Pulse or digital communications – Equalizers
Reexamination Certificate
2000-02-07
2004-05-25
Chin, Stephen (Department: 2634)
Pulse or digital communications
Equalizers
Reexamination Certificate
active
06741644
ABSTRACT:
BACKGROUND OF THE INVENTION
The present relation relates to data transmission channels and, more particularly, to a receiver having a pre-emphasis filter for mitigating inter-symbol interference effects from channels having low-pass filter characteristics.
In a typical communication system, analog signals are transmitted from a transmitter to a receiver through a transmission channel. A typical receiver includes an amplifier, an analog-to-digital (A/D) converter, an automatic gain controller and an equalizer. The amplifier amplifies the received analog signal, which is then converted by the A/D converter into a series of digital values that can be processed. The automatic gain controller monitors the signal power and adjusts the amplifier gain to restore the received signal to the desired power level. The equalizer equalizes the series of digital values to match a target response for the particular channel being used.
Various types of channels have been used, such as “twisted pair” wire line channels. Such channels are commonly used for transmission of data and voice over plain old telephone systems (POTS), digital subscriber loops (DSL) and their numerous variations, home, local and wide area networks (LAN and WAN), and other applications. These types of channels have limited bandwidths and therefore act as low-pass filters, which introduce undesired inter-symbol interference (ISI) effects in the signals being transmitted.
There are three general methods for mitigating or removing the undesired ISI effects of twisted pair cables. The first method uses a fully digital equalizer after the A/D converter to cancel the ISI effects, with no analog or mixed signal filtering. This fully digital equalization method sacrifices performance since the A/D converter introduces significant quantization noise to the series of digital values on which the equalizer operates. Due to the low-pass filter nature of twisted pair cables, the root mean square (RMS) signal power attenuates much more rapidly than the peak-to-peak signal power with increasing cable length. Hence, the ratio of the RMS signal power to the peak-to-peak signal power decreases as the cable length increases.
Since the A/D converter has a dynamic range that is determined by the peak-to-peak signal power and not the RMS signal power, the effective quantization noise of the A/D converter cannot be reduced by just optimizing the conversion window range of the A/D converter to the RMS signal power. For example, for a given peak-to-peak signal power and hence A/D conversion window, a factor of two decrease in the RMS signal power would translate to 6dB of additional quantization noise. This additional quantization noise would adversely affect the system performance unless it is mitigated by adding an extra bit to the A/D converter. Adding an extra bit to the A/D converter can be very expensive since the complexity of the A/D converter increases exponentially with the number of bits used.
The second method of mitigating the undesired ISI effects uses a complex multi-pole, multi-zero analog/mixed signal high-pass pre-emphasis filter to partially cancel the channel s low-pass filtering effect. This partial cancellation aids the equalizer in equalizing the received signal to the target response. However, a complex analog/mixed signal filter is very difficult to design accurately and is very expensive in terms of power consumption and silicon area in an integrated circuit application. The design of such filters requires the implementation of resistors and capacitors in silicon. Not only do these resistors and capacitors consume power and area, they are also very sensitive to manufacturing process variations and therefore provide very inaccurate filters that require an extensive tuning circuit to achieve the specified response. The difficulty in tuning and inaccuracy of such filters increase exponentially with the filter complexity, which makes it difficult or unproductive to expect and specify an accurate filter response.
The third method of mitigating the undesired ISI effects uses a fully analog equalizer. Analog equalization suffers from some of the same analog circuit implementation shortcomings discussed above. Hence, the use of a fully analog equalizer to cancel the channel ISI effects is also not a very attractive solution.
Improved filters and methods for efficiently and economically mitigating undesired ISI effects of channels having low-pass filter characteristics are desired.
SUMMARY OF THE INVENTION
One aspect of the present invention relates to a communications receiver provided for receiving a transmitted signal from a transmission channel having a low-pass filter characteristic. The receiver includes a receiver input for coupling to the channel and a switched capacitor pre-emphasis filter coupled to the receiver input. An analog-to-digital (A/D) converter is coupled to an output of the pre-emphasis filter. An equalizer is coupled to an output of the analog-to-digital converter.
Another aspect of the present invention relates to a communications transceiver, which includes a transmitter, a transmission channel coupled to the transmitter and having a frequency response with a low-pass filter characteristic, and a receiver. The receiver includes a switched capacitor pre-emphasis filter coupled to an output of the transmission channel, an analog-to-digital (A/D) converter coupled to an output of the pre-emphasis filter and an equalizer coupled to an output of the analog-to-digital converter.
Yet another aspect of the present invention relates to a method of removing inter-symbol interference (ISI) effects from an analog signal received from a transmission channel having a frequency response with a low-pass filter characteristic. In the method, the analog signal is filtered with a switched capacitor pre-emphasis filter having a frequency response that approximates an inverse of the frequency response of the transmission channel. The step of filtering produces a filtered analog signal in which a first portion of the ISI effects are removed. The filtered analog signal is then converted to a series of digital signals, and the series of digital signals are passed through a digital equalizer to remove a second portion of the ISI effects.
REFERENCES:
patent: 5479522 (1995-12-01), Lindemann et al.
patent: 5719326 (1998-02-01), Vulih et al.
patent: 6351506 (2002-02-01), Lewicki
Chen Ting-Yin
Dabiri Dariush
Dehghan Hossein
Chin Stephen
LSI Logic Corporation
Odom Curtis
Westman Champlin & Kelly
LandOfFree
Pre-emphasis filter and method for ISI cancellation in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Pre-emphasis filter and method for ISI cancellation in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Pre-emphasis filter and method for ISI cancellation in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3224309