Multiplex communications – Diagnostic testing
Reexamination Certificate
2001-02-15
2004-12-21
Kizou, Hassan (Department: 2662)
Multiplex communications
Diagnostic testing
C370S503000
Reexamination Certificate
active
06834040
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to communication systems such as telephone systems and, more particularly, to performance measurements, as for example voice clarity measurements, in such systems, and even more particularly to measurement synchronization between two stations in such systems, and specifically in voice over packet systems.
BACKGROUND OF THE INVENTION
Telephone companies have expended great efforts over many years to improve the quality of the voice communications that they have traditionally carried. Telephone systems operated by these companies are often referred to as public switched telephone network (PSTNs). While not perfect, voice quality in modem telephone systems has been improved by optimizing various system components for the dynamic range of the human voice and the rhythms of human conversation to the point that they can provide high quality service. High quality voice traffic does not require a large bandwidth but does require timely transmission.
Unlike PSTNs, however, networks which transmit data in discrete packets, such as those that use the Internet Protocol (IP), were developed to support non-real-time applications, such as file transfers and e-mail. These applications feature communication traffic that is bursty and typically requires much higher bandwidths than voice traffic does but is not as sensitive to delays and delay variations as PSTNs are. In addition, such network applications can compensate for packet loss by re-transmitting any lost packets, and the reception of data packets out of order does not present significant problems in data reconstruction.
Recent developments in communication systems have resulted in combining the traffic historically carried separately by telephone and data networks. The service provided by such systems is referred to as Voice over Packet (VoP). The more popular of VoP systems utilize the Internet Protocol (IP) and are commonly referred to as Voice over IP (VoIP) systems. VoP technologies have made maintaining voice quality at high levels more complex by compressing the voice signal and transmitting it in discrete packets. With voice traffic there is the need for timely packet delivery, often in networks that were not originally designed for these conditions. Transmission conditions that pose little threat to non-real-time data traffic can introduce severe problems to real-time packetized voice traffic. These conditions include real-time message delivery, gateway processes, packet loss, packet delay, and the utilization of nonlinear codecs.
Newer PSTN networks use digital-voice transmission for greater efficiency in their backbones. Digitizing analog voice signals often affects voice clarity. The VoIP gateway interconnects the PSTN with the IP network using voice and signaling schemes.
Voice quality as perceived by the user is subjective, but typically his perception of quality includes three key parameters: (1) signal clarity, (2) transmission delays, and (3) signal echos. While the impact on the user is subjective in nature, objective measurement techniques for each of these parameters has been developed. The clarity of a voice signal is generally described by how accurately the received signal reproduces that which was sent. Signal fidelity, lack of distortion, and intelligibility are key elements in the description of its clarity. Delay is the time that it takes to transmit a voice signal from the speaker to the listener. And, echo is the sound of the speaker's voice that he hears returning to him. Delay and echo can be annoyances and distractions to the user. Any delays in transmission and any echos should be imperceptible to him. A lack of clarity can also degrade the ability of the user to obtain information from the interchange and heighten the level of his frustration.
Packet loss is not uncommon in IP networks. As the network, or even some of its links become congested, router buffers fill and start to drop packets. Another cause of packet loss is route changes due to inoperative network links. An effect similar to packet loss occurs when a packet experiences a large delay in the network and arrives too late for use in reconstructing the voice signal. In the case of real-time voice information, packets must arrive within a relatively narrow time window to be useful in reconstructing the voice signal. Re-transmissions in the case of voice may add extensive delay to the reconstruction and cause clipping, or unintelligible speech.
Voice transmission in a VoP system are coded and decoded via a codec. A speech codec is a device which transforms analog voice into digital bit streams and vice versa. The term codec is a shortened form of coder/decoder. Some speech codecs also use compression techniques which remove less important parts of the signal in order to reduce the bandwidth required for the transmission. In other words, many codecs compress voice signals by preserving only those parts of the voice signal that are perceptually important.
The signal can experience delays from the time it takes for the system or network to digitize, form data packets, transmit, route, and buffer a voice signal. These delays can interfere with normal conversations.
Since users have become accustomed to PSTN levels of voice quality and compare the voice quality of other services to that typically obtained from a PSTN, for VoP services to be acceptable they must maintain this level of quality. Voice quality is now an important differentiating factor for VoP (voice-over-packet) networks and equipment. Consequently, measuring voice quality in a relatively inexpensive, reliable, and objective way has become very important.
One industry standard, objective method for measuring clarity in VoP networks is the perceptual speech-quality measurement (PSQM). PSQM evaluates the quality of voice signals in the same way that codecs encode and decode voice signals. PSQM evaluates whether a voice signal is distorted enough for a human to find it annoying and distracting. It compares a clean voice sample with a distorted version using a complex weighting method that takes into account perceptually important elements, such as the physiology of the human ear and cognitive factors related to what human listeners are likely to notice. PSQM uses an algorithm to provide a relative score that indicates just how different the distorted signal is from the original from the human listener's perspective. This distortion score corresponds closely to how a statistically large number of human listeners would react in the same test situation using.
Another important method for measuring perceived clarity is the PAMS (perceptual analysis-measurement system). PAMS uses a perceptual model similar to that of PSQM and provides a repeatable, objective means of measuring perceived voice quality. PAMS uses a different but effective signal-processing model and produces different types of scores.
One difficulty in performing either the PSQM or PAMS test is the synchronization of the original and received messages. Typically the user must press the Start button on the receiving station prior to pressing the start button on the transmitting station. The receiving station then must record for a period of time longer than that of the message that was sent. During the analysis phase which follows the recording phase the recorded message is compared to that of the original message. To obtain a meaningful measurement these two signals must be correlated in time, i.e., the recorded file must be scanned to locate the PSQM/PAMS signal. This correlation can be very expensive in terms of computational resources consumed. In addition, the requirement of activating the recording by the receiving station prior to that of the transmitting station makes automatic measurements difficult.
The disadvantages of this synchronization scheme include (1) it puts burden on the user to synchronize test, (2) it cannot realistically be scheduled to run at preselected times, (3) PSQM signal correlation will be very slow, perhaps taking on the orde
Agilent Technologie,s Inc.
Cho Hong
Kizou Hassan
LandOfFree
Measurement synchronization method for voice over packet... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Measurement synchronization method for voice over packet..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Measurement synchronization method for voice over packet... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3324215