Last Updated on December 7, 2023 by pg@petergamma.org
A paper which used a 16 channel OpenBCI EEG heatset with WIFI shield published in the year 2023:
https://www.mdpi.com/1424-8220/23/7/3763
where did they buy they WIFI shield? It is not available anymore from www.OpenBCI.com since many years.
First author if this papers was was Yeison Nolberto Cardona Alvarez:
https://www.linkedin.com/in/yeisoncardona/?originalSubdomain=co
- Python Ninja, Electronic Engineer, M.Sc., Doctoral student, Developer, BCI researcher.
- It is very likely that Yeison is programming at this very moment
The authors of this papers connected the OpenBCI setup wirelessly to a Raspberry Pi.
In the paper they show 16 EEG channels viszualised in OpenBCI GUI together with a time-frequency plot:
- We cannot see any signs of packet loss and cyclic noise spiking in the EEG signal, as we would expect if the authors would have used a faulty WIFI shield.
- The following table shows the maximal sampling rate of OpenBCI Cyton. It drops for TCP over WIFI from 16 KHz for 8 channels to 8 KHz for 16 channels.
Conclusion of the paper
We introduced a flexible, scalable, and integral OpenBCI framework for supporting EEG-based neurophysiological experiments. For such a purpose, the single-board OpenBCI Cyton was chosen, and a brand new set of drivers was developed to maximize the hardware benefits of the ADS1299. Our approach supports multiple sampling rates, packaging sizes, communication protocols, and free electrode placement, making it suitable for EEG data. Furthermore, an innovative feature for marker synchronization was also added. The system operates in a distributed manner, which allows for the controlled execution of critical processes such as acquisition, stimuli delivery, and real-time data analysis. Achieved results under a motor imagery paradigm demonstrate that the system’s robustness and stability are maintained through dedicated handling of the OpenBCI hardware. Realtime streaming is guaranteed within acceptable latency, and jitter ranges for closed-loop BCI compared to state-of-the-art approaches. The development environment provides a complete API, automatic background configuration, and a range of easy-to-use widgets for stimuli delivery, making it an ideal platform for BCI data processing and custom extension development.
Future work will extend the proposed framework to 32 and 64 EEG channels [47,48]. This work uses the OpenBCI Cyton board, one of the highest-performing hardware devices. However, we want to test upcoming acquisition boards integrating the most recent technology and communication protocols. Furthermore, close-loop approaches and advanced machine and deep learning algorithms will be tested to study poor skills issues with OpenBCI-based solutions [36,49].
Comments are closed.