1. Method for compressing data, comprising the steps of:
transforming data of a frame into a first sequence of real numbers by using an entropy increasing transform method;
quantizing said first sequence of real numbers of said frame to obtain a first sequence of integers;
quantizing said first sequence of integers of said frame using a predicted sequence of integers representing said first sequence of integers to produce a second sequence of integers; and
encoding said integers of said second sequence of said frame into a stream of bits representing a compressed sequence of integers of said frame.
2. Method according to claim 1, wherein quantizing said first sequence of integers of said frame comprises the step of:
for each coefficient of said second sequence of integers, selecting a relation between an integer of said first sequence of integers of said frame and the corresponding integer of a reference frame and the corresponding integer of the predicted sequence based on an evaluation value of said relation.
3. Method according to claim 2, wherein the step of selecting comprises the step of
for each coefficient of said second sequence of integers, comparing a first relation between the integer of said first sequence of integers of said frame and the corresponding integer of a reference frame and a second relation between said integer and the corresponding integer of the predicted sequence; and
determining the evaluation value for each relation based on respective relation in encoded form.
4. Method according to claim 2, further comprises the step of
determining an absolute value of respective relation, wherein the evaluation value for each relation is set to the corresponding absolute value.
5. Method according to claim 4, wherein the step of selecting comprises the step of selecting the relation having a lowest absolute value.
6. Method according to claim 2 or 3, wherein the step of selecting comprises the step of:
selecting said relation according to,
c
\u2032
=
c
–
r
\u2062
if
\u2062
\u2062
p
=
r
\u2062
\u2062
or
\u2062
\u2062
c
–
r
p
–
r
<
1
2
,
c
\u2032
=
c
–
p
\u2062
otherwise
\u2003where c\u2032 is an integer of the second sequence, p is the corresponding integer of the predicted sequence, c is the corresponding integer of the first sequence of a current frame, and r is the corresponding integer of a reference frame.
7. Method according to claim 4, wherein the step of quantizing said first sequence of integers of said frame further comprises the step of: if
p
\u2260
r
\u2062
\u2062
and
\u2062
–
1
2
\u2264
c
–
r
p
–
r
<
3
2
,
associating a control bit identifying the selected relation.
8. Method according to any one of claims 1-5 and 7, further comprising the step of:
storing said stream of bits as a compressed representation of said sequence of said frame.
9. Method according to any one of the claims 1-5 and 7, further comprising decompressing said compressed sequence by inverting the steps of transforming, quantizing said first sequence of real numbers, quantizing said first sequence of integers of said frame, and decoding in reverse order.
10. Method according to claim 9, wherein the step of inverting the step of quantizing said first sequence of integers comprises the steps of:
reconstructing a sequence of integers of a current frame according to,
c
=
c
\u2032
+
r
\u2062
if
\u2062
\u2062
p
=
r
\u2062
\u2062
or
\u2062
\u2062
c
–
r
p
–
r
<
1
2
,
c
=
c
\u2032
+
r
\u2062
otherwise
\u2003where c\u2032 is an integer of the compressed sequence, r is the corresponding integer of the reference frame, and c is the corresponding integer of the reconstructed sequence representing the first sequence of the current frame.
11. Method according to claim 10, wherein the step of reconstructing comprises the step of
if
if
\u2062
\u2062
p
\u2260
r
\u2062
\u2062
and
\u2062
–
1
2
\u2264
c
–
r
p
–
r
<
3
2
,
\u2003where p an integer of the predicted sequence and c is the corresponding integer of the current frame, using the associated control bit to identify the relation between an integer, c\u2032, of the compressed sequence, the corresponding integer, r, of the reference frame, and the corresponding integer, c, of the reconstructed sequence representing the first sequence of the current frame.
12. Method according to claim 10, further comprising the step of
storing the reconstructed sequence of integers.
13. Method according to any one of claims 1-5 and 7, wherein the predicted sequence is a simulated reconstructed sequence of a previous frame.
14. Method according to any one of the claims 1-5 and 7, wherein the entropy increasing transform method is a wavelet transform method.
15. System for compressing and decompressing data, comprising:
a storage device for storing data;
transform means arranged to transform a frame of data into a first sequence of real numbers;
compression processing means, comprising
quantization means arranged to quantize said first sequence of real numbers to produce a first sequence of integers;
adaptive quantization means arranged to quantize said first sequence of integers to produce a second sequence of integers by using a predicted sequence of integers representing said first sequence of integers; and
encoding means arranged to encode said integers of said second sequence of said frame into a stream of bits representing the compressed sequence of integers of said frame.
16. System according to claim 15, farther comprising
reconstruction means comprising
decoding means arranged to decode a bit stream representing a compressed sequence of integers into a third sequence of integers;
inverse adaptive quantization means arranged to inversely quantize said fourth sequence of integers to produce a reconstructed first sequence of integers by using the predicted sequence of integers representing said first sequence of integers; and
inverse quantization means arranged to inversely quantize said reconstructed first sequence of integers to produce a second sequence of integers second sequence of real numbers; and
inverse transform means arranged to inversely transform said sequence of real number to a reconstructed frame of data.
17. Computer readable medium comprising instructions for bringing a computer to perform the method according to any one of the claims 1-5 and 7.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
What is claimed is:
1. A method for continuously growing multiple semiconductor ribbons concurrently in a single crucible, the method comprising
providing a crucible having multiple meniscus shapers disposed in a spaced relationship therein;
forming a melt from a semiconductor material disposed in the crucible, the multiple meniscus shapers separating the melt into a plurality of distinct melt subregions;
arranging multiple pairs of strings relative to the multiple meniscus shapers, each pair of strings (i) having a fixed distance therebetween, (ii) emerging from one of the distinct melt subregions, and (iii) defining a pair of edges of a meniscus and controlling the width of a ribbon; and
continuously pulling the multiple pairs of strings away from a surface of the melt to form multiple discrete and substantially flat semiconductor ribbons.
2. The method of claim 1 wherein the semiconductor material is silicon, germanium, silicon carbide, or alloys thereof.
3. The method of claim 1 further comprising positioning multiple afterheaters relative to the ribbons wherein each afterheater is disposed adjacent a surface of at least one of the ribbons to control the thermal profiles of the ribbons.
4. The method of claim 1 further comprising arranging the multiple pairs of strings relative to the multiple meniscus shapers such that the ribbons are grown substantially parallel to one another in a face-to-face pattern.
5. The method of claim 1 further comprising arranging the multiple pairs of strings relative to the multiple meniscus shapers such that the ribbons are grown substantially in a head-to-tail pattern.
6. The method of claim 1 further comprising arranging the multiple pairs of strings relative to the multiple meniscus shapers such that the ribbons are grown in a pattern comprising adjacent ribbons in both a face-to-face and a head-to-tail pattern.
7. The method of claim 1 wherein the multiple meniscus shapers have identical size and shape.
8. The method of claim 1 wherein the multiple meniscus shapers have different size or shape.
9. The method of claim 1 wherein two to twenty ribbons are grown concurrently.
10. The method of claim 1 wherein the distance between each pair of strings varies.
11. The method of claim 1 further comprising pulling the multiple pairs of strings in a direction perpendicular to the melt surface.
12. The method of claim 1 further comprising pulling the multiple pairs of strings in a direction other than perpendicular to the melt surface.
13. A method for minimizing interference between adjacent ribbons in a multiple semiconductor ribbon growth system, the method comprising
forming a melt from a semiconductor material disposed in an open crucible;
partitioning the melt into a plurality of distinct melt subregions by disposing a plurality of meniscus shapers in the crucible, each melt subregion having a distinct melt surface defined by the meniscus shaper;
continuously growing multiple semiconductor ribbons, each of the ribbons being grown from a melt subregion by pulling a pair of spaced strings away from the distinct melt surface.
14. The method of claim 13 wherein the step of partitioning the melt into a plurality of distinct melt subregions comprises positioning at least one meniscus shaper on the melt surface.
15. The method of claim 13 wherein the semiconductor material is silicon.
16. The method of claim 13 further comprising positioning multiple afterheaters relative to the ribbons wherein each afterheater is disposed adjacent at least one of the ribbons to control the thermal profiles of the ribbons.
17. An apparatus for continuously growing multiple semiconductor ribbons concurrently in a single crucible, the apparatus comprising
a crucible for holding a melt of a semiconductor material;
multiple meniscus shapers arranged in a spaced relationship in the crucible to partition the melt into a plurality of distinct melt subregions;
multiple pairs of strings wherein each pair is disposed relative to one of the multiple meniscus shapers, each pair of strings (i) having a fixed distance therebetween, (ii) emerging from one of the melt subregions, (iii) defining a pair of edges of a meniscus, and (iv) defining a width of one of the multiple semiconductor ribbons as the pair of strings is pulled from the melt subregion; and
multiple afterheaters wherein each afterheater is disposed adjacent a surface of at least one of the semiconductor ribbons to control the thermal profiles of the semiconductor ribbons.
18. The apparatus of claim 17 wherein each pair of strings pass through a pair of holes in the crucible.
19. The apparatus of claim 17 wherein each afterheater is disposed adjacent an outer surface of a ribbon.
20. The apparatus of claim 17 wherein the distance between the strings of each pair of strings varies.
21. The apparatus of claim 17 further comprising a housing for isolating from the ambient environment the melt and a portion of the solidifying ribbon comprising a solid-liquid interface and having a temperature of 400 C. or higher.
22. The apparatus of claim 17 comprising two to twenty pairs of strings for concurrently growing two to twenty ribbons.