IoT-Controlled EEG Monitoring and Decision Support System for Autism Therapy Using Signal Analysis and Embedded Processing

Abstract

Autism spectrum disorder (ASD) is associated with irregular neural activity that complicates therapy planning when based solely on behavioral observation. This study introduces an Internet of Things (IoT)-based electroencephalography (EEG) analysis system designed to support real-time therapy recommendations for children with ASD. EEG signals were processed using the fast Fourier transform for noise reduction and evaluated through power spectrum density (PSD) using Welch's method to characterize spectral patterns. The system was implemented on a Raspberry Pi platform integrated with an IoT application for remote monitoring and automated therapy notification. EEG recordings were segmented into two 30-s loops from 16 channels with a sampling rate of 256 Hz. Spectral analysis showed dominant Delta activity with frequencies of 1.68 Hz and 1.75 Hz and PSD values reaching up to 1280.31 µV2 /Hz across segments. Repeated-measures analysis of variance indicated a statistically significant increase only in the Beta band (F(1,15) = 10.06, p = 0.006), while other frequency bands remained stable. Based on dominant-band detection, the system generated automated therapy recommendations and transmitted them to the IoT interface in real time. The results demonstrate the technical feasibility of integrating EEG signal analysis, embedded processing, and IoT communication for assistive therapy decision support in ASD management. Future work will extend the framework to multi-subject datasets to improve clinical generalizability.



Received: 12 December 2025 | Revised: 13 April 2026 | Accepted: 11 July 2026



Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.



Data Availability Statement

The KAU BCI datasets that support the findings of this study are openly available at https://www.academia.edu/8920466/Diagnosis_Autism_by_Fisher_Linear_Discriminant_Analysis_FLDA_via_EEG, reference number [16].



Author Contribution Statement

Melinda Melinda: Conceptualization, Methodology, Resources, Writing – review & editing, Supervision, Project administration, Funding acquisition. Nizam Albar: Software, Formal analysis, Data curation, Writing – original draft, Visualization. Yuwaldi Away: Methodology, Investigation. Karlisa Priandana: Validation, Writing – review & editing. Prima Dewi Purnamasari: Validation. Muhammad Saifullah Nur: Data curation. Nurlida Basir: Writing – review & editing. Syahrul Gazali: Validation, Resources.