In this final project it is done a simple prototype, not complex in order not to overload the packet network or the computational part, of a sensor network, which communicating through wireless body area network (WBAN) are able to characterize daily activities. The nodes used were the Adafruit HUZZAH32 from the company Adafruit, it’s a System on Chip, which incorporates a Wi-Fi module that has been used for the communication between devices.
Firstly, an analysis has been done of the available system. On the one hand, an analysis of the devices and on the other hand a study of one possible characterization from data already collected.
In a second phase, the software of the devices has been modified to create the sensor network and to communicate with each other. For this purpose, the Wi-Fi module of the devices was used, after which, once they were connected, a series of experiments were carried out for different scenarios. With these experiments it has been possible to set thresholds for the development of the final classification algorithm.
Finally, in a third phase, the different tests have been exposed according to the algorithm performed in the second phase.
The results obtained have shown that it is a valid algorithm for the characterization of activities. In addition, an accelerometer has been included to differentiate more activities.
On July 15, 2020, the master student Pablo Sarabia Ortiz read and defended his master thesis entitled “Design and evaluation of electromyography signal processing techniques using resource-constrained devices”. This master thesis is enclosed in the current B105 Electronic Systems Lab research topic of acquiring and processing electromyography (EMG) signals on the human body to achieve a wearable health device based on EMG signals.
Surface electromyographic (sEMG) is an acquisition technique based on recording muscles potential over the skin. sEMG based devices have a wide range of application: early diagnose and treatment of neurodegenerative diseases, tracking of daily activities, rehabilitation, and adaptive training. sEMG signals are complex and present different challenges like great amount of data, complex signals, and significant variations between subjects and days. For most of these applications is required to identify and classify the gestures or movements that the user is doing. This classification is a task that requires great amount of resources (memory and CPU). This thesis is focused in understanding the sEMG signal characteristics and designing a classifier for hand gestures, by using the custom acquisition board.
Picture of the hardware used for sEMG acquisition. On the left the electrodes, on top of the image a preamplifier and on the bottom right corner the stack of PCBs composed of the microcontroller and the ADC.
First, a quantitative analysis of the sEMG data was carried out by using parallel factor analysis (PARAFAC). The dataset used was NINAPRO, because it contains numerous different hand gestures performed by different subjects in different days. This PARAFAC analysis showed that is possible to reduce the number of channels from 16 to 4 without significant loss of information, as shown in the figure below. It also showed that most of the information is under the 350 Hz range. PARAFAC proved to be an interesting method for choosing the most significant channels in the dataset.
Process followed to do the PARAFAC analysis of the data from the NINAPRO dataset.
Second, an acquisition system to log the data to the computer was established. This acquisition system had 4 channels at a sampling rate of 500 Hz each. The data once logged was formatted and stored using MATLAB. Eight different gestures were performed, as shown in the figure. Then a support vector (SVM) machine classifier was trained obtaining an 99% accuracy in cross validation.
Table with all the gestures recorded for the master thesis.
Third, a two level three variables factorial design was
carried out to model the influence of the design variables in three features of
the classifier (execution time, memory footprint and accuracy). The three design
variables studied were: codification of the SVM, data precision (float32 or
float64) and length of the sample. The results shown that float64 should never
be used, and that there is always a tradeoff between classifier accuracy versus
the memory footprint and speed of the classifier. It was also identified the
memory footprint as the bottleneck for the use of the classifier in a resource-constrained
device. It was achieved a reduction of 1/14 of the original memory footprint
and a speedup of 233 times, however accuracy of the classifier lowered to 85%.
