Analog/digital conversion using successive approximation and redundant weighting
View Patent ↗Illustrative binary networks for analog/digital converters are described. For instance, an analog/digital converter may operate in accordance with a principle of successive approximation, and may include a plurality of N+1 circuit nodes each representing one of N binary weights and M≧1 redundant binary weights, wherein each binary weight corresponds to a different position of a redundant binary code formed as part of the approximation. Illustrative methods of operation of the binary network and/or the analog/digital converter are also described.
1. A binary network for an analog/digital converter operating in accordance with a principle of successive approximation, comprising a plurality of circuit nodes each representing one of N binary weights and a plurality of redundant binary weights, wherein each of the binary weights and the redundant binary weights corresponds to a different position of a redundant binary code formed as part of the approximation, wherein N is a whole number greater than one.
2. A binary network for an analog/digital converter operating in accordance with a principle of successive approximation, comprising a plurality of circuit nodes each representing one of N binary weights and at least one redundant binary weight, wherein each of the binary weights and the at least one redundant binary weight corresponds to a different position of a redundant binary code formed as part of the approximation, wherein N is a whole number greater than one, each of the binary weights and the at least one redundant binary weight has a weight value, and the weight value of one of the at least one redundant weight corresponds to the weight value of an Mth most significant weight of the N binary weights, M being a positive whole number smaller than N.
3. The binary network of claim 2 , wherein the binary network includes a plurality of N+M capacitors each corresponding to a different one of the N binary weights or the at least one redundant binary weight.
4. The binary network of claim 3 , wherein those of the capacitors representing a most significant binary weight through the Mth most significant weight and the at least one redundant binary weight are each switchably connectable to the input signal.
5. A binary network for an analog/digital converter operating in accordance with a principle of successive approximation, comprising a plurality of circuit nodes each representing one of N binary weights and at least one redundant binary weight, each of the binary weights and the at least one redundant binary weight having a weight value, wherein each of the binary weights corresponds to a different position of a redundant binary code formed as part of the approximation, wherein N is a whole number greater than one, and wherein the binary network includes a plurality of groups of capacitors, each group including a plurality of capacitors, each group corresponding to a different one of the binary weights or the at least one redundant binary weight.
6. An analog/digital converter operating in accordance with a principle of successive approximation, comprising:
a binary network configured to implement digital/analog conversion in accordance with N binary weights and a redundant binary weight, each of the binary weights and the redundant binary weight having a weight value, wherein each binary weight corresponds to a position of a redundant binary code formed as part of the approximation, wherein N is a whole number greater than one; and
a code converter coupled to the binary network and configured to convert the redundant binary code into a binary output code,
wherein:
the weight value of the redundant binary weight corresponds to the weight value of an Mth most significant binary weight of the N binary weights, M being a positive whole number less than N,
the redundant binary code includes a plurality of N bits each associated with a different one of the binary weights and a bit associated with the redundant binary weight, and
the code converter is configured to convert the redundant binary code into the binary output code by adding a value of the bit associated with the redundant binary weight in accordance with the weight value of the Mth most significant weight and subtracting an offset value.
7. The analog/digital converter of claim 6 , wherein the offset value is 2 (N−M−1) .
8. A method for analog/digital conversion of an input signal in accordance with a principle of successive approximation, the method comprising:
approximating an input signal using a binary network, wherein the binary network comprises a plurality of circuit nodes each representing one of N binary weights and at least one redundant binary weight, wherein each of the binary weights corresponds to a different position of a redundant binary code formed as part of the approximation, the redundant binary code has a number of bits equal to the number of binary weights and redundant binary weights, and N is a whole number greater than one; and
converting the redundant binary code into a binary output code having N bits.
9. The method of claim 8 , wherein the binary network includes a plurality of groups of capacitors, each group of capacitors corresponding to a different one of the N binary weights or the at least one redundant binary weights, the method further comprising:
receiving the input signal; and
connecting to the input signal the groups of capacitors representing the most significant binary weight through an Mth most significant weight and the at least one redundant binary weight, M being a positive whole number smaller than N.
10. A method for analog/digital conversion of an input signal in accordance with a principle of successive approximation, the method comprising:
approximating an input signal using a binary network, wherein the binary network comprises a plurality of circuit nodes each representing one of N binary weights and at least one redundant binary weight, wherein each of the binary weights corresponds to a different position of a redundant binary code having at least N+1 bits formed as part of the approximation, wherein N is a whole number greater than one and M is a positive whole number smaller than N, wherein each of the binary weights and the at least one redundant binary weight has a weight value, wherein the weight value of the redundant binary weight corresponds to the weight value of an Mth most significant weight of the N binary weights, and wherein the redundant binary code includes a plurality of N bits each associated with a different one of the binary weights and at least one bit associated with the at least one redundant binary weight, the method further comprising converting the redundant binary code into the binary output code by adding a value of the at least one bit associated with the at least one redundant binary weight in accordance with a weighting of the Mth most significant weight and subtracting an offset value.
11. The method of claim 10 , wherein the offset value is 2 (N−M−1) .