Imposing patient data privacy in wireless medical sensor networks through homomorphic cryptosystems

Abstract

In recent years, healthcare applications are regarded as hopeful fields for wireless sensor networks, where the patients have been monitored in hospitals and at their accommodating places. These wireless medical sensor networks are more endangered to intruding, modification, eavesdropping, impersonation and playback attacks than the wired networks. A lot of solutions have been developed to secure wireless medical sensor networks. The already available solutions can safeguard the patient data during communication, but unable to terminate the inside attack where the proprietor of the patient database discloses the delicate patient information. In this project, an empirical approach has been presented to safeguard the inside attack by utilizing multiple number of data servers to store patient data for further utility. The main contribution of this work is to distribute the patient information securely in multiple data servers and by employing the homomorphic cryptosystems (modified paillier/elgamal) to accomplish patients' data privacy. This is a secure method against both outside and inside attacks as long as all the data servers are not compromised.