The disturbed electrical activity of the heart which leads to Atrial Fibrillation (AF) poses a threat to the health and a lot of people in the country are suffering because of this. Its timely detection and treatment can increase the longevity of patients with AF. The proposed, Smartphone based inventive method of detection of AF utilizes Photoplythosmography (PPG) signal obtained from the fingertip using the camera in a smartphone. It differentiates Normal sinus rhythm (NSR) from AF, Premature Ventricular contraction (PVC) and Premature Atrial Contractions (PAC), with the combination of statistical methods like Root mean square of successive differences (RMSSD), in combination with Shannon Entropy (ShE) and Turning point ratio (TPR) method. The reformulated TPR is used along with pulse rise and fall times to increase the accuracy of detection. The acquired results show that this method can detect NSR with sensitivity of 0.9155, and gives overall accuracy of classification of 92% © 2018 IEEE.