Anomaly detection in medical WSNs using enclosing ellipse and chi-square distance

Abstract

In this paper, we propose an Anomaly Detection (AD) approach for medical Wireless Sensor Networks (WSNs). This approach is able to detect abnormal changes and to cope with unreliable or maliciously injected measurements in the network, without prior knowledge of anomalous events or normal data pattern. The main objective is to reduce the false alarms triggered by abnormal measurements. In our proposed framework, each sensor applies the Exponentially Weighted Moving Average (EWMA) for one-step forecasting. To reduce the energy consumed by periodic data transmission to the Local Processing Unit (LPU), the sensor transmits only when the data point (measured, expected) falls outside the dynamically updated ellipsoidal region enclosing the normal data. The LPU exploits correlation and uses chi-square distance for spatial analysis before raising a medical alarm. We evaluate our approach on real medical data set. Experimental results through computer simulation demonstrate that our proposed approach can achieve a good detection accuracy with low false alarm rate (less than 4%).

AUDIT: AnomaloUs data Detection and Isolation approach for mobile healThcare systems

“…Salem, Liu, and Mehaoua () presented a distributed anomaly detection approach for medical wireless body networks. Each wearable and implantable sensor performs local point anomaly detection based on the following steps: (a) one‐step forecast of the current sensor value using the exponentially weighted moving average (EWMA) and (b) compare the forecasted value with the measured value to detect deviation from normal behaviour.…”

Layer-Adaptive Communication and Collaborative Transformed-Domain Representations to Optimize Performance in Next-Generation WSNs

“…To optimize resources, a sensible decision is to deploy an equal percentage of nodes over different regions to ensure minimization of coverage holes, and elongation of network lifetime. Therefore, in this scenario, we propose to deploy 20% of the nodes in region R 1 and the rest 80% of the nodes to be distributed evenly over R 2,3,..., 8,9 regions as shown in Fig. 3(a).…”

“…Following the deployment of nodes and prior network initialization, the election of CHs is carried out in all R 2,3,..., 8,9 regions. Since the use of CHs in clustering techniques plays an important role to improve network lifespan, effective criterion for CHs election is equally necessary for further improving performance of the network.…”

“…The development of such networks demand extensive planning strategy along with superior tactical approaches for its working capabilities. This effective development motivates the existence of many real-time application scenarios such as security monitoring [2], environmental control [3], under-water quality measurement [4], battlefield surveillance [5], target detection [6], remote environmental monitoring [7], medical and healthcare [8], an amazingly increasing interest toward application like refineries, petrochemicals, underwater development facilities, and oil and gas platforms [9], and many more.…”

Performance Optimization in IoT-Based Next-Generation Wireless Sensor Networks

“…Comparison of Denoising Image Data Samples in Terms of PSNR/SSIM Denoising 256 × 256 grayscale Cameraman standard test data images over noise σ =[5,10,15,20,25, 50, 100] when received at a node µα. Each row represent an original image, a noisy image, a partially denoised, and a fully denoised image, respectively, corrupted by a specific level of additive white Gaussian noise (AWGN).…”