Devotion towards own body is one of the important factor considered in this era. The electrocardiogram (ECG)signal is an electrical activity of the human heart. It is used for diagnosing many cardiac diseases. Theequipment’s which provide results at run time and also accuracy maintained. To detect any falling event, aswell as the relative electrocardiogram (ECG) signal of the user, a multi-thread method is proposed with thehelp of new technology of Raspberry Pi and classification methods. Wireless communication is done throughWi-Fi or bluetooth which provides flexibility and extendibility in embedded devices. The goal of this project isthe classification of an ECG signal into normal and abnormal classes to detect type of arrhythmia and toachieve this Artificial Neural Network (ANN) based cardiac arrhythmia disease diagnosis is used. ECG signalclassification done by EMB (empirical mode decomposition) method for accurate ECG signal using MATLAB.The classification performance is evaluated using measures such as mean squared error (MSE), classificationspecificity, sensitivity, accuracy and precision. The classifier achieves the maximum accuracy of 100%.