Source-linked AI summary
PLATOS: A Power and Latency-Aware Task-Oriented Scheduling Strategy for Healthcare IoT in Fog Computing
Mohammed Alaa Ala'anzy, Zulfiqar Ahmad, Zhanar Mukash
TL;DR
Healthcare IoT fog systems must manage latency and energy constraints, especially for battery-operated devices. PLATOS uses four-phase task-oriented scheduling to optimize both objectives, reducing energy consumption by 18.72% and latency by 8.65% versus FTDM in simulation.
Problem
Healthcare IoT fog environments must process data with low latency and energy use while many devices remain battery-powered.
Method
PLATOS categorizes HIoT tasks and selects fog resources through sequential latency optimization, power optimization, and decision-making phases.
Results
18.72% lower energy consumption and 8.65% lower latency were achieved by PLATOS compared with FTDM in simulation.
Takeaways & Limitations
PLATOS improves the efficiency and responsiveness of HIoT systems within the reported simulation evaluation.
Takeaways & Limitations
Future work will test PLATOS in large-scale real-world HIoT deployments beyond simulation environments.
Abstract
from arXiv · showhide
Healthcare Internet of Things (HIoT) technology is revolutionising the healthcare industry by enabling real-time data collection and analysis for personalised patient care. However, the rapid expansion of HIoT technology introduces challenges such as increased latency and higher energy consumption in fog computing environments, particularly when managing battery-operated devices. To address these issues, this work proposes a novel scheduling strategy that optimises both power consumption and latency through task-oriented scheduling for HIoT tasks. The proposed strategy, named PLATOS (Power and Latency Aware Task Oriented Scheduling), is implemented in four sequential phases. In the first phase, HIoT tasks are categorised into three groups: priority-oriented, storage-oriented, and computational-oriented. The second phase focuses on latency optimisation by identifying the fog computing resources that yield the lowest execution delay for each task category. In the third phase, power optimisation is achieved by selecting the resources that minimise energy consumption. Finally, in the decision-making phase, high-performance fog resources are allocated to high-priority tasks while the remaining tasks are scheduled based on a mapped list derived from the latency and power optimisation phases. Simulation experiments conducted in iFogSim2 demonstrate that PLATOS reduces energy consumption by 18.72% and latency by 8.65% when compared to the state-of-the-art. These improvements enhance the efficiency and responsiveness of HIoT systems and contribute to more effective patient care and proactive healthcare service delivery.
1 Introduction
Healthcare IoT fog environments must process growing, sensitive data with low latency and limited energy, while accommodating heterogeneous devices and scalable resource allocation. PLATOS addresses these constraints through task-oriented scheduling that jointly optimizes latency and power across four phases.
- Motivation: HIoT combines connected medical devices, wearables, and sensors for real-time health data collection and transmission.This supports patient monitoring, early diagnosis, and data-driven healthcare decisions.
- Challenges: Battery-powered IoT devices require energy-efficient task allocation to extend operation and maintain network efficiency.Resource allocation must assign tasks to devices capable of executing them with minimal energy use.
- Challenges: Healthcare IoT resource allocation must address low-latency processing, data security, device heterogeneity, and scalability.Distributed fog environments complicate latency minimization, while sensitive data may require additional processing and encryption.
- PLATOS strategy: PLATOS categorizes HIoT tasks as priority-oriented, storage-oriented, or computational-oriented before optimizing resource selection.The strategy is designed for power and latency optimization through task-oriented scheduling.
- PLATOS strategy: PLATOS orders resources by minimum latency and power consumption, then maps them to tasks during decision-making.High-performance fog resources serve high-priority tasks, while remaining tasks use the mapped resource list.
- Evaluation: The strategy is evaluated in simulation using energy consumption and latency, with results compared against existing state-of-the-art strategies.The paper evaluates PLATOS through a simulation environment rather than a real-world deployment.
2 Related Works
Prior work develops fog-based scheduling and resource-allocation methods for reducing latency, energy use, and deadline violations in heterogeneous IoT and healthcare settings. These approaches include task classification, heuristic optimization, fault tolerance, and energy-aware allocation, but real-world validation remains an identified need.
- Fog scheduling objectives: Fog schedulers have targeted network usage, delay, heterogeneous task deployment, energy consumption, deadline violations, and QoS requirements.Proposed methods include fuzzy logic and semi-greedy algorithms for mapping tasks to constrained fog resources.
- Research boundary: Further research and testing are needed to validate the EEIoMT framework in real-world settings and assess its potential impact on patient outcomes.The stated boundary is validation beyond the reported framework evidence.
- Healthcare fog management: FTDM manages healthcare IoT data and task or node failures through a two-way mechanism, with simulation improvements over GKS.The scheme is positioned for remote treatment scenarios requiring energy-aware and fault-tolerant data management.
- Optimization approaches: Optimization-based approaches include MGWO, CHMPAD, enhanced semi-greedy methods, and OEeRA for distributing or offloading IoT tasks.These methods use combinations of metaheuristics, fuzzy logic, resource allocation, and fault remediation.
- Fault tolerance: Fault-tolerant scheduling research reschedules tasks after fog-device failure and seeks to limit the resulting performance degradation.Reactive fault tolerance and modified particle swarm optimization are used to identify executable replacement nodes.
- Healthcare IoT scheduling: Healthcare fog scheduling research prioritizes rapid processing of remote-monitoring tasks to address latency from transferring data between sensors and the cloud.Task classification methods distinguish high-, medium-, and low-importance tasks based on patient health status.
3 System Design and Model
The system models task allocation across terminal devices, fog nodes, and cloud servers, then applies PLATOS to balance latency and power for HIoT tasks. PLATOS categorizes tasks, ranks resources by latency and power, and maps those rankings for final allocation.
- System Architecture: The Cloud-Fog architecture connects terminal devices to fog nodes and cloud servers through a fog broker that selects suitable execution resources.The broker analyzes task characteristics and schedules requests according to transmission delay and energy criteria.
- Task Scheduling Model: Each task is described by size, length, type, and deadline, while each fog node is described by storage capacity, computing capacity, and energy capacity.Task types are normal, moderate, or critical; task length is measured in Millions of Instructions, and node computing capacity in MIPS.
- Task Scheduling Model: The scheduling problem allocates IoT tasks to fog or cloud resources while optimizing latency and energy consumption under resource constraints.The formulation includes computing capacity, storage capacity, battery capacity, response-time, and binary assignment constraints.
- PLATOS Strategy: PLATOS is a four-stage power- and latency-aware scheduling strategy for HIoT tasks.The stages organize task handling and resource optimization.
- Task Orientation: The task-orientation stage classifies HIoT tasks as priority-oriented, storage-oriented, or computationally-oriented for targeted allocation.Priority tasks are scheduled first, storage-oriented tasks require memory, and computationally-oriented tasks require processing power.
- Optimization and Decision Making: The latency- and power-optimization stages rank resources by ascending latency and power usage before the decision-making stage maps these rankings.The final stage allocates high-performance resources to high-priority tasks and integrates both optimization results.
4 Evaluation Methods
The evaluation uses iFogSim2 to simulate healthcare IoT fog scenarios, including cardiovascular health monitoring, and measures energy consumption and latency. The study reports improvements in both metrics for PLATOS compared with FTDM.
- Simulation Environment: The evaluation uses the iFogSim2 simulation environment to assess PLATOS under different configurations and operating conditions.iFogSim2 supports mobility models, service migration, and distributed edge/fog clusters.
- Simulation Scenario: The simulated application is Cardiovascular Health Monitoring, which uses smart healthcare IoT sensors for ECG monitoring and heart-disease diagnosis.The application models a loop of smart sensors that sense and transmit ECG signals.
- Evaluation Metrics: The experiments focus on energy consumption and latency as performance parameters for healthcare IoT applications.These metrics reflect efficiency and responsiveness where battery-powered devices and real-time analysis are common.
- Results: PLATOS demonstrates significant improvements in energy consumption and latency compared with the FTDM strategy.The conclusion identifies both metrics as improved relative to FTDM.
5 Experiments, Results, and Discussion
The experiments evaluate PLATOS for energy consumption, latency, and scalability across fog-device deployments. PLATOS reduces energy consumption and latency relative to FTDM while maintaining lower energy-latency measures as deployments scale.
- Energy Consumption: 18.72% lower energy consumption is achieved by PLATOS than by the existing FTDM strategy.The strategy avoids excessive power drain on individual devices and balances load more effectively.
- Latency: 8.65% lower latency is reported for PLATOS, enhancing healthcare-system responsiveness.The evaluation treats latency as critical for timely healthcare decision-making.
- Evaluation Metrics: The DEI metric combines the number of fog devices, energy consumption, and latency, while adding 10^-6 to avoid division by zero.DEI is used to evaluate fog-device efficiency across energy and latency dimensions.
- Scalability: DEI increases more favorably for PLATOS than for the benchmark as the number of fog devices grows, indicating better efficiency at larger scales.The comparison covers deployments with 15, 30, 45, and 60 fog devices.
- Scalability: PLATOS maintains lower ELDP than the benchmark as fog devices increase, with a more controlled increase beyond 45 devices.The benchmark shows a linear ELDP increase, whereas PLATOS slightly decreases from 45 to 60 devices.
6 Conclusion
PLATOS schedules HIoT tasks through task categorisation, latency and power optimisation, and decision-making across fog resources. In iFogSim2 simulations, it improved energy consumption and latency relative to the state-of-the-art, while future work targets fault tolerance, dynamic learning, and real-world validation.
- PLATOS approach: PLATOS implements four stages, beginning by categorising HIoT tasks into priority-oriented, storage-oriented, and computationally-oriented groups.The remaining stages identify low-delay resources, select energy-efficient resources, and use their mapped list for scheduling.
- Results: 18.72% lower energy consumption and 8.65% lower latency were achieved than the present state-of-the-art strategy in iFogSim2 simulations.The reported comparison evaluates the proposed PLATOS method against the state-of-the-art strategy.
- Results: PLATOS improved fog-resource efficiency and scalability as the number of fog devices increased, according to the DEI and ELDP measures.The ELDP was reported as showing greater scalability and efficiency particularly with more fog devices.
- Future work: Future work will add fault tolerance, integrate machine learning for dynamic task classification and resource allocation, and test PLATOS in large-scale real-world HIoT deployments.These extensions are intended to validate effectiveness beyond simulation environments.