Estimated Power Values Research Articles

Estimation of Ankle Joint Power during Walking Using Two Inertial Sensors.

(1) Background: Ankle joint power, as an indicator of the ability to control lower limbs, is of great relevance for clinical diagnosis of gait impairment and control of lower limb prosthesis. However, the majority of available techniques for estimating joint power are based on inverse dynamics methods, which require performing a biomechanical analysis of the foot and using a highly instrumented environment to tune the parameters of the resulting biomechanical model. Such techniques are not generally applicable to real-world scenarios in which gait monitoring outside of the clinical setting is desired. This paper proposes a viable alternative to such techniques by using machine learning algorithms to estimate ankle joint power from data collected by two miniature inertial measurement units (IMUs) on the foot and shank, (2) Methods: Nine participants walked on a force-plate-instrumented treadmill wearing two IMUs. The data from the IMUs were processed to train and test a random forest model to estimate ankle joint power. The performance of the model was then evaluated by comparing the estimated power values to the reference values provided by the motion tracking system and the force-plate-instrumented treadmill. (3) Results: The proposed method achieved a high accuracy with the correlation coefficient, root mean square error, and normalized root mean square error of 0.98, 0.06 w/kg, and 1.05% in the intra-subject test, and 0.92, 0.13 w/kg, and 2.37% in inter-subject test, respectively. The difference between the predicted and true peak power values was 0.01 w/kg and 0.14 w/kg with a delay of 0.4% and 0.4% of gait cycle duration for the intra- and inter-subject testing, respectively. (4) Conclusions: The results of this study demonstrate the feasibility of using only two IMUs to estimate ankle joint power. The proposed technique provides a basis for developing a portable and compact gait monitoring system that can potentially offer monitoring and reporting on ankle joint power in real-time during activities of daily living.

Prediction of power output at different running velocities through the two-point method with the Stryd™ power meter

BackgroundThe force- and power-velocity (F–V and P–V, respectively) relationships have been extensively studied in recent years. However, its use and application in endurance running events is limited. Research questionThis study aimed to determine if the P–V relationship in endurance runners fits a linear model when running at submaximal velocities, as well as to examine the feasibility of the “two-point method” for estimating power values at different running velocities. MethodsEighteen endurance runners performed, on a motorized treadmill, an incremental running protocol to exhaustion. Power output was obtained at each stage with the Stryd™ power meter. The P–V relationship was determined from a multiple-point method (10, 12, 14, and 17 km·h−1) as well as from three two-point methods based on proximal (10 and 12 km·h−1), intermediate (10 and 14 km·h−1) and distal (10 and 17 km·h−1) velocities. ResultsThe P–V relationship was highly linear ( r = 0.999). The ANOVAs revealed significant, although generally trivial (effect size < 0.20), differences between measured and estimated power values at all the velocities tested. Very high correlations ( r = 0.92) were observed between measured and estimated power values from the 4 methods, while only the multiple-point method ( r2 = 0.091) and two-point method distal ( r2 = 0.092) did not show heteroscedasticity of the error. SignificanceThe two-point method based on distant velocities (i.e., 10 and 17 km·h−1) is able to provide power output with the same accuracy than the multiple-point method. Therefore, since the two-point method is quicker and less prone to fatigue, we recommend the assessment of power output under only two distant velocities to obtain an accurate estimation of power under a wide range of submaximal running velocities.

Detection of steering direction using EEG recordings based on sample entropy and time-frequency analysis.

Monitoring driver's intentions beforehand is an ambitious aim, which will bring a huge impact on the society by preventing traffic accidents. Hence, in this preliminary study we recorded high resolution electroencephalography (EEG) from 5 subjects while driving a car under real conditions along with an accelerometer which detects the onset of steering. Two sensor-level analyses, sample entropy and time-frequency analysis, have been implemented to observe the dynamics before the onset of steering. Thus, in order to classify the steering direction we applied a machine learning algorithm consisting of: dimensionality reduction and classification using principal-component-analysis (PCA) and support-vector-machine (SVM), respectively. The results showed an increase of the sample entropy and the estimated power values in the theta and alpha frequency bands, 100 ms before the onset of steering. The detection of steering direction depicted that sample entropy gives a higher classification accuracy (73.5% ±6.8) as compared to that of using the estimated power for theta and alpha frequency bands (62.6% ±5.6).

Distributed algorithms for sum rate maximization in multi-cell downlink OFDMA with opportunistic DF relaying

This paper considers a multi-cell orthogonal frequency division multiple access (OFDMA) downlink system with several decode-and-forward (DF) relay stations (RSs) aiding the base station (BS) transmissions. The problem considered is the maximization of the system sum rate with a total power constraint in each cell. An iterative semi-distributed resource allocation (RA) algorithm is first proposed to optimize mode selection (decision whether relaying should be used or not and which relay), subcarrier assignment (MSSA), and power allocation (PA), alternatively. During the MSSA stage, the problem is decoupled into subproblems which can be solved distributively in linear time. During the PA stage, an algorithm based on single condensation and Lagrange duality (SCLD) is designed to optimize PA with the tentative MSSA results. The convergence of the SCLD-based RA algorithm is theoretically guaranteed and an local optimum is reached after convergence. To solve the formulated problem autonomously, a modified iterative water-filling (IWF) algorithm is further proposed. Specifically, each cell autonomously optimizes its own sum rate with the estimated power values of the received interferences from the other cells. An optimum algorithm is proposed to solve the local RA problem in each cell. Through numerical experiments, the convergence of the two proposed algorithms as well as their benefits compared with a centralized algorithm (CA) are illustrated.