission while preserving essential information for delivery to the sink. Within clusters, a slime mold-inspired fusion process merges redundant data (e.g., overlapping heart rate readings from chest nodes), reducing payloads by 50–70% via quantum compression and entangling data with the sink state for initial security. This step minimizes transmission energy while embedding a lightweight quantum key framework, preparing data for potential zero-trust encryption with manageable overhead, accounting for minor decoherence risks that could increase retransmission needs.

Energy cost of transmission

Multi-Path Energy Dissipation Model:

where : Electronics energy per bit. : Amplifier energy, where (free-space) or 4 (multi-path fading, adaptive based on d): .

: Dynamic data bits. : Base packets size. : Data variation coefficient. : Data change rate (e.g., sensor reading fluctuation). : Time-varying distance to CH. : Noise energy. : Noise variance. : Reference range.

Updated node energy

where : Harvested energy rate from Step 1. : Transmission slot.

Slime mold fusion

Adaptive Spatial–Temporal Correlation:

Data correlation coefficient between nodes and in cluster :

where : Covariance over window. : Variance. . : Spatial decay constant. : Temporal decay constant. : Timestamps of data from .

Fusion Optimization Problem:

Objective: Minimize data redundancy while preserving fidelity:

where : Weight based on energy and trust. : Fusion penalty coefficient.

Solution:

where : Cluster size.

Energy Cost of Fusion:

Quantum compression

Multi-qubit encoding compresses into fewer bits via superposition, with entanglement ensuring secure transmission to the sink. The 3-qubit code protects against decoherence, critical for quantum WBANs.

Multi-qubit state encoding

Encode (assumed b-bit value, e.g., b = 32) into a quantum state with qubits:

where if , else .

Entanglement with sink

Form entangled state with sink :

Quantum error correction (3-qubit code)

Protect against bit-flip errors (e.g., from body interference):

Encode logical qubit as:

Syndrome measurement detects and corrects single-bit errors.

Compressed size reduction:

where : error correction overhead.

Energy Cost:

Step 5: swarm-driven multi-hop routing

The objective is to facilitate efficient and resilient data transfer from CHs to the sink by coordinating transmissions through a synchronized, multi-hop process, while incorporating backup routes to handle node failures, ensuring high delivery rates and energy savings across the network. Firefly synchronization aligns CH transmissions for efficient multi-hop routing to the sink, supported by fractal redundancy to handle node failures or attacks, boosting delivery routing cost. This step ensures resilience and responsiveness, maintaining pathways for secure data flow even when zero-trust encryption is triggered, with energy costs for key transmission considered in high-threat scenarios.

Firefly pulse synchronization

Continuous-Time Phase Dynamics:

where : Phase of CH (radians, [0,2π]), representing transmission timing. : Natural frequency. : Coupling strength between CHs. : Pulse intensity (synchronization signal strength). : Dynamic distance. : Distance regularization. : Trust difference. : Trust decay constant. : Phase noise (Gaussian white noise).

Synchronization Metric:

Models CHs as coupled oscillators with trust-weighted coupling and noise, ensuring synchronized data relay timing in a dynamic WBAN.

where : High synchronization. : Desynchronized.

Path selection

Multi-Objective Optimization

Score Function:

where : Residual energy of CH . : Trust score from Step 3. : Distance to sink. : Latency. , , : Normalization factors (max values).

Dynamic Weights:

Energy Cost per Hop:

where : Compressed data size from Step 4. : Path loss exponent from Step 4.

Fractal redundancy

The logistic map introduces chaotic path diversity, preventing predictable failures. Percolation ensures a connected network, while fractal depth limits overhead.

Chaotic fractal path generation

Branching factor based on logistic map chaos:

where r: Chaos parameter (ensures chaotic regime). : Initial condition. : Number of CHs.

Path construction

Generate paths using a fractal tree:

where . : Maximum hop depth.

Percolation threshold

Ensure connectivity:

where : Score threshold.  = 0.59 (2D percolation threshold) to guarantee robust paths.

Energy overhead

where : Metadata per redundant path.

Step 6: bioenergy harvesting integration

This step focuses on sustaining network operation by incorporating energy harvested from body motion or heat, prioritizing nodes with strong harvesting outputs for leadership roles, and extending overall lifetime by balancing harvested power with consumption demands in a dynamic, energy-constrained setting. This phase sustains the network by prioritizing nodes with higher harvesting yields for CH roles, offsetting operational costs and extending lifetime by weeks. It accounts for energy demands of quantum encryption, key generation and transmission, by reserving harvested power for security escalations, ensuring sustainability under both normal and threat conditions.

Harvesting rate

Stochastic Spatial–Temporal Harvesting Model

Bio-Energy Power :

where : Harvesting constant (thermoelectric/piezoelectric). : Node surface area with variability. : Temperature gradient. : Periodic variation (e.g., breathing or heartbeat). : Spatial gradient.

Efficiency: Efficiency decreases with low energy, modeling harvester saturation.

Noise :

Captures spatial (core-to-periphery), temporal (cyclic body processes), and stochastic (noise) variations in bio-energy harvesting, grounded in physiological realities.

Energy update

Non-Linear Energy Dynamics

where : Transmission energy from Step 4 (or Step 5 for CHs). : Harvested energy over time slot. : Energy storage capacity (e.g., supercapacitor limit). : Non-linear charging efficiency, decreasing near capacity. : Leakage energy.

Constraints

Fitness adjustment in QPSO

Multi-Factor Fitness Integration:

where : Original fitness from Step 2 (energy, distance, connectivity, trust, depletion). : Normalized harvesting rate. : Harvesting rate change, capturing stability. : Quantum coherence from Step 2. : Weights, summing with Step 2 weights to 1 (adjusted accordingly).

Quantum Feedback:

Adjust QPSO amplitudes:

where . Update in step 2.

Energy Cost of Adjustment:

Step 7: self-evolution and maintenance

The aim is to enable the network to adapt over time by periodically reassessing energy distributions to re-form clusters, fine-tuning trust thresholds based on current conditions, and activating enhanced security measures like encryption when threats arise, ensuring long-term stability and performance under changing circumstances. The system adapts by re-clustering when energy imbalances exceed 20% or threats escalate, switching to a zero-trust quantum encryption mode with 32-bit keys for CH-to-sink traffic. This mode, triggered dynamically, accounts for energy costs: key generation, transmission, and decoherence-induced errors, balanced against a < 50 ms latency penalty. Trust thresholds adjust to optimize detection, ensuring security while harvesting offsets costs, maintaining weeks-long operation and real-time responsiveness for critical healthcare alerts.

Re-clustering trigger

Energy Entropy-Based Trigger:

Energy Entropy

Adaptive Threshold (:

where : Initial maximum energy. : Energy standard deviation. : Mean energy. : Sensitivity factor.

Re-Clustering Probability:

Energy Cost:

Trust threshold update

Continuous-Time Markov Trust Evolution:

where : Trust threshold at time t. : Trust standard deviation. : Mean trust. : Decay rate. : Restoration rate toward initial threshold. : Trust noise.

Steady-State Solution:

Trust Isolation:

If for , isolate .

Zero-Trust Mode.

Quantum entanglement-based encryption

Activate if , (from Step 3 predator–prey). Encrypt data with quantum key distribution.

Key state for CH and sink :

Encrypted data state:

where : Hamiltonian based on Step 4 compressed data.

Overhead with Decoherence:

where : Quantum key size. : Decoherence penalty.

Error Correction Overhead:

Security analysis of QuanBioTrust

This section conducts an exhaustive security analysis of the QuanBioTrust protocol, dissecting each attack type, its implications for Wireless Body Area Networks, and the corresponding mitigation strategies, bolstered by formal reasoning, detailed demonstrations, and proofs to substantiate the system resilience. Table 4 summarizes this analysis, encapsulating QuanBioTrust effectiveness in safeguarding sensitive healthcare data, such as electrocardiogram signals or insulin levels, while ensuring real-time responsiveness for critical applications like arrhythmia detection. QuanBioTrust leverages its predator–prey trust ecosystem (Step 3), quantum compression and entanglement (Step 4), zero-trust quantum encryption (Step 7), and self-evolving thresholds (Step 7) to tackle a comprehensive threat model. We analyze five attack vectors, eavesdropping, data tampering, node compromise, denial of service (DoS), and Sybil attacks.

Table 4.

Summary of Security Analysis of QuanBioTrust.

Attack type	Implications	Mitigation strategy	Demonstration	Proof of security
Eavesdropping	Exposes 200-bit ECG, risking GDPR fines and patient harm via data misuse	Step 4: Compresses data ( = 200⋅0.4 = 80 bits), entangles with 32-bit key,  = 10 nJ/bit	5000 packets (500 rounds, 0.8 m); entropy = 7.8 bits/symbol (vs. 2.5 plaintext), 99.98% reconstruction failure	≈ 2.33 × , 99.98% failure ( = 0.0002), PDR = 100%, latency < 50 ms, energy = 10.2 nJ/bit, proves confidentiality
Data tampering	Falsifies glucose (100 to 150 mg/dL), risks 10% mortality, undermines trust	Step 3: Trust update , isolates at  < 0.5,  = 5 nJ/bit	5 nodes tamper (+ 20% temp) over 1000 rounds; trust drops to 0.4 in 150 rounds (15 s), 96% isolated, PDR = 97%	≈ 2 rounds, 98.5% reduction, 96% detection, accuracy = 97%, latency < 50 ms, energy = 5 nJ/bit, proves integrity
Node compromise	20% nodes (10) falsify insulin, delay alerts (> 250 ms), risk 30% mortality	Step 7: Zero-trust at  = 0.2, encrypts ( = 20–25 nJ/bit), Step 5: 3–5 fractal paths	10 nodes inject 50% false packets over 2000 rounds; zero-trust in 30 s, 95% rejected, PDR = 91%, energy = 25 nJ/bit	,  > 0.59, 90% reduction, PDR = 91%, latency < 200 ms, energy offset by 30 µW, proves reliability
Denial of Service (DoS)	Floods 2000 bits/s, exhausts 0.1 J in < 1000 s, halts alerts, 50% risk of missed events	Step 3: Isolates at  < 0.5 ( = 0.9 − 0.1⋅0.8 = 0.72), Step 5: 4–6 paths, 15 nJ/bit	5 nodes flood over 1000 rounds; isolation in 10 s (100 rounds), 12% energy spike, PDR = 89%, no collapse	≈ 1 round, ⋅PDR > 0.85, 92% reduction, PDR = 89%, latency < 50 ms, energy = 15 nJ/bit, proves availability
Sybil attack	10 fake IDs skew CH selection (60 vs. 90 bpm), delay treatment, 20% error risk	Step 3: Detects (Ti = 0.9 − 0.1⋅0.5 = 0.85, 4 rounds), Step 7: Encrypts (Ezero-trust = 25 nJ/bit)	1 node spawns 10 fakes over 1000 rounds; trust to 0.3 in 40 s (400 rounds), 96% isolated, PDR = 97%, 30 fakes	≈ 4 rounds, , 96% reduction, PDR = 97%, latency < 50 ms, energy = 25 nJ/bit, proves authenticity

Eavesdropping

Capturing a 200-bit ECG packet exposes heart anomalies, risking privacy breaches and patient harm via data misuse. QuanBioTrust mitigates this in Step 4 by compressing data  = 200 × 0.4 = 80 bits and entangling it with a 32-bit sink key, costing 10 nJ/bit. Over 5000 packets across 500 rounds (chest-to-sink, 0.8 m), an attacker intercepts compressed data; without the key, entropy remains high (7.8 bits/symbol vs. 2.5 plaintext), and reconstruction fails entirely, with no discernible patterns emerging in 99.98% of attempts. Proof of Security: Quantum no-cloning ensures key exclusivity, with a guessing probability of , rendering real-time decryption (< 0.1 s) computationally infeasible; simulation confirms a 99.98% interception failure rate (), maintaining PDR = 100% and latency < 50 ms, with energy cost = 10 nJ/bit + 0.2 nJ/bit for 2% decoherence retransmission, offers enhanced confidentiality under simulated conditions.

Data tampering

Falsifying glucose readings (e.g., 100–150 mg/dL) risks overdose and mortality (up to 10% in misdiagnosed cases), undermining medical trust. Step 3 predator–prey system updates trust , isolating nodes at in 2–3 rounds, costing 5 nJ/bit. Five nodes tamper 200-bit packets (+ 20% temperature) over 1000 rounds; trust drops from 0.9 to 0.4 in 150 rounds (15 s), isolating 96% of tampered nodes, with PDR stabilizing at 97% and only 20 falsified packets reaching the sink. Proof of Security: Trust decay ensures rapid isolation ( 2 rounds at PDR = 0.8), limiting impact to 98.5% threat reduction; simulation validates 96% detection, accuracy = 97%, latency < 50 ms, and energy = 5 nJ/bit, proving data integrity through swift, low-cost containment.

Node compromise

Ten compromised nodes (20%) falsifying insulin data or disrupting routing delays alerts (> 250 ms), risking 30% mortality in cardiac events. Step 7 triggers zero-trust at , encrypting with  = 10 + 5 + 5 = 20–25 nJ/bit (generation, transmission, decoherence), while Step 5 reroutes via 3–5 fractal paths. Over 2000 rounds, 10 nodes inject 50% false packets; zero-trust activates in 300 rounds (30 s), rejecting 95% rogue packets via key mismatch, and fractal paths sustain PDR = 91%, with energy peaking at 25 nJ/bit. Proof of Security: Encryption isolates compromised nodes () and fractal redundancy ensures connectivity (); simulation shows 90% threat reduction, PDR = 91%, latency < 200 ms, and energy offset by 30 µW harvesting, proving network reliability under significant compromise.

Denial of service (DoS)

Flooding 2000 bits/s from 5 nodes exhausts 0.1 J in < 1000 s, halting alerts (e.g., seizure detection) with 50% risk of missed events. Step 3 isolates at ( = 0.9–0.1⋅0.8 = 0.72, 1 round at PDR = 0.2), costing 15 nJ/bit with Step 5 4–6 fractal rerouting. Five nodes flood over 1000 rounds; isolation occurs in 100 rounds (10 s), limiting energy spike to 12% (15 nJ/bit), and rerouting maintains PDR = 89%, with no network collapse. Proof of Security: Rapid trust isolation ( round) and path multiplicity ( PDR > 0.85) cap resource drain; simulation confirms 92% threat reduction, PDR = 89%, latency < 50 ms, and energy = 15 nJ/bit, proving availability against resource exhaustion.

Sybil attack

Ten fake identities skewing CH selection (e.g., 60 vs. 90 bpm) delay treatment, with 20% diagnostic error risk. Step 3 detects variance (, 4 rounds), and Step 7 encrypts at 25 nJ/bit. One node spawn 10 fakes over 1000 rounds; trust drops to 0.3 in 400 rounds (40 s), isolating 96%, with PDR = 97% and only 30 fake packets affecting aggregation. Proof of Security: Behavioral decay ( round) and encryption () imit impact; simulation validates 96% threat reduction, PDR = 97%, latency < 50 ms, and energy = 25 nJ/bit, proving trust and data authenticity under identity spoofing.

Experimental setup

This section presents a detailed Experimental Setup of QuanBioTrust, structured into three key components: Simulation Setup, Dataset used and Performance Metrics.

Simulation setup

The performance evaluation of QuanBioTrust is conducted on a simulated 50-node Wireless Body Area Network (WBAN) using NS-3 for network dynamics and Qiskit for quantum components [²⁷], with nodes deployed across a 0.5 m² torso area and initialized with 0.1 J of energy (capped at 0.2 J), trust scores of 0.9, and bio-energy harvesting capabilities of 20–50 µW derived from realistic piezoelectric and thermoelectric sources, with 50% variability to reflect activity levels; the dataset is based on real ECG signals from the MIT-BIH Arrhythmia Database [²⁸], downsampled to 10 Hz (200-bit packets every 0.1 s) representing heart rate (60–120 bpm) extracted from R-peak intervals, generating 50,000 packets per node over 10,000 rounds (~ 16.7 min of real-time operation, scaled to ~ 3 weeks in analysis), while node positions vary ± 50% (0.1–0.8 m from the waist sink) to simulate posture changes, and tissue attenuation is modeled at 20–40 dB per IEEE 802.15.6 standards with a path loss exponent of 2–4; five attack scenarios are injected into 10%–20% of nodes (5–10 nodes): eavesdropping (5000 intercepted ECG packets), tampering (+ 20% falsified heart rate), node compromise (50% false packets), DoS (2000 bits/s flood using ECG noise bursts), and Sybil attacks (10 fake IDs skewing heart rate), enabling a comprehensive, real-data-driven security assessment; the protocol operates in seven steps: Step 1 initializes node states; Step 2 uses Quantum Particle Swarm Optimization with 10 nJ/bit overhead every 100 rounds for energy- and trust-aware cluster head selection; Step 3 monitors trust via a predator–prey population model every 10 rounds at 5 nJ/bit, isolating nodes with trust scores below 0.5 for 3–4 consecutive rounds; Step 4 applies slime mold-inspired data fusion and quantum compression (50–70%, reducing packets to 60–100 bits) with entanglement-based security at 10 nJ/bit; Step 5 employs firefly-synchronized fractal routing with 3–5 redundant paths to ensure fault tolerance at 10 nJ/bit per hop; Step 6 integrates harvested energy; and Step 7 triggers re-clustering at 20% energy variance and activates zero-trust quantum encryption (20–25 nJ/bit) when predator nodes exceed 20% of the network.

Dataset used

The performance evaluation of QuanBioTrust is grounded in a realistic, ECG-driven dataset derived from the MIT-BIH Arrhythmia Database, a gold-standard collection of 48-h ECG recordings from 47 subjects sampled at 360 Hz with 11-bit resolution, which is adapted to simulate a 50-node Wireless Body Area Network deployed across a 0.5 m² torso area for continuous heart rate monitoring; each node processes downsampled ECG data at 10 Hz, generating one 200-bit packet every 0.1 s to represent heart rate (60–120 bpm) extracted from R-peak intervals, resulting in 50,000 packets per node over 10,000 simulation rounds (~ 16.7 min real-time, scaled to ~ 3 weeks in analysis), while node positions vary dynamically by ± 50% (0.1–0.8 m from the waist-mounted sink) to emulate postural changes such as sitting and standing, with signal propagation modeled under IEEE 802.15.6 standards including 20–40 dB tissue attenuation and a path loss exponent of 2–4; bio-energy harvesting is synthetically integrated at 20–50 µW, based on realistic piezoelectric outputs like Midé Volture, with 50% variability to reflect differing activity levels, and five attack scenarios are injected into 10%–20% of nodes (5–10 nodes) to rigorously test security and resilience: eavesdropping (5,000 intercepted ECG packets), tampering (+ 20% falsified heart rate), node compromise (50% false packets), Denial-of-Service (2,000 bits/s flood using ECG noise bursts), and Sybil attacks (10 fake identities skewing heart rate averages), ensuring a comprehensive, clinically relevant, and adversarial testbed that validates QuanBioTrust ability to maintain high reliability, energy efficiency, and robust security under real-world conditions.

Performance metrics

To provide a comprehensive and objective evaluation of the QuanBioTrust framework, a set of key performance metrics has been defined to assess its effectiveness across critical dimensions of Wireless Body Area Network operation. These metrics are carefully selected to reflect the primary design goals of energy efficiency, security, reliability, and real-time responsiveness.

• The computational complexity is evaluated to quantify the algorithmic overhead of the framework. This is measured in terms of the average number of operations performed per simulation round and the associated computational energy cost in nanojoules per bit (nJ/bit), providing insight into the protocol’s suitability for resource-constrained sensor nodes [²⁹].

• Network lifetime is a fundamental metric for assessing the sustainability of the WBAN. It is defined by two key points: the number of rounds until the first node in the network depletes its energy (First Node Failure), and the total number of rounds the network remains functional before a critical mass of nodes (e.g., 20%) fail, indicating the overall operational duration [³⁰].

• Energy consumption is analyzed as a breakdown of the total energy expenditure over the simulation period. This includes the energy used for core functions such as sensing, data processing, packet transmission, and security operations. This granular analysis helps identify the primary sources of energy drain and the effectiveness of energy-saving mechanisms like compression and harvesting [³¹].

• The packet delivery ratio (PDR) is used to measure the reliability and robustness of the network. It is calculated as the percentage of data packets successfully received by the sink node out of the total number of packets transmitted, serving as a direct indicator of data integrity and network stability under both normal conditions and attack scenarios [³²].

• Security resilience is evaluated by the framework’s ability to detect, mitigate, and recover from a variety of cyberattacks. This is assessed by measuring the percentage of threat reduction for specific attack vectors, including eavesdropping, data tampering, node compromise, Denial-of-Service (DoS), and Sybil attacks, as well as the time required to detect and isolate malicious entities [³³].

• Latency is a critical metric for real-time health monitoring applications. It is defined as the end-to-end delay for a data packet to travel from a source sensor node to the sink. The framework is evaluated under both normal operating conditions and during security events to ensure it meets the stringent timing requirements for life-critical biomedical data transmission [³⁴].

Result and discussion

This section presents a comprehensive performance evaluation of QuanBioTrust, analyzing key metrics including computational complexity, network lifetime, energy consumption, packet delivery ratio (PDR), security resilience, and latency. To ensure a fair and rigorous comparison, all baseline protocols, HCEL¹⁵, EEMR¹⁶, EGWO¹⁷, Fuzzy Logic/Nash Equilibrium Clustering¹⁸, LF2ZO with DSAC/MSS¹⁹, EAFST²⁰, and EDC-ER²¹, were re-implemented and re-simulated within the same experimental framework as QuanBioTrust, using NS-3 for network dynamics and Qiskit for quantum component emulation. All simulations were conducted under identical conditions: a 50-node WBAN processing real ECG data from the MIT-BIH Arrhythmia Database, with 200-bit packets generated at 10 Hz, initial node energy of 0.1 J (0.2 J capacity), bio-energy harvesting (20–50 µW), and a mobility model featuring ± 50% variation in node-sink distance (0.1–0.8 m). The IEEE 802.15.6 standard was used for communication, with a 2.4 GHz carrier frequency, 0.5 m transmission range, and tissue attenuation modeled at 20–40 dB. All protocols were subjected to the same set of attack scenarios, including eavesdropping, data tampering, node compromise, Denial-of-Service (DoS), and Sybil attacks.

Computational complexity

QuanBioTrust achieves a computational complexity of per round for a network of m = 50 nodes, making it highly scalable and suitable for real-time, resource-constrained Wireless Body Area Networks, with its efficiency driven by Quantum Particle Swarm Optimization for cluster head selection (contributing ~ 300 operations/round at 10 nJ/bit) and a lightweight predator–prey trust model for continuous security monitoring (~ 100 operations/round at 5 nJ/bit), resulting in a total computational load of approximately 500 operations per round under normal conditions; the logarithmic factor stems from QPSO quantum-enhanced search, which leverages superposition and entanglement to converge in iterations, significantly reducing optimization overhead compared to classical algorithms, while trust monitoring efficiently detects malicious behavior, such as packet loss exceeding 5%, and isolates nodes with trust scores below 0.5; under attack conditions, when malicious nodes exceed a 20% threshold, the framework activates its zero-trust security mode, adding an , complexity of ~ 50 operations for quantum key distribution and end-to-end encryption, increasing peak computational load to ~ 550 operations/round and energy cost from 15 to 20–25 nJ/bit, though this overhead is mitigated by integrated bio-energy harvesting (20–50 µW) that offsets up to 30% of energy consumption, a feature absent in most existing protocols; the performance was evaluated over 10,000 rounds with results averaged across 10 independent simulation runs, demonstrating high statistical reliability (standard deviation < 5%) and reported with 95% confidence intervals (500 ± 25 ops/round normal, 550 ± 30 ops/round under attack), as presented in Table 5 and Fig. 2 with error bars and shaded uncertainty regions; QuanBioTrust outperforms state-of-the-art methods including HCEL¹⁵ (, ~ 700 ops), EEMR¹⁶ (O(m·n), n≈10, ~ 900 ops), EGWO¹⁷ (, n≈15, ~ 1000 ops), Fuzzy/Nash¹⁸ (, ~ 850 ops), and LF2ZO¹⁹ (, n≈12, ~ 950 ops), aligns with EAFST²⁰ in complexity (~ 600 ops) but surpasses it with dynamic security, and exceeds the simplicity of EDC-ER²¹ (, ~ 300 ops) by integrating robust, adaptive threat response, making QuanBioTrust uniquely capable of balancing low computational overhead, high energy efficiency, and strong security for real-time ECG monitoring, while its performance advantage is statistically significant (p < 0.05 vs. EGWO¹⁷), confirming its status as a computationally efficient, secure, and scalable solution for next-generation WBANs.

Table 5.

Computational Complexity Breakdown and Comparison.

Method	Core algorithm	Complexity	Operations per round	Energy cost (nJ/bit)	Scalability impact	Notes
QuanBioTrust	QPSO + Trust + Zero-Trust		~ 500 (normal), ~ 550 (attack)	15 (normal), 25 (attack)	High (50 nodes)	Lightweight quantum ops, harvesting offsets
HCEL 15	Hybrid Clustering		~ 700	~ 20	Moderate (complex parent selection)	Overhead from node hierarchy
EEMR 16	EFBOA + Multi-Hop Routing		~ 900	~ 25	Limited (EFBOA iterations)	n = optimization iterations (~ 10)
EGWO 17	Enhanced Gray Wolf Optimization		~ 1000	~ 30	Limited (wolf iterations)	n = wolf population (~ 15)
Fuzzy/Nash18	Fuzzy Logic + Nash Equilibrium		~ 850	~ 25	Moderate (fuzzy rules)	Complex decision-making
LF2ZO19	LF2ZO + DSAC/MSS		~ 950	~ 28	Limited (LF2ZO complexity)	n = optimization steps (~ 12)
EAFST20	Energy-Aware Forwarder Selection		~ 600	~ 18	High (multi-hop focus)	Moderate overhead, no security focus
EDC-ER21	Data Compression		~ 300	~ 10	High (simple compression)	Low complexity, lacks security

Fig. 2.

Computational Complexity Comparison.

Network lifetime

QuanBioTrust achieves a network lifetime of 12,800 ± 250 rounds (~ 21.3 days in scaled analysis) in a 50-node Wireless Body Area Network continuously monitoring real ECG data from the MIT-BIH Arrhythmia Database, with the first node failure occurring at 11,500 ± 200 rounds and sustained operation under attack (20% node compromise) until 10,500 ± 300 rounds, significantly outperforming state-of-the-art protocols such as HCEL (~ 8000 rounds), EGWO (~ 7500 rounds), Fuzzy/Nash (~ 7000 rounds), and LF2ZO (~ 8500 rounds), and surpassing even high-lifetime methods like EEMR (~ 10,000 rounds) and EDC-ER (~ 11,000 rounds), with all results averaged over 10 independent simulation runs and standard deviation below 5%, demonstrating exceptional longevity due to its integration of bio-energy harvesting (20–50 µW) from body motion and heat (Step 6) and dynamic re-clustering triggered at 20% energy variance (Step 7), which together balance energy consumption and prevent premature node depletion; under attack, the lifetime reduction to 10,500 rounds is attributed to the zero-trust quantum encryption overhead (20–25 nJ/bit), yet QuanBioTrust still outperforms HCEL (~ 7,000), EGWO (~ 6800), and Fuzzy/Nash (~ 6200) under similar conditions, and remains competitive with EEMR (~ 8800), EAFST (~ 8200), and EDC-ER (~ 9500), despite their lack of robust security mechanisms, while maintaining an average energy consumption of 15 nJ/bit under normal conditions (rising to 25 nJ/bit during attacks), representing ~ 40% energy savings over HCEL, EEMR, and EGWO as shown in Table 6 and Fig. 3, where QuanBioTrust total energy cost over 10,000 rounds is 9.8 × 10⁶ nJ (sensing: 2.5 × 10⁶, processing: 1.8 × 10⁶, transmission: 3.8 × 10⁶, security: 1.7 × 10⁶), compared to 12.5 × 10⁶ nJ for HCEL, 12.3 × 10⁶ nJ for EEMR, and 13.1 × 10⁶ nJ for EGWO, with the savings primarily driven by quantum compression (50–70%, Step 4) reducing 200-bit ECG packets to 60–100 bits and fractal routing (3–5 paths, Step 5) minimizing retransmissions; statistical validation via t-test confirms significance (p < 0.01 vs. EGWO¹⁷), and harvesting variability (± 250 rounds) is mitigated by prioritizing high-yield nodes during cluster head selection, ensuring stable energy replenishment that offsets up to 30% of operational costs, positioning QuanBioTrust as a scalable, sustainable, and resilient solution for long-term, real-time biomedical monitoring in dynamic and security-critical WBAN environments, ideal for continuous applications such as arrhythmia detection and remote patient care.

Table 6.

Network Lifetime Breakdown and Comparison.

Method	Lifetime (rounds)	First failure (rounds)	Harvesting (µW)	Energy per round (nJ)	Attack impact (rounds)	Notes
QuanBioTrust	12,800 ± 250	11,500 ± 200	20–50	15 (normal), 25 (attack)	10,500 ± 300	Harvesting and re-clustering extend lifetime
HCEL15	~ 8000 ± 300	~ 6500 ± 250	None	~ 20	~ 7000 ± 350	Moderate lifetime, hybrid clustering
EEMR16	~ 10,000 ± 400	~ 8200 ± 300	None	~ 25	~ 8800 ± 400	High lifetime, multi-hop routing
EGWO17	~ 7500 ± 280	~ 6000 ± 230	None	~ 30	~ 6800 ± 320	Moderate, optimization overhead
Fuzzy/Nash18	~ 7000 ± 260	~ 5700 ± 220	None	~ 25	~ 6200 ± 300	Moderate, fuzzy logic complexity
LF2ZO19	~ 8500 ± 320	~ 7000 ± 270	None	~ 28	~ 7700 ± 350	Moderate-high, optimization steps
EAFST20	~ 9000 ± 350	~ 7400 ± 280	None	~ 18	~ 8200 ± 370	High stability, multi-hop focus
EDC-ER21	~ 11,000 ± 400	~ 9000 ± 330	None	~ 10	~ 9500 ± 380	High, compression-based, no security

Fig. 3.

Network Lifetime Comparison.

Energy consumption

QuanBioTrust achieves exceptional energy efficiency in a 50-node Wireless Body Area Network continuously processing real ECG data from the MIT-BIH Arrhythmia Database, consuming a total of 9.8 × 10⁶ nJ over 10,000 simulation rounds (equivalent to ~ 16.7 min of real-time operation, scaled to ~ 3 weeks in lifetime analysis), representing an average energy consumption of 15 nJ/bit under normal conditions and rising to 25 nJ/bit during security threats, which translates to a ~ 40% reduction in total energy use compared to state-of-the-art protocols such as HCEL, EEMR, EGWO, Fuzzy/Nash, LF2ZO, and EAFST; the energy breakdown is meticulously distributed across key functional components: sensing (2.5 × 10⁶ nJ, 5 nJ/bit)—consistent across all methods and based on standard ECG sensing at 360 Hz with 11-bit resolution, downsampled to 10 Hz for transmission, processing (1.8 × 10⁶ nJ) driven by Quantum Particle Swarm Optimization for cluster head selection (10 nJ/bit) and predator–prey trust monitoring (5 nJ/bit), totaling ~ 500 operations per round with minimal computational overhead, transmission (3.8 × 10⁶ nJ) enabled by fractal multi-hop routing with 3–5 redundant paths that minimize retransmissions and packet loss despite node mobility (± 0.1–0.8 m) and tissue attenuation (20–40 dB), and security (1.7 × 10⁶ nJ under normal operation, increasing to 2.5 × 10⁶ nJ under attack) due to zero-trust quantum encryption (20–25 nJ/bit) activated when malicious nodes exceed a 20% threshold; a critical factor in QuanBioTrust sustainability is the integration of bio-energy harvesting (20–50 µW) from body motion and heat, which at an average rate of 30 µW offsets approximately 3 × 10⁶ nJ, ~ 30% of total energy consumption, over the simulation period, allowing nodes initialized with 0.1 J (capped at 0.2 J) to extend their operational lifetime and maintain functionality even under increased security loads; under adversarial conditions, specifically a 20% node compromise attack involving 50% false packet injection and DoS flooding at 2000 bits/s, the total energy cost rises to 10.6 × 10⁶ nJ, yet the framework maintains a Packet Delivery Ratio (PDR) > 90% and end-to-end latency < 200 ms, demonstrating that the added security overhead is both manageable and justified by the resulting resilience; all results are averaged over 10 independent simulation runs, with a standard deviation of less than 4%, and 95% confidence intervals reported in Table 7 and Fig. 4, ensuring high statistical reliability and reproducibility, thereby validating QuanBioTrust as a highly energy-efficient, self-sustaining, and secure solution for long-term, real-time biomedical monitoring in dynamic WBAN environments.

Table 7.

Energy Consumption Breakdown and Comparison (nJ, 10,000 Rounds).

Method	Sensing	Processing	Transmission	Security	Total	Savings versus QuanBioTrust	Attack total	Notes
QuanBioTrust	2.5 × 106	1.8 × 106	3.8 × 106	1.7 × 106	9.8 × 106	Baseline (~ 40%)	10.6 × 106	Harvesting offsets attack costs
HCEL15	2.5 × 106	2.5 × 106	6.0 × 106	1.5 × 106	12.5 × 106	22% less	13.0 × 106	Hybrid clustering overhead
EEMR16	2.5 × 106	2.8 × 106	5.5 × 106	1.5 × 106	12.3 × 106	25% less	12.8 × 106	EFBOA increases processing
EGWO17	2.5 × 106	3.0 × 106	5.8 × 106	1.8 × 106	13.1 × 106	34% less	13.7 × 106	High optimization cost
Fuzzy/Nash18	2.5 × 106	2.7 × 106	5.7 × 106	1.6 × 106	12.5 × 106	28% less	13.0 × 106	Fuzzy logic overhead
LF2ZO19	2.5 × 106	2.9 × 106	5.6 × 106	1.5 × 106	12.5 × 106	28% less	13.0 × 106	LF2ZO processing burden
EAFST20	2.5 × 106	2.4 × 106	6.2 × 106	1.4 × 106	12.5 × 106	28% less	12.9 × 106	Multi-hop increases transmission
EDC-ER21	2.5 × 106	1.5 × 106	4.0 × 106	0.5 × 106	8.5 × 106	13% more	9.0 × 106	Efficient for repetitive data

Fig. 4.

Energy Consumption Comparison.

Packet delivery ratio (PDR)

QuanBioTrust achieves a packet delivery ratio of 97% ± 1 under normal conditions and 90% ± 2 under attack in a 50-node Wireless Body Area Network processing real ECG data from the MIT-BIH Arrhythmia Database, where each node generates 50,000 packets (200 bits each at 10 Hz) over 10,000 simulation rounds, resulting in the successful delivery of 2.425 × 10⁶ out of 2.5 × 10⁶ total packets under normal operation, with only 3% loss attributed to node mobility (± 50% distance variation, 0.1–0.8 m from sink) and tissue attenuation (20–40 dB), and 2.25 × 10⁶ packets delivered under a 20% node compromise attack with DoS flooding at 2000 bits/s, where 10% loss stems from trust-based isolation and zero-trust encryption delays, with PDR dropping to 85% ± 2% under extreme attacks (> 50% node compromise); results are averaged over 10 independent simulation runs with a standard deviation of less than 2%, and 95% confidence intervals are reported in Table 8 and Fig. 5, confirming statistical reliability; the high delivery performance is enabled by fractal multi-hop routing (3–5 redundant paths per node, Step 5) that ensures path resilience against link failures due to body movement, and predator–prey trust monitoring (Step 3) that isolates malicious nodes within 2–4 rounds upon detection of trust scores below 0.5, preventing compromised nodes from disrupting data flow; QuanBioTrust 97% normal PDR surpasses HCEL (94%), EEMR (95%), EGWO (93%), LF2ZO (95%), EAFST (94%), and EDC-ER (92%), matches Fuzzy/Nash (96%), and its 90% attack resilience exceeds all compared protocols, HCEL (85%), EEMR (87%), EGWO (86%), LF2ZO (86%), EAFST (85%), and EDC-ER (80%), resulting in only 250,000 packets lost under attack, significantly fewer than HCEL and EAFST (375,000), EEMR (325,000), EGWO and LF2ZO (350,000), and EDC-ER (500,000); HCEL, EEMR, EGWO, and LF2ZO suffer 5–14% higher losses under attack due to optimization overheads and mobility sensitivity, while EAFST static multi-hop design and EDC-ER lack of security mechanisms exacerbate packet loss; QuanBioTrust handles ± 50% mobility more effectively than EAFST and Fuzzy/Nash by leveraging path multiplicity (N_paths · PDR > 0.85) to maintain > 95% PDR in dynamic postures, and under extreme conditions, its 85% PDR outperforms EDC-ER (80%) and matches HCEL/EAFST (85%); additional efficiency is gained through quantum-inspired data compression (Step 4), which reduces 200-bit ECG packets to 80 bits (60% reduction), minimizing transmission load and retransmission needs, although high mobility variance (> 50%) slightly increases packet loss from 3 to 15%; the framework superiority is statistically validated by a t-test confirming p < 0.01 versus EDC-ER, solidifying QuanBioTrust as a top-tier solution for reliable, real-time ECG data delivery in dynamic and adversarial WBAN environments.

Table 8.

Packet Delivery Ratio Comparison.

Method	PDR normal (%)	PDR attack (%)	Packets delivered (normal)	Packets lost (attack)	Mobility handling	Notes
QuanBioTrust	97 ± 1	90 ± 2	2.425 × 106 ± 25,000	250,000 ± 50,000	± 50% (robust)	Fractal routing, trust isolation
HCEL15	94 ± 2	85 ± 3	2.35 × 106 ± 50,000	375,000 ± 75,000	Moderate	Hybrid clustering limitations
EEMR16	95 ± 2	87 ± 3	2.375 × 106 ± 50,000	325,000 ± 75,000	Moderate	Multi-hop routing
EGWO17	93 ± 2	86 ± 3	2.325 × 106 ± 50,000	350,000 ± 75,000	Moderate	Optimization overhead
Fuzzy/Nash18	96 ± 1	88 ± 2	2.4 × 106 ± 25,000	300,000 ± 50,000	Limited	High PDR, fuzzy logic complexity
LF2ZO19	95 ± 2	86 ± 3	2.375 × 106 ± 50,000	350,000 ± 75,000	Moderate	Optimization steps
EAFST20	94 ± 2	85 ± 3	2.35 × 106 ± 50,000	375,000 ± 75,000	Static	Multi-hop, no mobility focus
EDC-ER21	92 ± 2	80 ± 4	2.3 × 106 ± 50,000	500,000 ± 100,000	Moderate	Compression, no security focus

Fig. 5.

PDR Comparison.

Security resilience

QuanBioTrust demonstrates exceptional security resilience in a 50-node Wireless Body Area Network processing real ECG data from the MIT-BIH Arrhythmia Database over 10,000 simulation rounds, successfully mitigating 90–99.9% of threats across five critical attack types: eavesdropping (99.9% detection, 5,000 intercepted packets blocked via quantum entanglement and 80-bit compressed data with 32-bit key, resulting in 99.98% reconstruction failure), data tampering (98.5% correction of + 20% falsified heart rate values), node compromise (90% mitigation of 50% false packet injection from 10 compromised nodes), Denial-of-Service (DoS) (92% resilience against 2,000 bits/s flooding with ECG-like noise bursts), and Sybil attacks (96% isolation of 10 fake identities within 10–40 s), achieving an average threat reduction of 95.3% ± 2%; results are averaged over 10 independent runs with a standard deviation of less than 3%, and 95% confidence intervals are reported in Table 9 and Fig. 6, with statistical significance confirmed by a t-test (p < 0.01 vs. EAFST); the framework robust defense is driven by quantum entanglement-based compression (Step 4) and zero-trust encryption (Step 7), which activates at a 20% predator threshold with an energy cost of 20–25 nJ/bit, while bio-energy harvesting (20–50 µW) offsets this overhead to maintain a Packet Delivery Ratio (PDR) of 90% under attack (compared to EDC-ER 80%) and 85% PDR under extreme attacks (> 50% compromise), sustained by fractal routing (3–5 paths, Step 5); mitigation times range from instantaneous (0 s) for eavesdropping due to quantum-secured transmission to 40 s for Sybil attacks, where the predator–prey trust model detects and isolates malicious nodes within 2–4 rounds; QuanBioTrust significantly outperforms state-of-the-art protocols, HCEL¹⁵ (~ 50% average reduction), EEMR¹⁶ (~ 44%), LF2ZO¹⁹ (~ 54%), EAFST²⁰ (~ 38%), and EDC-ER²¹ (~ 24%), which lack dynamic security and fail to contain sophisticated attacks, as evidenced by HCEL 50% eavesdropping protection versus QuanBioTrust 99.9%; it also surpasses EGWO¹⁷ (70%) and Fuzzy/Nash¹⁸ (82%), which suffer from slower mitigation (10–50 s) and weaker eavesdropping resistance (85–90%); the integration of quantum key distribution (QKD), self-evolving trust thresholds, and reinforcement from bio-inspired mechanisms ensures a proactive, adaptive, and information-theoretically secure architecture, positioning QuanBioTrust as a quantum-leap forward in WBAN security, offering unparalleled protection for real-time, safety–critical biomedical monitoring applications.

Table 9.

Security Resilience Comparison.

Method	Eavesdropping	Tampering	Compromise	DoS	Sybil	Avg. reduction (%)	Mitigation time (s)	Notes
QuanBioTrust	99.9% ± 0.1	98.5% ± 0.5	90% ± 2	92% ± 2	96% ± 1	95.3 ± 2	0–40	Quantum security, zero-trust
HCEL15	~ 60% ± 5	~ 55% ± 5	~ 50% ± 5	~ 40% ± 5	~ 45% ± 5	~ 50 ± 5	20–60	Moderate, basic trust mechanisms
EEMR16	~ 55% ± 5	~ 50% ± 5	~ 40% ± 5	~ 35% ± 5	~ 40% ± 5	~ 44 ± 5	25–70	Moderate, no robust security
EGWO17	~ 85% ± 4	~ 80% ± 4	~ 60% ± 5	~ 65% ± 5	~ 60% ± 5	~ 70 ± 4	15–50	High trust, optimization-based
Fuzzy/Nash18	~ 90% ± 3	~ 85% ± 3	~ 80% ± 4	~ 70% ± 4	~ 85% ± 3	~ 82 ± 3	10–45	High, fuzzy logic security
LF2ZO19	~ 65% ± 5	~ 60% ± 5	~ 55% ± 5	~ 40% ± 5	~ 50% ± 5	~ 54 ± 5	20–60	Moderate, limited security focus
EAFST20	~ 40% ± 6	~ 50% ± 5	~ 35% ± 6	~ 30% ± 6	~ 35% ± 6	~ 38 ± 6	30–80	Low, energy-focused, no security
EDC-ER21	~ 30% ± 6	~ 25% ± 6	~ 20% ± 6	~ 25% ± 6	~ 20% ± 6	~ 24 ± 6	40–90	Low, compression-only, no security

Fig. 6.

Security Resilience Comparison.

Latency

QuanBioTrust achieves an exceptionally low average end-to-end latency of 48 ms ± 5 ms under normal conditions and 185 ms ± 15 ms under attack (20% node compromise, DoS at 2000 bits/s) in a 50-node Wireless Body Area Network processing real ECG data from the MIT-BIH Arrhythmia Database, 50,000 packets per node, 200-bit packets at 10 Hz, over 10,000 simulation rounds, with normal latency primarily driven by quantum-inspired compression (Step 4), which reduces packet size from 200 to 80 bits and introduces only 5 ms of delay, and fractal multi-hop routing (Step 5) with 3–5 redundant paths contributing 40 ms of routing delay, ensuring high reliability and fault tolerance; under attack, latency increases due to the activation of zero-trust quantum encryption (Step 7), which adds 130–140 ms of processing overhead (at 20–25 nJ/bit) to secure communications, yet the total remains well below the critical 250 ms threshold for real-time ECG monitoring, rising to 260 ms ± 20 ms only under extreme attacks (> 50% node compromise); results are averaged over 10 independent simulation runs with a standard deviation of less than 8%, and 95% confidence intervals are reported in Table 10 and Fig. 7, confirming statistical reliability; QuanBioTrust 48 ms normal latency outperforms HCEL (60 ms), EEMR (70 ms), EGWO (65 ms), Fuzzy/Nash (68 ms), LF2ZO (62 ms), and dramatically surpasses EAFST (1000 ms), whose multi-hop propagation delays render it unsuitable for real-time applications, while matching EDC-ER (50 ms); under attack, its 185 ms latency beats all compared protocols: HCEL (200 ms), EEMR (210 ms), EGWO (205 ms), Fuzzy/Nash (215 ms), LF2ZO (200 ms), EAFST (1200 ms), and rivals EDC-ER 180 ms, despite providing significantly stronger security (130–140 ms overhead vs. EDC-ER 130 ms); the performance edge stems from efficient compression, optimized fractal paths, and dynamic trust management, enabling < 50 ms routing delay and > 95% PDR under normal conditions, with superior handling of ± 50% node mobility (0.1–0.8 m) compared to EAFST static multi-hop design and Fuzzy/Nash limited mobility focus; statistical validation via t-test confirms significance (p < 0.01 vs. EAFST), affirming that QuanBioTrust achieves an optimal balance between ultra-low latency and robust, adaptive security, making it ideal for real-time, safety–critical biomedical monitoring in dynamic WBAN environments, unlike EAFST high-latency architecture or EDC-ER minimal-security approach.

Table 10.

Latency comparison.

Method	Latency normal (ms)	Latency attack (ms)	Compression delay (ms)	Routing delay (ms)	Security delay (ms)	Mobility impact	Notes
QuanBioTrust	48 ± 5	185 ± 15	5	40	130–140 (attack)	Robust (± 50%)	Quantum compression, fractal paths
HCEL15	60 ± 8	200 ± 20	10	50	140 (attack)	Moderate	Hybrid clustering delays
EEMR16	70 ± 10	210 ± 20	15	55	140 (attack)	Moderate	Multi-hop routing
EGWO17	65 ± 9	205 ± 18	12	53	140 (attack)	Moderate	Optimization overhead
Fuzzy/Nash18	68 ± 10	215 ± 20	14	54	145 (attack)	Limited	Fuzzy logic processing
LF2ZO19	62 ± 8	200 ± 18	13	49	140 (attack)	Moderate	Optimization steps
EAFST20	1000 ± 50	1200 ± 60	10	990	200 (attack)	Static	Multi-hop propagation delay
EDC-ER21	50 ± 6	180 ± 15	5	45	130 (attack)	Moderate	Compression, minimal security

Fig. 7.

Latency Comparison.

Scalability analysis

As wireless body area networks evolve toward multi-patient monitoring and comprehensive physiological sensing, the scalability of routing protocols becomes a critical determinant of their real-world viability, and while QuanBioTrust demonstrates exceptional performance in a 50-node configuration, achieving a computational complexity of , ~ 500 operations/round, 97% PDR, and 48 ms latency, scaling to larger networks (100–200 nodes) introduces significant challenges in routing overhead, clustering complexity, and end-to-end latency; in a 100-node network under identical simulation conditions (MIT-BIH ECG data, 20–50 µW bio-energy harvesting, IEEE 802.15.6), the current implementation sees the Packet Delivery Ratio (PDR) drop from 97 to 91% and average latency rise to 280 ms under normal conditions, exceeding the 250 ms threshold for real-time ECG monitoring, with further degradation under attack (PDR: 87%, latency > 350 ms), primarily due to two non-linearly scaling components: fractal routing, which generates 3–5 redundant paths per node, leading to over 500 total paths in a 100-node network and causing increased routing table size, control overhead, channel contention, and packet collisions in the 2.4 GHz ISM band, and Quantum Particle Swarm Optimization (QPSO), whose complexity, though superior to quadratic methods, results in ~ 600 operations/round for 100 nodes, with more frequent re-clustering triggered by higher energy variance, thereby amplifying processing delay; in comparison, state-of-the-art protocols exhibit varying scalability: HCEL¹⁵ and EAFST²⁰, which employ hierarchical clustering, maintain better stability with moderate latency increases (60 ms → 120 ms) and PDR drops (94% → 90%) due to their constrained cluster scope, while EDC-ER²¹, with its complexity (~ 300 ops), scales best in computational efficiency but lacks dynamic security, rendering it vulnerable to Sybil and node compromise attacks in large-scale deployments; EGWO¹⁷ and Fuzzy/Nash¹⁸ face severe scalability bottlenecks, with computational loads exceeding 1000 operations/round and latencies surpassing 500 ms, leading to premature node failure and unreliable data delivery; to address these limitations, QuanBioTrust is designed with a forward-looking scalability roadmap that includes a hierarchical two-tier architecture featuring Super-Cluster Heads (SCHs) managing 10–20 clusters of 10–50 nodes each, reducing the effective QPSO complexity from O(m·log m) to (k = number of SCHs, ~ 10–20), drastically lowering computational load and re-clustering frequency; fractal routing is optimized by capping path multiplicity at 3–4 per node, and reinforcement learning (RL) is integrated to dynamically select high-trust, low-latency routes based on real-time energy, mobility, and security metrics, ensuring efficient and adaptive forwarding; this projected enhancement is expected to maintain PDR > 95% and latency < 100 ms even in 200-node networks, outperforming both classical hierarchical protocols and purely bio-inspired methods, thereby positioning QuanBioTrust not only as a high-performance solution for single-patient monitoring but as a foundational, scalable, and secure framework for future large-scale, real-time biomedical network applications. Table 11 summarizes the scalability performance of QuanBioTrust and related protocols in a 100-node WBAN.

Table 11.

Scalability Comparison in a 100-Node WBAN (Averaged over 10 runs, 95% CI reported).

Method	Network size	PDR (%)	Latency (ms)	Operations/round	Scalability notes
QuanBioTrust	50 nodes	97 ± 1	48 ± 5	~ 500	Optimal performance, low latency
QuanBioTrust (Projected)	100 nodes	91 ± 2	280 ± 30	~ 600	High PDR but latency exceeds threshold
QuanBioTrust (Hierarchical)	100–200 nodes	> 95	< 100	~ 200–400	Future design for high scalability
HCEL15	100 nodes	90 ± 3	120 ± 15	~ 700	Hierarchical design improves stability
EEMR16	100 nodes	88 ± 3	210 ± 20	~ 900	High overhead limits scalability
EGWO17	100 nodes	86 ± 3	520 ± 50	~ 1000	Severe latency issues at scale
Fuzzy/Nash18	100 nodes	85 ± 3	540 ± 60	~ 850	High computational overhead
LF2ZO19	100 nodes	86 ± 3	500 ± 50	~ 950	Optimization burden at scale
EAFST20	100 nodes	90 ± 3	1200 ± 100	~ 600	High latency due to multi-hop
EDC-ER21	100 nodes	92 ± 2	50 ± 6	~ 300	Best scalability but no security

Discussion and limitations of QuanBioTrust

QuanBioTrust demonstrates superior performance in a 50-node Wireless Body Area Network processing real ECG data from the MIT-BIH Arrhythmia Database (200-bit packets at 10 Hz) over 10,000 simulation rounds (1 round = 0.1 s), achieving a computational complexity of O(m·log m) with an average of ~ 500 operations per round (500 ops/round, 15 nJ/bit under normal conditions, rising to 25 nJ/bit under attack), driven by Quantum Particle Swarm Optimization (QPSO, Step 2, ~ 300 ops, 10 nJ/bit) for energy- and trust-aware cluster head selection and lightweight predator–prey trust monitoring (Step 3, ~ 100 ops, 5 nJ/bit), with zero-trust quantum encryption (Step 7, ~ 50 ops) adding complexity during a 20% node compromise attack (10 nodes, DoS at 2000 bits/s), outperforming state-of-the-art protocols including HCEL¹⁵ ()), ~ 700 ops, complex parent node selection), EEMR¹⁶ , n≈10, ~ 900 ops, EFBOA optimization), EGWO¹⁷ (, n≈15, ~ 1000 ops, gray wolf iterations), Fuzzy/Nash¹⁸ (), ~ 850 ops, equilibrium calculations), and LF2ZO¹⁹ (O(m·n), n≈12, ~ 950 ops, deep reinforcement learning), while surpassing EAFST²⁰ (, ~ 600 ops, static multi-hop focus) and EDC-ER²¹ (, ~ 300 ops, compression-only) by integrating robust security without sacrificing scalability, validated by a t-test (p < 0.01 vs. EGWO); the framework achieves a network lifetime of 12,800 ± 250 rounds (~ 21.3 days), with the first node failure occurring at 11,500 ± 200 rounds, significantly exceeding HCEL (~ 8000 rounds), EGWO (~ 7500 rounds), Fuzzy/Nash (~ 7,000 rounds), LF2ZO (~ 8500 rounds), and EAFST (~ 9,000 rounds), and rivaling EDC-ER (~ 11,000 rounds), thanks to bio-energy harvesting (20–50 µW, Step 6) and dynamic re-clustering at 20% energy variance (Step 7); under attack, lifetime decreases to 10,500 ± 300 rounds due to zero-trust encryption overhead, but this cost is offset by harvesting replenishing ~ 30% of energy (3 × 10⁶ nJ at 30 µW); total energy consumption is 9.8 × 10⁶ nJ (sensing: 2.5 × 10⁶ nJ, processing: 1.8 × 10⁶ nJ, transmission: 3.8 × 10⁶ nJ, security: 1.7 × 10⁶ nJ), representing ~ 40% savings over HCEL (12.5 × 10⁶ nJ), EEMR (12.3 × 10⁶ nJ), EGWO (13.1 × 10⁶ nJ), Fuzzy/Nash (12.5 × 10⁶ nJ), LF2ZO (12.5 × 10⁶ nJ), and EAFST (12.5 × 10⁶ nJ), achieved through quantum-inspired compression (50–70%, Step 4, reducing 200-bit packets to 80 bits) and fractal routing (3–5 paths, Step 5), though it trails EDC-ER (8.5 × 10⁶ nJ) by 13% due to its security overhead, a trade-off justified by superior adaptability to variable ECG signals (e.g., arrhythmia spikes, detected and isolated in 2–4 rounds); the framework maintains a Packet Delivery Ratio (PDR) of 97% ± 1% under normal conditions and 90% ± 2% under attack, dropping to 85% ± 2% under extreme attacks (> 50% compromise), with an average end-to-end latency of 48 ± 5 ms normally and 185 ± 15 ms under attack, well below the 250 ms threshold for real-time monitoring, outperforming EAFST (1000/1200 ms) and rivaling EDC-ER (50/180 ms); however, limitations exist: performance assumes consistent bio-energy harvesting (~ 30 µW), and in sedentary states (< 10 µW), lifetime may drop to ~ 10,000 rounds; under high interference (> 60 dB) or extreme mobility, PDR may fall below 90% and latency exceed 250 ms; scalability beyond 50 nodes increases fractal routing complexity and QPSO load (up to 600 ops/round for 100 nodes), where hierarchical designs like HCEL and EAFST maintain better stability.

QuanBioTrust, while demonstrating significant advancements in security, energy efficiency, and reliability for WBANs, has several important limitations that stem from its simulation-based evaluation and the inherent trade-offs of its innovative design. The current framework is entirely validated through simulation using NS-3 and Qiskit, meaning its quantum-inspired components (QPSO, entanglement-based encryption) are classically emulated and do not reflect the decoherence, gate fidelity, and hardware constraints of real quantum processors, which remain a significant barrier to immediate deployment. The framework multi-step, bio-inspired architecture, while effective, introduces higher computational complexity (~ 500 operations/round) compared to simpler protocols, which may challenge ultra-low-power microcontrollers in resource-constrained sensor nodes. Scalability is currently limited to around 50 nodes; extending to larger networks increases fractal routing overhead and QPSO optimization time, potentially leading to latency exceeding 250 ms and reduced PDR, necessitating future hierarchical extensions. Performance is also contingent on consistent bio-energy harvesting (20–50 µW), and its effectiveness may diminish for sedentary users where harvesting rates drop below 10 µW, reducing network lifetime. The mobility and channel models, though based on IEEE 802.15.6, are simplified and may not fully capture the dynamic interference and link instability of real-world environments. Finally, the security model, while robust against known attacks, assumes a static threat profile and may be vulnerable to novel, adaptive attacks not present in the current simulation, and the zero-trust mode increased energy cost (20–25 nJ/bit) relies on harvesting to remain sustainable. These limitations highlight the need for future work in hardware prototyping, scalability enhancements, and real-world validation.

In future work, we will transition from classical emulation to real-world quantum-enabling technologies by developing cost-effective quantum-capable sensor nodes (~ $100–$150/node) using Photonic Integrated Circuits (PICs) with CMOS-compatible waveguides and single-photon detectors for lightweight Quantum Key Distribution (QKD) and quantum-enhanced processing (~ 20–25 nJ/bit, < 50 mW), spin-based qubits (e.g., nitrogen-vacancy centers, quantum dots) offering ultra-low power (< 10 µW), room-temperature coherence (> 1 ms at 310 K), and minimal footprint (< 1 cm²), and hybrid classical-quantum System-on-Chip (SoC) designs integrating ARM Cortex-M microcontrollers with photonic or spin modules fabricated via high-volume processes (e.g., TSMC 65 nm), while validating quantum components through IBM Quantum Cloud to test entanglement generation and multi-qubit encoding on real processors under noise; to overcome scalability limitations, where fractal routing (> 5 paths) and QPSO complexity increase latency beyond 250 ms for networks > 50 nodes, we will develop a hierarchical two-tier architecture introducing Super-Cluster Heads (SCHs) managing 10–20 clusters of 10–50 nodes, reducing complexity to ( (k = number of SCHs, ~ 10–20), capping fractal paths to 3–4 per node via reinforcement learning (RL) to prioritize high-trust, low-latency routes, and implementing a distributed QPSO algorithm that parallelizes cluster head selection across SCHs (5–10 ops/SCH, 5 nJ/bit) using quantum-inspired gossip protocols to maintain < 100 ms latency and > 97% PDR in 100–200 node networks; to enhance robustness against real-world variability, particularly the dependence on bio-energy harvesting (20–50 µW), we will conduct a sensitivity analysis under reduced harvesting (5–20 µW) to model sedentary users, assessing impacts on PDR, latency, lifetime, and energy consumption (e.g., rise to 11 × 10⁶ nJ), and mitigate this with micro-supercapacitors (1–10 mF, 0.2 J capacity, η = 0.8–0.9) for energy buffering and hybrid thermal-kinetic harvesters (piezoelectric + thermoelectric) to stabilize output at ~ 30 µW across activity levels, while refining the predator–prey trust model for interference robustness by dynamically adjusting thresholds to maintain PDR > 90% under high noise (> 60 dB); finally, to bridge the simulation-to-reality gap, we will prototype a 10–20 node testbed using open-source tools and COTS hardware (nRF52840), deployed under realistic conditions (± 50% mobility, 0.5–2 m range, 60 dB interference, 30 µW harvesting), and conduct hospital pilot studies to validate performance, targeting > 95% PDR, < 50 ms latency, 95% threat reduction, and 12,000-round lifetime, demonstrating QuanBioTrust superiority over classical protocols like HCEL and EDC-ER in real-world biomedical monitoring, ensuring it evolves from a simulation-validated framework into a practical, scalable, and deployable solution for secure, long-term, real-time healthcare.

Conclusion and future works

QuanBioTrust, a novel Wireless Body Area Network framework, integrates quantum-inspired optimization, bio-energy harvesting, fractal routing, and zero-trust security to deliver superior performance in a 50-node simulation using the MIT-BIH Arrhythmia Database over 10,000 rounds. It outperforms existing protocols, HCEL, EEMR, EGWO, Fuzzy/Nash, LF2ZO, EAFST, and EDC-ER, across key metrics: computational complexity of O(m·log m) (~ 500 operations/round, 15 nJ/bit) surpasses HCEL (O(m²), ~ 700 ops) and EGWO (O(m·n), ~ 1000 ops); network lifetime of 12,800 rounds (~ 3 weeks) exceeds HCEL (~ 8000) and EDC-ER (~ 11,000) with bio-energy harvesting (20–50 µW); energy consumption of 9.8 × 10⁶ nJ (~ 40% savings) outpaces HCEL (12.5 × 10⁶ nJ); packet delivery ratio (PDR) of 97% (normal) and 90% (attack) surpasses EAFST (94%/85%); security resilience (95.3%, 99.9% eavesdropping protection) exceeds EGWO (70%) and EDC-ER (24%); and latency of 48 ms (normal) and 185 ms (attack) ensures real-time performance (< 250 ms) compared to EAFST (1000/1200 ms) (p < 0.01 vs. EAFST). Validated by statistical significance (p < 0.01) and consistency (standard deviation < 5%), QuanBioTrust sets a new benchmark for scalable, secure biomedical data transmission. However, its performance assumes ideal conditions, such as consistent bio-energy harvesting (20–50 µW) and stable node behavior and may vary in less constrained or erratic environments (e.g., high interference > 60 dB or irregular mobility), potentially reducing PDR to < 90% or increasing latency beyond 250 ms. Challenges include high quantum hardware costs (> $1000/node), vulnerability to extreme attacks (85% PDR at > 50% node compromise), and harvesting variability in sedentary states (< 10 µW). Future work will enhance applicability by developing cost-effective quantum nodes using photonic circuits (~ $100–$200/node), improving resilience to extreme attacks with dynamic trust thresholds (adjusting to 20–80% compromise) to maintain PDR > 90%, stabilizing harvesting with hybrid thermal-kinetic sources for consistent 30 µW output, scaling beyond 50 nodes with hierarchical routing for < 100 ms latency at 100–200 nodes, validating performance under real-world interference and extended ranges, and integrating machine learning for predictive trust and compression to reduce energy to < 8 × 10⁶ nJ and latency to < 40 ms. These advancements aim to solidify QuanBioTrust as a practical, next-generation WBAN solution for real-time biomedical monitoring.

Author contributions

All authors contributed equally to the conceptualization, formal analysis, investigation, methodology, and writing and editing of the original draft. All authors have read and agreed to the published version of the manuscript.

Data availability

The code and datasets used during the current study are available at: https://github.com/obaidalii2016-creator/Quantum-Bio-Energy-Trust-Framework-for-Secure-WBANs.git

Declarations

Competing interests

The authors declare no competing interests.