Voice quality optimization on multi-codec calls

Details Voice quality optimization on multi-codec calls Voice quality optimization on multi-codec calls

1998-10-03

2003-07-29

Vincent, David (Department: 2661)

Multiplex communications

Pathfinding or routing

Through a circuit switch

C370S255000, C370S465000

Reexamination Certificate

active

06600740

ABSTRACT:

FIELD OF THE INVENTION
The present invention is generally related to communication networks including wireless telephony communication networks, communicating voice calls between an originating network and a terminating network, and more particularly to a communication network having dissimilar compression and decompression equipment, such as codecs, in the speech path.
BACKGROUND OF THE INVENTION
Communications networks, including wireless communication networks, typically include an originating network, a terminating network, and a communication link exchanging voice and data between these networks. In the case of telephony networks, analog speech signals are typically digitized through digital sampling prior to transmission over the communication link and then converted back to analog at the terminating network. To increase the capacity of the communication network, these digitized voice calls routed over the communication link are typically compressed through the use of compression and decompression equipment, commonly referred to as a codec or as a vocoder. Typically, a codec resides at both the originating end and at the terminating end of a call, whereby the digitized voice is compressed by an encoding algorithm in a forward direction, and decompressed by a decoding algorithm at the receiving end. The decompressed voice signal is ultimately converted back to voice through the use of a digital to analog (D/A) converter. The compressed digitized voice signals are typically routed over a communication link, such as a public switched telephone network (PSTN) in a pulse code modulated (PCM) format, typically at 64 kbps.
The repeated use of compression and decompression equipment (codecs) in a speech path yields poor speech quality. In particular, the use of different voice codecs in different networks exacerbates the problem. As, the speech compression becomes more widespread in communication networks e.g., through the expanding use of cellular networks and “voice over the Internet,” this voice degradation problem becomes more troublesome.
Current codecs are based on conversion from the 64 kbps PCM encoding used in the PSTN and back again, and do not consider the previous or subsequent use of other codecs in the network handling a voice call. However, in many instances, information about the originating network, and the codec used at the originating network, is available at the transit network and at the terminating end. In addition, information about the terminating network, and the terminating network codec, is available to the originating end.
There is desired an improved communication network and method of transmitting voice calls across the network having multiple codecs which improves the quality of voice calls over the communication network.
SUMMARY OF THE INVENTION
The present invention achieves technical advantages as a communication network having multiple codecs whereby the originating network and the terminating network provide information of the resident codec to each other, and codec encoding and decoding algorithms are responsively altered to improve voice quality. In one embodiment, the originating network provides codec information indicating the encoding algorithm to the terminating network, and the terminating network alters the codec decoding algorithm to better match the encoding algorithm and improve voice quality. In a second embodiment of the invention, the terminating network provides information about the decoding algorithm to the originating network, whereby the originating network alters the codec encoding algorithm to better match the decoding algorithm and improve voice quality. In both embodiments, information about one network codec algorithm is provided to the other network to allow one network to adjust and match its codec algorithm to the other to improve voice quality.
The first embodiment of the present invention comprises a communication system comprising an originating network. The originating network comprises a transmitter generating an electrical signal representative of speech. The originating network further comprises an originating voice codec coupled to the transmitter encoding the electrical signal according to an encoding algorithm. A signal device is coupled to the transmitter and generates a codec signal link indicative of the encoding algorithm utilized by the originating voice codec.
The communication system further comprises a communication link coupled to the originating network, and a terminating network comprising a receiver and a terminating voice codec coupled between the communication link and the receiver decoding the received encoded electrical signal. The terminating voice codec has a decoding algorithm. The terminating network further comprises a processing device coupled to the communications link identifying the encoding algorithm as a function of the received codec signal, and responsively alters the decoding algorithm as a function of the codec signal. The terminating voice codec has a decoding algorithm. The terminating network further comprises a processing device coupled to the communications link identifying the encoding algorithm as function of the received codec signal, and responsively alters the decoding algorithm as a function of the codec signal. The terminating voice codec recreates the original encoded electrical signal as a function of the identified encoding algorithm. The processing device analyzes the encoded electrical signal as a function of the identified encoding algorithm. The processing device examines the digital speech stream and looks for encoding artifacts including values present that are useful in the enhanced decoding process to identify encoding parameters to recreate the original encoded electrical signal. The signal device sends a start/sync flag to the terminating voice codec to indicate the start of an encoding period, which is typically 20 milliseconds in GSM networks. Preferably, the originating voice codec encodes the electrical signal in PCM format such as 64 Kbps PCM. The communication link preferably comprises a public switched telephone network (PSTN).
According to the second embodiment to the present invention, a communication system communicates an encoded electrical signal representative of speech and comprises a terminating network. The terminating network comprises a receiver, and a terminating voice codec coupled to the receiver. The terminating voice codec decodes the encoded electrical signal according to a decoding algorithm. The terminating network further includes a signaling device coupled to the terminating voice codec and generates a codec signal indicative of the decoding algorithm. The communication system further comprises a communications link and an originating network including a transmitter generating the electrical signal. An originating voice codec is coupled between the transmitter and the communications link and generates the encoded electrical signal according to an encoding algorithm. The originating network further includes a processing device coupled to the communications link and identifies the decoding algorithm as a function of the codec signal, and responsively alters the encoding algorithm as a function of the codec signal. The processing device may send a start/sync flag to the terminating voice codec to indicate the start of an encoding period. The communication link preferably comprises a PSTN.
According to a method of a third embodiment of the present invention, the method comprises communicating an encoded signal representative of speech across a communication link between an originating network including an encoder having an encoder algorithm, and a terminating network including a decoder having a decoder algorithm. The method comprises the steps of the originating network sending a codec signal to the terminating network indicative of the encoding algorithm. The terminating network receives the codec signal and identifies the encoding algorithm of the function of the received codec signal. The terminating network then modifies

Affiliated with

Also associated with

Rating

Voice quality optimization on multi-codec calls does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Voice quality optimization on multi-codec calls, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Voice quality optimization on multi-codec calls will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3036574