1. A dual gate oxide high-voltage semiconductor device, comprising:
a buried oxide layer formed over a semiconductor substrate;
a silicon layer formed over the buried oxide layer;
a top oxide layer formed over the silicon layer;
a first gate oxide formed over the silicon layer adjacent the top oxide layer; and
a second gate oxide formed over a portion of the first gate oxide.
2. The device of claim 1, wherein the silicon layer comprises a source region, a body region, and a drift region.
3. The device of claim 2, wherein the first gate oxide is formed over the drift region, the body region, and the source region.
4. The device of claim 2, wherein the second gate oxide is formed over the first gate oxide between the top oxide layer and the body region.
5. The device of claim 1, further comprising a field plate formed over the top oxide layer, the first gate oxide, and the second gate oxide.
6. The device of claim 1, wherein the first gate oxide has a thickness in a range of approximately 300-600A, and wherein the second gate oxide has a thickness in a range of approximately 900-1200A.
7. The device of claim 1, wherein the first gate oxide has a length of approximately 3-4 m, and wherein the second gate oxide has a length of approximately 1-2 m.
8. A dual gate oxide high-voltage semiconductor device, comprising:
a buried oxide layer formed over a semiconductor substrate;
a silicon layer formed over the buried oxide layer, wherein the silicon layer comprises a source region, a body region, and a drift region;
a top oxide layer formed over the silicon layer;
a first gate oxide formed over the silicon layer adjacent the top oxide layer; and
a second gate oxide formed over a portion of the first gate oxide between the top oxide layer and the body region.
9. The device of claim 8, further comprising a field plate formed over the top oxide layer, the first gate oxide and the second gate oxide.
10. The device of claim 8, wherein the first gate oxide has a thickness in a range of approximately 300-600A, and wherein the second gate oxide has a thickness in a range of approximately 900-1200A.
11. The device of claim 8, wherein the first gate oxide has a length of approximately 3-4 m, and wherein the second gate oxide has a length of approximately 1-2 m.
12. The device of claim 8, wherein a thickness of approximately 1200A for the second gate oxide results in an increase from approximately 1 e12cm2 to approximately 2 e12cm2 for a maximum allowable charge, and wherein a decrease of approximately 30% for a specific-on-resistance, of the device.
13. A method for forming a dual gate oxide high-voltage semiconductor device, comprising:
forming a buried oxide layer over a semiconductor substrate;
forming a silicon layer over the buried oxide layer;
forming a top oxide layer over the silicon layer;
forming a first gate oxide adjacent the top oxide layer over the silicon layer; and
forming a second gate oxide over the first gate oxide.
14. The method of claim 13, wherein forming the first gate oxide, and forming the second gate oxide comprises:
growing the first gate oxide adjacent the top oxide layer over the silicon layer;
applying a mask over the first gate oxide; and
growing the second gate oxide over a portion of the first gate oxide.
15. The method of claim 13, wherein forming the silicon layer comprises forming a silicon layer having a source region, a body region, and a drift region over the buried oxide layer.
16. The method of claim 15, wherein forming the first gate oxide comprises forming a first gate oxide over the drift region, the body region, and the source region.
17. The method of claim 15, wherein forming the second gate oxide comprises forming a second gate oxide over the first gate oxide between the top oxide layer and the body region.
18. The method of claim 13, further comprising:
increasing a maximum allowable charge of the device from approximately 1 e12cm2 to approximately 2 e12cm2; and
decreasing a specific-on-resistance of the device by approximately 30%.
19. The method of claim 13, wherein forming the first gate oxide and forming the second gate oxide comprises:
forming a first gate oxide having a thickness in a range of approximately 300-600A adjacent the top oxide layer over the silicon layer; and
forming a second gate oxide having a thickness in a range of approximately 900-1200A over the first gate oxide.
20. The method of claim 13, further comprising forming a field plate over the top oxide layer, the first gate oxide, and the second gate oxide.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A method, comprising:
receiving a symbol that has been modulated with multiple bits in accordance with a signal constellation comprising multiple constellation points;
computing a soft metric for a given bit in the received symbol by:
finding a first constellation point that is closest to the received symbol in accordance with a distance measure;
identifying a row or column of the constellation points that is closest to the first constellation point in accordance with the distance measure, and over which a value of the given bit is constant and is opposite to the value of the given bit in the first constellation point;
searching over at least the identified row or column for a second constellation point that is closest to the received symbol in accordance with the distance measure; and
calculating the soft metric based on the first and second constellation points; and
reconstructing the value of the given bit in the received symbol based on the soft metric.
2. The method according to claim 1, wherein searching for the second constellation point comprises searching over the identified row or column and over an additional partial subset of the constellation points that is predefined for the first constellation point.
3. The method according to claim 2, wherein searching for the second constellation point comprises accessing a Look-Up Table (LUT), which specifies a respective partial subset of the constellation points for each constellation point, so as to select the additional partial subset of the constellation points.
4. The method according to claim 1, wherein the signal constellation comprises rows and columns of the constellation that are not parallel to respective signal space axes.
5. The method according to claim 1, wherein calculating the soft metric comprises calculating a difference or a ratio between respective distances of the first and second constellation points from the received symbol.
6. The method according to claim 1, wherein calculating the soft metric comprises conditionally inverting the soft metric based on a sign of the given bit in the first constellation point.
7. The method according to claim 1, wherein searching for the second constellation point comprises deciding whether to search only over the identified row or column, or over the identified row or column and an additional predefined subset of the constellation points, based on a Signal to Noise Ratio (SNR) estimate.
8. The method according to claim 1, wherein the distance measure comprises a Max-Log approximation of a Maximum Likelihood (ML) estimate.
9. Apparatus, comprising:
a front-end receiver, which is configured to receive a symbol that has been modulated with multiple bits in accordance with a signal constellation comprising multiple constellation points; and
a baseband processor, which is configured, for a given bit in the received symbol, to find a first constellation point that is closest to the received symbol in accordance with a distance measure, to identify a row or column of the constellation points that is closest to the first constellation point in accordance with the distance measure, and over which a value of the given bit is constant and is opposite to the value of the given bit in the first constellation point, to search over at least the identified row or column for a second constellation point that is closest to the received symbol in accordance with the distance measure, to compute the soft metric based on the first and second constellation points, and to reconstruct the value of the given bit in the received symbol based on the soft metric.
10. The apparatus according to claim 9, wherein the baseband processor is configured to search for the second constellation point over the identified row or column and over an additional partial subset of the constellation points that is predefined for the first constellation point.
11. The apparatus according to claim 10, and comprising a Look-Up Table (LUT) that specifies a respective partial subset of the constellation points for each constellation point, wherein the baseband processor is configured to access the LUT so as to select the additional partial subset of the constellation points.
12. The apparatus according to claim 9, wherein the signal constellation comprises rows and columns of the constellation that are not parallel to respective signal space axes.
13. The apparatus according to claim 9, wherein the baseband processor is configured to compute the soft metric by calculating a difference or a ratio between respective distances of the first and second constellation points from the received symbol.
14. The apparatus according to claim 9, wherein the baseband processor is configured to conditionally invert the soft metric based on a sign of the given bit in the first constellation point.
15. The apparatus according to claim 9, wherein the baseband processor is configured to decide whether to search only over the identified row or column, or over the identified row or column and an additional predefined subset of the constellation points, based on a Signal to Noise Ratio (SNR) estimate.
16. The apparatus according to claim 9, wherein the distance measure comprises a Max-Log approximation of a Maximum Likelihood (ML) estimate.
17. A method, comprising:
receiving a symbol that has been modulated with multiple bits in accordance with a signal constellation comprising multiple constellation points;
computing a soft metric for a given bit in the received symbol by:
finding a first constellation point that is closest to the received symbol in accordance with a distance measure;
storing a first distance between the first constellation and the received symbol to serve as a term of the soft metric;
finding a second constellation point for which a value of the given bit is opposite from the value of the given bit in the first constellation point;
storing a second distance between the second constellation and the received symbol to serve as an opposite term of the soft metric; and
conditionally inverting the soft metric based on a sign of the given bit in the first constellation point; and
reconstructing the value of the given bit in the received symbol based on the soft metric.
18. The method according to claim 17, wherein finding the second constellation point comprises searching over at least a row or column of the constellation points that is closest to the first constellation point in accordance with the distance measure.
19. Apparatus, comprising:
a front-end receiver, which is configured to receive a symbol that has been modulated with multiple bits in accordance with a signal constellation comprising multiple constellation points; and
a baseband processor, which is configured, for a given bit in the received symbol, to find a first constellation point that is closest to the received symbol in accordance with a distance measure, to store a first distance between the first constellation and the received symbol to serve as a term of a soft metric for the given bit, to find a second constellation point for which a value of the given bit is opposite from the value of the given bit in the first constellation point, to store a second distance between the second constellation and the received symbol to serve as an opposite term of the soft metric, to conditionally invert the soft metric based on a sign of the given bit in the first constellation point, and to reconstruct the value of the given bit in the received symbol based on the soft metric.
20. The apparatus according to claim 19, wherein the baseband processor is configured to finding the second constellation point by searching over at least a row or column of the constellation points that is closest to the first constellation point in accordance with the distance measure.
21. A method, comprising:
receiving a multi-bit symbol that has been modulated with multiple bits in accordance with a signal constellation comprising multiple constellation points;
identifying a first constellation point having a minimal distance to the received symbol in accordance with a distance measure;
computing multiple soft metrics for the respective multiple bits of the received symbol based on the minimal distance of the first constellation point, without re-identifying the first constellation point or re-calculating the minimal distance for each soft metric; and
decoding the multiple bits of the received symbol based on the soft metrics.
22. Apparatus, comprising:
a front-end receiver, which is configured to receive a multi-bit symbol that has been modulated with multiple bits in accordance with a signal constellation comprising multiple constellation points; and
a baseband processor, which is configured to identify a first constellation point having a minimal distance to the received symbol in accordance with a distance measure, to compute multiple soft metrics for the respective multiple bits of the received symbol based on the minimal distance of the first constellation point, without re-identifying the first constellation point or re-calculating the minimal distance for each soft metric, and to decode the multiple bits of the received symbol based on the soft metrics.