Background: In the recent times, e-health applications have been trending for providing timely and effective services to the patients. Remote patient monitoring technology plays a significant role in managing the health of patients through computerized systems and tiny devices known as sensors. These wearable sensor devices can measure the patient’s physiological data and provide personal care to patients when integrated with remote monitoring system. Precise information about patients’ physiological parameters along with their activities and location is highly beneficial for determination of patient’s exact health condition. Sometimes, non-repeated but un-necessary data is transmitted which shows the same status of patient’s health condition for extended period. This redundant data transmission only increases processing overhead, transmission time, energy consumption and leads to wastage of memory. Method: The multi-stage context-aware data filtering technique proposed in this work aims to reduce the overall volume of body sensor data that needs to be transmitted which helps in significantly reducing aforementioned overheads. The proposed data filtering technique is divided into three stages, where first stage is context based range determination, wherein the Normal Range (NR) is modified based on geographical and environmental context to yield Context Based Range (R). In the second stage, deviation of the data is calculated from the Context Based Range (R) acquired from previous stage. The third and final stage carries out selective transmission of data with priority to the sensor values where the deviation is maximum from the context based normal range. The performance of proposed filtering technique has been measured using Castalia Framework. Results: The simulation carried out using OMNET++ based Castalia framework shows that the proposed context based data filtering algorithm results in reduction of 14.5% in the volume of data to be transmitted to the health server. Conclusion: The proposed algorithm filters out the false positives in the sensed values of physiological parameters thereby achieving significant reduction in the volume of data to be transmitted.
Volume 11 | 07-Special Issue
Pages: 943-953