Integrated network selection scheme for remote healthcare systems

The quality of service parameters (QoS) needed for remote patient monitoring are generally known to the users in linguistic terms e.g. low, medium, high etc. However, the real time values observed for available networks are crisp. This paper presents an integrated approach for network selection in heterogeneous network for m-health applications by obtaining QoS requirements in linguistic terms from clinical parameters in crisp format. We propose to generate a Trust Matrix which provides an alternate approach to writing large number of fuzzy rules to draw fuzzy inferences. Trust Value obtained from each row of Trust Matrix is compared with Desired Trust Value calculated as per level of critical nature of the patient/disease. Most suitable network is selected to transmit vital information of the patient to the medical practitioner. The proposed approach is simulated for different patients at different levels of critical nature of a disease, taking diabetes as a clinical case study. The results show that based on the level of disease, network selected is different for different levels.