Adaptive VoIP with Audio Watermarking for Improved Call Quality and Security

In this paper we describe a novel adaptive method of speech quality control which may be used to adjust three call parameters: speech codec configuration, playout buffer size, and amount of FEC (Forward Error Correction) mechanism information during VoIP (Voice over Internet Protocol) call under changing network conditions. This solution differs from previously proposed because it utilizes audio watermarking techniques as a communication channel between calling parties to send information about the change in quality of the call. Assessment of the call quality is performed with non-intrusive objective methods like E-Model [9, 10] and expressed in MOS scale. Moreover we show how to use the proposed speech control mechanism in conjunction with adaptation of security measures applied for VoIP traffic. Thus, we gain a tradeoff between the quality and security of the call which is very a important and still unsolved issue for IP telephony.