A precise sensor fault detection technique using statistical techniques for wireless body area networks

Abstract

One of the major challenges in wireless body area networks (WBANs) is sensor fault detection. This paper reports a method for the precise identification of faulty sensors, which should help users identify true medical conditions and reduce the rate of false alarms, thereby improving the quality of services offered by WBANs. The proposed sensor fault detection (SFD) algorithm is based on Pearson correlation coefficients and simple statistical methods. The proposed method identifies strongly correlated parameters using Pearson correlation coefficients, and the proposed SFD algorithm detects faulty sensors. We validated the proposed SFD algorithm using two datasets from the Multiparameter Intelligent Monitoring in Intensive Care database and compared the results to those of existing methods. The time complexity of the proposed algorithm was also compared to that of existing methods. The proposed algorithm achieved high detection rates and low false alarm rates with accuracies of 97.23% and 93.99% for Dataset 1 and Dataset 2, respectively.

ADSBAN: Anomaly detection system for body area networks utilizing IoT and machine learning

“…Subscribers (4) subscribe to the specific topic. The network traffic is captured and sent to a computer (5), where data preprocessing is carried out. ML models are trained using the input features, and finally, an ADS (6) detects the anomalies and generates an alert.…”

“…It allows for model comparison based on its ability to identify positive and negative samples. It is calculated as per Equation (5).…”

“…The collected data is preprocessed with the help of statistical techniques and simultaneously transferred to the server to monitor health. 5 Further, the preventive decision is made with the help of ML or DL techniques. The number of sensors depends upon the vital signs of the human body to be monitored and is generally proportional to the complexity of the monitoring system.…”

“…d. KNN-hyperparameters: "n neighbors ": list (range (1, 31)), "weights": ["uniform," "distance"], "metric": ["euclidean," "manhattan"]. e. XGboost-Hyperparameters: "n estimators ": [50, 100, 200], "learning rate ": [0.001, 0.01, 0.1, 1], "max depth ":[3,4,5], "subsample": [0.5, 0.7, 1.0], "colsample bytree ": [0.5, 0.7, 1.0]. f. AdaBoost-Hyperparameters: "n estimators ": [50, 100, 200], "learning rate ": [0.001, 0.01, 0.1, 1], "base estimator ": [DecisionTreeClassifier(max depth = d) for d in[1,2,3]].…”

Towards an Accurate Faults Detection Approach in Internet of Medical Things Using Advanced Machine Learning Techniques

“…In [ 13 ], a link failure is detected before its occurrence using probability computation. The authors of [ 14 ] proposed a system which consists of two stages to detect fault; in the first stage they used a Pearson correlation coefficient to determine the physiological parameters strongly correlated to the actual value, and in the second stage they use statistical measures such as means and standard deviations, to obtain useful information about the capability of the actual sensor. A Markov Chain Model is also used [ 10 ] where a confusion matrix with nine states was determined.…”