Browsing by Author "El-Hennawey, Samy"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Echo control in VoIP
    (Canadian Acoustics, 2007) El-Hennawey, Samy
    The challenges faced in echo control in voice over the Internet Protocol (VoIP) are discussed. There are primarily two classes of echo in telephone conversation. The first class is due to the use of a 2-to-4 wire conversion using hybrid circuits. Echo control of this class is achieved by a line echo canceller (LEC). The other class is due to the use of handsfree mechanisms where there is significant acoustical feedback coupling between the receiver speaker and the receiver microphone. Echo control of this class is achieved using an acoustic echo canceller (AEC). The design of double-talk detectors is a challenge for VoIP gateway design. Another challenge seen in deploying VoIP with voice gateways to interface with PSTN is that some PSTN local loops have strange characteristics. It is highly desirable to measure echo at the end user of an IP phone with multi-vendor networks. An important impairment, that is to be measured, is echo with its echo path loss and its associated round-trip delay.
  • Thumbnail Image
    Item
    Embedded Real-Time WDiCB Quality Analysis System
    (ntl. Engineering Consortiu, 2006) El-Hennawey, Samy
    Voice quality is essential in any communication systems when speech is transmitted from one end to another passing through several impairments. New installation of voice over Internet protocol (VoIP) systems has been increasingly popular in the past few years and will continue to spread both in the carrier and enterprise sectors. In addition to impairments introduced in the public switched telephone network (PSTN), VoIP systems include additional degrading factors such as latency, delay jitter, and packet loss. In order to provide good quality of service (QoS), a QoS monitoring module is sought. This mandates the exis-tence of an embedded subsystem that assesses the voice quality in each live call. This is the main concern of this paper. The existing assessment techniques either lack being nonintrusive for application in live calls or lack the assess-ment accuracy by either probing on the packet network sta-tistics only or relying on the received voice samples. The idea presented in this paper overcomes those shortcomings