Performance Analysis of Secure Real-time Transport Protocol Using Elliptic Curves

Abstract

The growing popularity of streaming audio/video is enhancing the popularity of real-time media applications and services. People use real-time media applications for voice/video calls as well. Real-time Transport Protocol is a well-known protocol solely used for real-time data transmission, which uses the Voice over IP technology. Nevertheless, like other Internet applications, the security aspect has become a significant issue for real-time media applications. To protect users’ privacy, real-time data such as streaming video or audio/video calls can be encrypted during the conversation. Some real-time applications use symmetric key encryption to protect the media data. However, this cannot ensure the security of real-time media. An asymmetric key encryption technique can be a better choice in this context. Among other asymmetric key encryption schemes, Elliptic Curve Cryptography (ECC) performs better due to its comparatively smaller key size and lesser computation time. We must choose the elliptic curves judiciously to encrypt the voice call, because audio encryption may increase end-to-end delay, which worsens the quality of the call. The appropriate elliptic curves can be chosen from a set of elliptic curves by analyzing their performance on real-time audio encryption. In this work, we have implemented an ECC based encryption technique in a softphone to encrypt real-time voice calls. We have used 15 elliptic curves to measure the performance of audio calls, and based on the result, have proposed some suitable elliptic curves for real-time audio encryption.