Browse Prior Art Database

Method for packet loss resilient tone detection

IP.com Disclosure Number: IPCOM000017141D
Publication Date: 2003-Jul-22
Document File: 3 page(s) / 72K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method for packet loss resilient tone detection. Benefits include improved functionality and an improved user experience.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Method for packet loss resilient tone detection

Disclosed is a method for packet loss resilient tone detection. Benefits include improved functionality and an improved user experience.

Background

              Tone detection is a critical component in telecommunication systems for call initiation, control, and termination purposes. Many different tones have defined industry-standard roles for various types of communications, including voice call, fax transmission, and data access. With the advent of the Internet, many conventional communication services have gradually migrated to the IP-based networks. New communication services running over IP have appeared. However, because of the fundamental difference of the Internet from conventional circuit networks, certain components used in communications are no longer adequate to provide good performance. One such component is the tone detection mechanism popularly used in converged communications networks.

              Packet networks are not reliable because packets that carry media or tones are subject to packet loss, delay, or jitter. A jitter-buffering algorithm can compensate for jitter. Delay can make a packet too late to be useful and it must be dropped. Lost packets cannot be recovered. When a packet is lost, an algorithm on the receiver side typically compensates for the effect of packet loss but does not completely recover it. In addition, packet loss compensation does not work for packets that carry tones. A packet-loss-compensated tone is not recognizable by a tone detector due to severe frequency distortion.

              A tone transmitted over the IP network may exhibit cadence distortion because portions of it are cut off due to packet loss. As a result, a single tone may be detected as two or more tones, causing communication failure for the user. 

              The loss of a single packet is the most common packet-loss occurrence, though more than one packet can be lost. Statistically, the probability of more than one packet getting lost is extremely low, if the statistical distribution is Gaussian. Therefore, this disclosure describes the single packet-loss case.

              When a single packet is lost in a packet stream, an empty gap is created in the stream packet time interval. In IP telephony, packet time intervals are typically 10 ms to 30 ms. A gap of 10 ms to 30 ms has the effect of a single packet loss in the transporting of tones.

              All tones in telephony communications...