Multiplex communications – Duplex – Transmit/receive interaction control
Reexamination Certificate
2001-02-09
2004-12-14
Marcelo, Melvin (Department: 2663)
Multiplex communications
Duplex
Transmit/receive interaction control
C370S290000, C379S406050, C379S406080
Reexamination Certificate
active
06831900
ABSTRACT:
TECHNICAL FIELD OF THE INVENTION
The present invention is directed, in general, to digital signal processing and, more specifically, to a filter cell and an echo canceling system employing the same.
BACKGROUND OF THE INVENTION
The existing public switched telephone network represents a significant capital investment that has taken place in great part over the last 80 years. The public switched telephone network was originally designed for voice services (so-called plain old telephone service) and was entirely analog.
Originally, the public switched telephone network included “local loops,” which connected homes and businesses to central office switches. This allowed anyone having a telephone connected to the central office switch to call one another. A given central office typically only covers a relatively small area.
To allow people to call one another from greater distances, central office switches were interconnected by analog trunks. Unfortunately, signal quality suffered greatly as distances increased. Filters and amplifiers improved quality, but only to a limited extent.
Over time, however, the analog trunks (that may be thought of as forming the “backbone” of the public switched telephone network) were replaced with land-based microwave, satellite and optical fiber links. Public switched telephone network signals (“traffic”) were digitized for transmission over the backbone, significantly improving signal quality, service and reliability.
To maximize utilization of the backbone, an assumption was made that, at the time, seemed straightforward. The assumption was based on the observation that public switched telephone network traffic consisted of human speech, which by its nature occupies frequencies below 4 kilohertz (kHz).
Thus, it was decided that higher frequencies were of limited value and represented a waste of bandwidth if the traffic were to be digitized and transmitted. The higher frequencies were, as a result, discarded when signals were initially digitized. The net effect was that more conversations were carried over a given microwave, satellite or fiber link.
While truncating the frequencies above 4 kHz was of no consequence to the transmission of speech, the same proved not to be true for data. In the quest for speed, computer modems have attempted to use as much bandwidth as possible, and in the most clever manner. Unfortunately, even in view of the most clever modems, the 4 kHz digitization cutoff has imposed an apparent limit on the speed of such devices. Unfortunately, the analog local loops have unjustly taken most of the blame for the speed limitation.
Digital subscriber line (DSL), developed over the past few years, presents a novel solution to the speed limitation conundrum. According to DSL, local loops are employed to carry speech in a stream at normal frequencies (exclusively below 4 kHz). The local loops, however, are also called upon to carry data in a stream at frequencies exclusively above 4 kHz. DSL termination circuits located at the home or business and the central office combine and separate the voice and data streams as they enter and leave the local loop. Once separated at the central office, the voice stream is digitized for relay over the public switched telephone network backbone as before, and by employing the existing infrastructure. The data stream, however, is sent through the public switched telephone network or another network (such as the Internet via a different path), without imposition of the 4 kHz artificial bandwidth limits.
One form of DSL, Asymmetrical DSL (ADSL) was designed with the Internet particularly in mind and accordingly emphasizes downloading of data over uploading of data (which is the nature of Internet “surfing”). ADSL uses the frequency spectrum between 0-4 kHz for the plain old telephone service stream and 4 kHz to 2.2 MHZ for the data stream. Depending on the design, length and conditions of the local loop, ADSL can offer speeds up to 9 Mbits/s (Mbps) for downstream (network to user) and up to 800 Kbps for upstream (user to network) communication.
Another form of DSL, High-Bit Rate DSL (HDSL) is a technology extension of DSL. HDSL is a symmetric transport medium, meaning that it provides 1.544 Mbps transmission speed both downstream and upstream over distances as far as 12,000 feet, without repeaters. Because about 20% of loops are longer than 12,000 feet, the industry has developed a repeater for HDSL that effectively doubles the span's reach to 24,000 feet. HDSL is based on a technology called adaptive equalization, which digitally analyzes and then compensates for distortion, imperfections in the copper line itself as well as adverse environmental conditions, throughout the transmission process. Furthermore, HDSL transmits full-duplex signals on each pair of wire and uses echo cancellation to separate the receive signals from the transmit signals.
To enhance and build on that inherent symmetry, standards bodies are now working on HDSL's next generation, called HDSL2. HDSL2 promises full-duplex T-Carrier 1 (T1) payload over one copper loop, while still delivering the same time and cost efficiencies for T1 deployment that HDSL offers. Carriers everywhere are running out of copper in their local loop plants. One of HDSL2's key benefits will focus squarely on and alleviate that concern. Essentially, the technology will double the number of available T1 lines because it requires only a single copper pair, compared with the two pairs required by the standard HDSL. As a result, HDSL2 may replace standard HDSL for most T1 deployments in the future, although HDSL will remain an option in those cases in which there may still be some engineering reasons for deploying a two-loop solution. One example is with long loops in excess of 12,000 feet, where span-powered HDSL repeaters may still be necessary. Beyond just workhorse T1 deployment, HDSL2 also should prove to be a viable competitive technology for Internet access applications that require symmetrical data delivery.
HDSL2, therefore, further enhances the noteworthy advantages associated with DSL. First, DSL-based technology does not require local loops to be replaced. Second, DSL-based technology overcomes the 4 kHz digitization barrier without requiring changes to existing public switched telephone network voice-handling equipment. Third, DSL-based technology requires relatively little equipment to combine and later separate speech and data streams. Fourth, DSL-based technology allows speech and data to occur simultaneously over the same local loop. HDSL2 now promises full-duplex T1 payload over one copper loop, while still delivering the same time and cost efficiencies for T1 deployment that its predecessor, HDSL, offers.
Some technical challenges, however, remain for HDSL2. One is designing a transceiver that can accommodate the full-duplex T1 payload in conjunction with the standard defined by American National Standards Institute (ANSI) committee T1E1.4 (June 1995), which is incorporated herein by reference. In conjunction therewith, echo cancellation techniques may be employed in the digital signal processing portion of the transceiver to reduce the magnitude of an echo associated with the information transmitted over the HDSL2 network.
For a DSL-based system, an echo of the transmit signal usually occurs at the point where the transmit path and the receive path are joined together. A hybrid circuit, typically used to connect these two paths to the twisted-pair telephone line, generates the echo which propagates down the receive path along with any receive signal from the far end. An echo may easily be 15 decibels (dB) greater than a strong receive signal thereby making processing of the receive signal at least more difficult. As the receive signal becomes weaker, acceptable processing of the receive signal becomes even more difficult due to the overriding effects of the echo. Therefore, echo cancellation is usually employed in a transceiver employed in the DSL-based system.
Digital adaptive echo cancellers are located between the transm
Agere Systems Inc.
Marcelo Melvin
LandOfFree
Filter cell, method of deriving an echo component and an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Filter cell, method of deriving an echo component and an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Filter cell, method of deriving an echo component and an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3326567