Coded data generation or conversion – Analog to or from digital conversion – Digital to analog conversion
Reexamination Certificate
1995-12-07
2001-05-15
Young, Brian (Department: 2819)
Coded data generation or conversion
Analog to or from digital conversion
Digital to analog conversion
Reexamination Certificate
active
06232903
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to noise cancellation circuits and particularly but not exclusively to noise cancellation circuits for use with D/A and A/D converters.
BACKGROUND OF THE INVENTION
In an oversampled digital to analogue (D/A) converter, such as a sigma-delta modulator, a digital input signal is to be converted to an analogue output signal. The input signal is first interpolated to produce an oversampled signal, which is modulated by a digital modulator to generate an N-bit signal. The N-bit signal has a low-frequency component resembling the input signal and a high-frequency “quantization noise”, which is typically removed with a low pass filter.
For an N-bit D/A converter there must be at least one (N=1) dynamic analogue element, such as a current source, which is used to generate the analogue signal. Each additional element used gives an increase in signal to noise ratio (SNR) of 6 dB, so it is desirable to have a large N if possible.
However, a problem with this arrangement is that each analogue element will be slightly mismatched with respect to the others, and the oversampling gives rise to harmonic distortion if the elements, which are arranged in an array, are always used in the same order.
Methods of dynamic element matching are known, which seek to disrupt the order of using the elements by selecting a new starting position in the array, based on the last used element (such that the elements are used substantially equally). However, for some signal levels, the above matching method gives rise to a pattern of element usage that produces low frequency tone.
This invention seeks to provide a noise cancellation circuit which mitigates the above mentioned disadvantages.
SUMMARY OF THE INVENTION
According to a first aspect of the present invention there is provided a noise cancellation circuit for use with a digital signal, comprising: a plurality of equally weighted cells for receiving the digital signal and for providing an analogue output signal in dependence upon the value of the received digital signal; and switching means for dynamically switching a number of the plurality of cells according to a sequencing scheme, wherein the sequencing scheme comprises a first sequence arranged to switch each of the plurality of cells an equal number of times, and a second sequence arranged to define one of the plurality of cells as a starting position for the first sequence, such that low frequency tone generation within the analogue output signal is substantially eliminated.
Preferably the switching means is a decoder having a plurality of output lines, such that each line is coupled to switch one of the plurality of cells. The switching means preferably includes a random number generator, arranged to provide a random number, such that the starting position for the first sequence is defined randomly at each occurrence of the second sequence.
Preferably the plurality of cells is an array of switchable current sources. Alternatively, the plurality of cells is preferably an array of switchable capacitors.
According to a second aspect of the present invention there is provided a noise cancellation method, comprising the steps of: receiving a digital signal; modulating the digital signal with a delayed digital signal to provide a modulated signal; quantizing the modulated signal for providing a quantized signal, and dynamically switching a number of cells in dependence upon the value of the quantized signal, to provide an analogue output signal, the switching being performed according to a sequencing scheme, wherein the sequencing scheme comprises a first sequence arranged to switch each of the plurality of cells an equal number of times, and a second sequence arranged to define one of the plurality of cells as a starting position for the first sequence, such that low frequency tone generation within the analogue output signal is substantially eliminated.
Preferably the second sequence includes a random step, such that the starting position for the first sequence is defined randomly at each occurrence of the second sequence. A random number is preferably generated at each occurrence of the second sequence, to provide a value for the random step.
Preferably the first sequence includes an incremental step, each of the plurality of cells being used once per occurrence of the first sequence. The first and second sequences are preferably used alternately by the sequencing scheme.
In this way the use of the cells is managed such that harmonic distortion caused by mismatching of the elements is reduced.
REFERENCES:
patent: 3982172 (1976-09-01), Van De Plassche
patent: 4125803 (1978-11-01), Van De Plassche
patent: 5084701 (1992-01-01), Sauerwald
patent: 5221926 (1993-06-01), Jackson
patent: 5293166 (1994-03-01), Ta
patent: 5305004 (1994-04-01), Fattaruso
patent: 5508702 (1996-04-01), Estrada et al.
patent: 2080059A (1981-07-01), None
Afek Yachin
Koifman Vladimir
Liberman Sergio
Motorola Inc.
Young Brian
LandOfFree
Sequencing scheme for reducing low frequency tone generation... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Sequencing scheme for reducing low frequency tone generation..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Sequencing scheme for reducing low frequency tone generation... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2541079