Miscellaneous active electrical nonlinear devices – circuits – and – Specific identifiable device – circuit – or system – Unwanted signal suppression
Reexamination Certificate
2000-06-23
2003-06-24
Le, Dinh T. (Department: 2816)
Miscellaneous active electrical nonlinear devices, circuits, and
Specific identifiable device, circuit, or system
Unwanted signal suppression
C327S558000
Reexamination Certificate
active
06583662
ABSTRACT:
FIELD OF THE INVENTION
The present invention generally relates to high speed data communications wherein signal information is processed both in digital and analog forms. More specifically, the invention is related to an adaptable smoothing filter, which solves problems associated with integration density, power consumption, and the need for an external non-integrated transmit filter and a buffer for asymmetric digital subscriber line circuit cards.
BACKGROUND OF THE INVENTION
With the advancement of technology, and the need for instantaneous information, the ability to transfer digital information from one location to another, such as from a central office (CO) to a customer premise (CP), has become more and more important.
In a digital subscriber line (DSL) system, data is transmitted from a CO to a CP via a transmission line, such as a two-wire pair, and is transmitted from the CP to the CO as well, either simultaneously or in different communication sessions. The same transmission line might be utilized for data transfer by both sites or the transmission to and from the CO might occur on two separate lines. In its most general configuration, a DSL circuit card at a CO is comprised of a digital signal processor (DSP) which receives information from a data source and sends information to an analog front-end (AFE). The AFE serves as the interface between an analog line, such as the two-wire pair, and the DSP. The AFE converts digital data, from the DSP, into a continuous-time analog signal when processing data from the CO to the CP. The AFE also converts a received continuous-time analog signal into a digital data stream when processing data that originates at the CP that is sent to the CO. A digital to analog converter (DAC) may be followed in series by a clocked (sampled-data) analog switched-capacitor (SC) filter. Residual high-frequency energy, either spectral images resulting from digital sampling or high-frequency content from other sources not removed by the SC filter, may have to be removed as necessary to comply with output signal standards. More specifically, the residual high-frequency energy should be removed in order to meet or exceed the prescribed spectral energy mask of the particular communication standard.
A smoothing filter typically receives an input signal comprising a number of discrete magnitudes that remain relatively constant over each DAC clock period and corresponding to the resolution of the DAC. The DAC output signal may be filtered by a SC filter before being forwarded to the input of the smoothing filter. Since the output of the SC filter may comprise a clocked or sampled signal, the smoothing filter may be used to complete the digital to analog signal conversion process by creating a more analog like continuous signal.
Continuous-time smoothing filters are traditionally implemented as low-order filters, which typically seek to remove spectral images due to conversion from sampled continuous-time signals. The task of transmit signal filtering is usually left to off-chip filters. Furthermore, integrated continuous-time smoothing filters traditionally do not seek to drive off-chip loads.
For the continuous-time smoothing filter to be integrated, it has to make use of available on-chip components, typically, amplifiers, resistors, and capacitors. These integrated components typically have relatively large manufacturing tolerances. As a result, an integrated circuit that relies on the resistance and capacitance values of the various components of the circuit, such as a smoothing filter, must be over designed in order to function correctly or the circuit must be adjustable to compensate for the integrated component manufacturing variations. The precision of the adjustment or compensation depends on the filter performance requirements. The output of a continuous-time smoothing filter can be brought under a prescribed spectral mask using a number of various filter types. However, if power dissipation and hence the number of operational amplifiers, and filter accuracy are issues for the overall AFE, the selection of an appropriate filter type becomes more difficult.
For example, a 3
rd
order Chebyshev filter would suffice, but it would require rather accurate adjustment in order to realize appropriate cut-off frequencies for CO-xDSL transmit applications. A 4
th
order filter reduces the accuracy requirements to 3-bits, but the pole Q of the filter becomes too high for a single operational amplifier filter to meet the spectral mask required for a xDSL transmit signal. If a 6
th
order filter was used, the smoothing filter would not need to be tuned at all. However, such a filter would require no less than 5 operational amplifiers (resulting in a significant power requirement) to realize the relatively high pole-Q values needed to meet the required transfer function.
The overall functionality of an AFE is particular to the specific DSL application, wherein factors such as signal bandwidth, data rate, data reach, signal quality, power budget, and different applicable standards combine to determine the optimum AFE. In order to minimize application specific implementations of AFEs across the many DSL applications, it is desired to create a high-performance configurable AFE. Considering the many flavors of asymmetric DSL (ADSL) applications, the adaptability problem becomes more apparent. Focusing on ADSL applications, there are a number of different implementation standards including: discrete multi-tone (DMT) T1.413, discrete multi-tone-frequency division multiplexed (DMT-FDM), discrete multi-tone-echo cancelled (DMT- EC), discrete multi-tone-G.lite (DMT-G.lite), carrierless amplitude/phase—rate adaptive digital subscriber line (CAP-RADSL), and discrete multi-tone over integrated services digital network (DMT-ISDN). Hereinafter, the aforementioned ADSL applications will be denoted xDSL. For each xDSL application, the optimum AFE configuration varies. Subsequently, the overall AFE configuration and the configuration of an anti-aliasing filter within the AFE may vary accordingly.
In CO-DSL modem applications, multiple DSL transceivers may be co-located within the same equipment or even located on the same printed circuit board. Competitive local-exchange carriers (CLECS) often rent equipment space from the various local telephone companies on a volume basis. As a result, DSL transceiver density and power efficiency are important factors for CLECs to consider when entering local DSL service markets. Transceiver density and power efficiency are important to the various telephone companies as well, as higher transceiver density and reduced power requirements directly reduce overhead and operating costs, respectively, for the CO operators.
Transceiver density is a function of the minimum necessary component area to construct the various circuits of the transceiver and the required transmit power of each transceiver. DSPs and the various other components within the transceiver need power in order to function. The desired movement of electrons to, from, and through the various components encounters resistance, which leads to the dissipation of heat. Generally, the faster DSPs and the other components are run, the more power the devices need to overcome resistance and the more power will be dissipated within the various circuits of the transceiver. Furthermore, transmit power and power dissipation requirements directly influence component sizes. As a result, transmit power and power dissipation requirements become limiting factors for transceiver density. Accordingly, there is a need for a highly integrated AFE circuit that removes the requirement for external filters and buffers while retaining the capability to support multiple ADSL communications standards.
SUMMARY OF THE INVENTION
In light of the foregoing, the invention is a continuous-time smoothing filter that may be used as the output stage of a high-performance AFE with an increased integration level that accommodates multiple xDSL applications in a power efficient manner. The continuous-time smoothing f
GlobespanVirata Inc.
Le Dinh T.
Thomas Kayden Horstemeyer & Risley
LandOfFree
Circuit and method for implementing an integrated... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Circuit and method for implementing an integrated..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Circuit and method for implementing an integrated... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3159458