Method and device for compressing and expanding data pattern

• US 6,661,839 B1
• Filed: 11/20/1998
• Issued: 12/09/2003
• Est. Priority Date: 03/24/1998
• Status: Expired due to Fees

• Current Assignee: Advantest Corporation
• Original Assignee: Advantest Corporation
• Inventors: Yamaguchi, Takahiro, Ishida, Masahiro, Tilgner, Marco
• Primary Examiner(s): Ghebretinsae, Temesghen
• Application Number:
  US09/194,220
• Time in Patent Office: 1,845 Days
• Field of Search: 375/240, 375/224, 714/738, 714/741, 341/51
• US Class Current: 375/240
• CPC Class Codes: G01R 31/31921 using compression techniques, e.g. patterns sequencer
• CPC Class Codes: H03M 7/3084 using adaptive string matching, e.g. the Lempel-Ziv method
• CPC Class Codes: H03M 7/46 Conversion to or from run-length codes, i.e. by representing the number of consecutive digits, or groups of digits, of the same kind by a code word and a digit indicative of that kind

Abstract

There are provided methods each of which is for efficiently compressing a test pattern to be applied to an IC for testing. The number of data changes φ and a data entropy H of a pattern for each pin of an IC are obtained and then the test pattern is divided and the divided patterns are distributed to a block for φ that is equal to or less than a threshold value φ_M (φ<φ_M), a block for φ>φ_M and for H that is equal to or less than a threshold value H_M (H<H_M), and a block for H>H_M (411). The block for φ<φ_M is compressed by a run length compressing method, the block for φ>φ_M and H<H_M is compressed by the run length compressing method after application of Burrows wheeler transform, and the block for H>H_M is compressed by an LZ compressing method.

Claims

1. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method including the steps of:

counting a number of data changes of each said divided unit data sequence as a factor of difference in the data structure;

measuring an appearing probability of each symbol in each said divided unit data sequence and calculating an entropy of the data as a factor of difference in the statistical characteristics of data based on measured result of the appearing probabilities of the symbols;

comparing the number of data changes with a threshold value of the number of data changes and also comparing the entropy of the data with the threshold value of entropy;

dividing the inputted data to the respective unit data sequences and distributing the respective divided unit data sequences in accordance with the comparison results to a plurality of compressing blocks;

compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data;

dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

2. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and selectively distributing the respective divided unit data sequences in accordance with difference in their data structure into a plurality of compressing blocks, wherein step (a) includes;

selectively distributing the respective divided unit data sequences in accordance with difference in their data structures into a plurality of blocks;

counting a number of data changes of each said divided unit data sequence as a factor of difference in the data structure; and

comparing said counted number of data changes with a threshold value; and

selectively distributing the respective divided unit data sequences to a plurality of blocks in accordance with the comparison result;

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

3. The data compressing method according to claim 2, wherein said compressing step (b) includes the steps of:

applying a run length compressing method to a block in which the unit data sequences having the number of data changes less than the threshold value are contained, and

applying another compressing method to another block in which the unit data sequences having the number of data changes greater than the threshold value are contained.

4. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and selectively distributing the respective divided unit data sequences in accordance with difference in their data structure into a plurality of compressing blocks, wherein step (a) includes;

selectively distributing the respective divided unit data sequences in accordance with difference in their data structures into a plurality of blocks;

determining whether the periodicity of data of each said divided unit data sequence is larger than a threshold value; and

distributing the respective unit data sequences to a plurality of blocks in accordance with the determination result, and

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block, applying Burrows Wheeler transform (hereinafter referred to as BW transform) to a block in which the unit data sequences having a large periodicity of data larger than the threshold value are contained, and applying a run length compressing method to data transformed by BW transform;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

5. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and selectively distributing the respective divided unit data sequences in accordance with difference in their data structure into a plurality of compressing blocks, wherein step (a) includes;

selectively distributing the respective divided unit data sequences in accordance with difference in their statistical characteristic of data into a plurality of blocks;

measuring an appearing probability of each symbol in each said unit data sequence and calculating an entropy of data from thus measured result;

comparing the entropy of the data with a threshold value; and

selectively distributing the respective divided unit data sequences to a plurality of blocks in accordance with the comparison results;

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

6. The data compressing method according to claim 5, wherein a Huffman compressing method is applied in said compressing step (b) to a block having an entropy less than the threshold value among the divided blocks.

7. The data compressing method according to claim 5, wherein

an LZ compressing method is applied in said compressing step (b) to a block having an entropy less than the threshold value.

8. The data compressing method according to claim 5, wherein an arithmetic coding compressing method is applied in said compressing step (b) to a block having an entropy less than the threshold value.

9. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and distributing the respective divided unit data sequences in accordance with difference in their data structure to a plurality of compressing blocks, wherein step (a) includes;

counting a number of data changes of each said divided unit data sequence as a factor of difference in the data structure;

measuring an appearing probability of each symbol in each said divided unit data sequence;

calculating an entropy of the data based on measuring result of the appearing probabilities of the symbols;

comparing the number of data changes with a threshold value of the number of data changes and also comparing the entropy of the data with an entropy threshold value; and

selectively distributing the respective divided unit data sequences in accordance with the comparison results to a plurality of blocks which include a first block for containing the unit data sequence having its number of data change equal to or less than the threshold value, a second block for containing the unit data sequence having its number of data change larger than the threshold value and its entropy of the data equal to or less than the entropy threshold value, and a third block for containing the unit data sequence having its number of data change larger than the threshold value and its entropy of the data larger than the entropy threshold value; and

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes;

compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

10. The data compressing method according to claim 9, wherein

said compressing step (b) includes the steps of,

applying a run length compressing method to the first block, and

applying the run length compressing method after performing BW transform at least one time to the second block.

15. The data compressing method according to any one of claims 2 to 5 to 10, 13 and 14, wherein

the input data is test pattern which is applicable to a semiconductor integrated circuit for its testing, and

the unit data sequences are test sequences of the test pattern, each said test sequence being applicable to each pin of the semiconductor integrated circuit under test.

13. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data, to an expanded data wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and selectively distributing the respective divided unit data sequences in accordance with difference in their data structure into a plurality of compressing blocks, wherein step (a) includes;

selectively distributing the respective divided unit data sequences in accordance with difference in their data structures into a plurality of blocks;

applying BW transform to each the divided unit data sequence;

counting a number of data changes of each the unit data sequence transformed by BW transform;

calculating a compression rate from the number of data changes;

obtaining the maximum value of compression rate and the number of application times of BW transform when the maximum value of compression rate is obtained;

comparing the maximum value of compression rate with a threshold value of compression rate; and

selectively distributing the divided unit data sequences into a plurality of blocks in accordance with the comparison results;

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

11. The data compressing method according to any one of claims 2, 5 and 13, further including the steps of:

inputting environmental parameters; and

calculating the threshold values using the inputted environmental parameters.

12. The data compressing method according to any one of claims 2, 5 and 13, wherein

said dividing step (a) further includes the steps of:

calculating threshold values for respective possibly appearable unit data sequences;

estimating compression rates of the unit data sequences using the calculated threshold values;

optimizing the threshold value so as to maximize the compression rate; and

said compressing step (b) includes the steps of;

using the optimized threshold value.

14. The data compressing method according to claim 13, wherein

said compressing step (b) comprises the steps of:

applying one of an LZ compressing method, a Huffman compressing method, and an arithmetic coding compressing method to a first block for containing the unit data sequence having compression rate equal to or greater than the threshold value; and

applying a run length compressing method to a second block for containing the unit data sequence having compression rate less than the threshold value after performing BW transform the number of times by which the maximum compression rate is obtained.

47. The data compressing method according to claim 11, wherein

the inputted data is test pattern which is applicable to a semiconductor integrated circuit for its testing, and

the unit data sequences are test sequences of the test pattern, each said test sequence being applicable to each pin of the semiconductor integrated circuit under test.

48. The data compressing method according to claim 12, wherein

the inputted data is test pattern which is applicable to a semiconductor integrated circuit for its testing, and

the unit data sequences are test sequences of the test pattern, each said test sequence being applicable to each pin of the semiconductor integrated circuit under test.

16. A data expanding method for expanding an inputted compressed data which has a uniform data type and is composed of a plurality of compressed unit data sequences, the respective compressed unit data sequences being compressed by different compressing methods according to difference at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted compressed data into the respective compressed unit data sequences, which are in turn distributed to a plurality of blocks in accordance with their compressing methods used for compression of the respective compressed unit data sequences, so that each block contains the compressed unit data sequences compressed by the same compressing method which is different from that of the other block, wherein the inputted compressed data is divided into compressed unit data sequences directly compressed by a run length compressing method and compressed unit data sequences compressed by the run length compressing method after application of BW transform; and

(b) expanding the compressed unit data sequences contained in the respective blocks by using different expanding methods each corresponding to the compressing method of each blocks, wherein the compressed unit data sequences directly compressed by the run length compressing method are expanded by the run length expanding method to be outputted, and the compressed unit data sequences compressed by the run length compressing method after application of BW transform are expanded by the run length expanding method and thereafter the expanded data are transformed by inverse BW transform to be outputted.

17. A data expanding method for expanding an inputted compressed data which has a uniform data type and is composed of a plurality of compressed unit data sequences, the respective compressed unit data sequences being compressed by different compressing methods according to difference at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted compressed data into the respective compressed unit data sequences, which are in turn distributed to a plurality of blocks in accordance with their compressing methods used for compression of the respective compressed unit data sequences, so that each block contains the compressed unit data sequences compressed by the same compressing method which is different from that of the other block, wherein the inputted compressed data is a compressed data of a test pattern which is applicable to a semiconductor integrated circuit for its testing, and which is composed of a plurality of compressed test sequences, the respective test sequences being compressed by different compressing methods in accordance with difference at least either in their data structures or statistical characteristics of data and being applicable to respective pins of the semiconductor integrated circuit under test; and

(b) expanding the compressed unit data sequences contained in the respective blocks by using different expanding methods each corresponding to the compressing method of each block, wherein the expanded unit data sequences are the expanded test sequences each correspondingly applicable to each said pin.

18. The data expanding method according to claim 17, wherein

said dividing step (a) comprises a step of dividing the inputted compressed data into compressed unit data sequences compressed by a run length compressing method and compressed unit data sequences compressed by the other compressing methods, and

wherein

the compressed unit data sequences compressed by the run length compressing method are expanded, in said expanding step (b), by a run length expanding method.

19. The data expanding method according to claim 17, wherein

in said dividing step (a), the inputted compressed data is divided into compressed unit data sequences directly compressed by a run length compressing method and compressed unit data sequences compressed by the run length compressing method after application of BW transform, and

wherein

in said expanding step (b), the compressed unit data sequences directly compressed by the run length compressing method are expanded by the run length expanding method to be outputted, and the compressed unit data sequences compressed by the run length compressing method after application of BW transform are expanded by the run length expanding method and thereafter the expanded data are transformed by inverse BW transform to be outputted.

20. The data expanding method according to 17, wherein

in said dividing step (a), if there are compressed unit data sequences compressed by an LZ compressing method in the inputted compressed data, those compressed unit data sequences are separated from the other compressed unit data sequences, and

wherein

in said expanding step (b), the compressed unit data sequences compressed by the LZ compressing method are expanded by an LZ expanding method.

21. The data expanding method according to claim 17, wherein

in said dividing step (a), the inputted compressed data is divided into compressed unit data sequences compressed by a Huffman compressing method and compressed unit data sequences compressed by the other compressing methods, and

wherein

in said expanding step (b), a Huffman expanding method is applied to the compressed unit data sequences compressed by the Huffman compressing method.

22. The data expanding method according to claim 17, wherein

in said dividing step (a), the inputted compressed data is divided into compressed unit data sequences compressed by an LZ compressing method and compressed unit data sequences compressed by the other compressing methods, and

wherein

in said expanding step (b), an LZ expanding method is applied to the compressed unit data sequences compressed by the LZ compressing method.

23. The data expanding method according to claim 17, wherein

in said dividing step (a), the inputted compressed data is divided into compressed unit data sequences compressed by an arithmetic coding compressing method and compressed unit data sequences compressed by the other compressing methods, and

wherein

in said expanding step (b), an arithmetic coding expanding method is applied to the compressed unit data sequences compressed by the arithmetic coding compressing method.

24. The data expanding method according to claim 17, wherein

in said dividing step (a), the inputted compressed data is divided into compressed unit data sequences directly compressed by a run length compressing method, compressed unit data sequences compressed by the run length compressing method after application of BW transform, and compressed unit data sequences compressed by the other compressing methods, and

wherein

in said expanding step (b), a run length expanding method is applied to the compressed unit data sequences directly compressed by the run length compressing method to output the expanded data, the run length expanding method is applied to the compressed unit data sequences compressed by the run length compressing method after application of BW transform and thereafter inverse BW transform is applied to the expanded data the number of times that BW transform was applied thereto at the compression time to output the expanded data, and an expanding method corresponding to one of the other compressing methods is applied to each of the compressed unit data sequences compressed by the other compressing methods to output the expanded data.

28. The test pattern compressing apparatus according to claim 22, wherein said set of the compressing means further comprise compressing means other than said run length compressing means and said BWT run length compressing means.

29. The test pattern compressing apparatus according to claim 22, wherein said compressing method determining means comprises:

BW transform optimizing means for calculating the appropriate number of application times for applying BW transform to the inputted test sequence;

threshold value storage means for storing a threshold value of compression rate;

compression rate comparator means for comparing maximum value of compression rate obtained from said BW transform optimizing means with the threshold value stored in said threshold value storage means; and

compressing method selecting means for selecting a compressing method to compress a test sequence in accordance with the comparison result obtained from said compression rate comparator means and the appropriate number of application times BW transform obtained from said BW transform optimizing means.

49. The test pattern compressing apparatus according to claim 28, wherein said compressing method determining means comprises:

BW transform optimizing means for calculating the appropriate number of application times for applying BW transform to the inputted test sequence;

threshold value storage means for storing a threshold value of compression rate;

compression rate comparator means for comparing maximum value of compression rate obtained from said BW transform optimizing means with the threshold value stored in said threshold value storage means; and

compressing method selecting means for selecting a compressing method to compress a test sequence in accordance with the comparison result obtained from said compression rate comparator means and the appropriate number of application times BW transform obtained from said BW transform optimizing means.

25. A test pattern compressing apparatus comprising

a plurality of compressing means each having a different compressing method from one another;

dividing means for dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test; and

compressing method determining means for selecting suitable compressing means out of a plurality of compressing means in accordance with a data structure of each test sequence to compress the corresponding test sequence in the basis of each pin.

26. The test pattern compressing apparatus according to claim 25, wherein said compressing method determining means and said set of the compressing means are provided for each pin of said integrated circuit under test, and said dividing means supplies each test sequence to compressing method determining means of a corresponding pin.

27. The test pattern compressing apparatus according to claim 25 or 26, wherein said set of the compressing means comprise run length compressing means for compressing a test sequence by a run length compressing method, BW transform means for applying BW transform at least one time to the test sequence, and BWT run length compressing means for supplying the transformed output of BW transform to said run length compressing means.

44. An automatic semiconductor integrated circuit test system for automatically testing a semiconductor integrated circuit comprising:

a test pattern compressing apparatus recited in claim 25;

a large capacity storage apparatus for storing compressed test pattern data;

a pattern generator for storing the test pattern data and for generating a test pattern; and

test pattern expanding means for restoring a compressed test pattern data to an original test pattern without any information loss.

45. The automatic semiconductor integrated circuit test system according to claim 44, wherein said test pattern expanding means is comprised of a plurality of test sequence expanding means for restoring test sequence data each being compressed for each pin in parallel in the basis of respective pins.

30. A test pattern compressing method including the steps of:

dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test; and

selecting an appropriate compressing method in accordance with a data structure of each divided test sequence to compress the test sequence in the basis of each pin.

32. The test pattern compressing method according to claim 30 or 31, further including the steps of:

determining an appropriate compressing method adaptive for an inputted test sequence; and

applying to the test sequence a run length compressing method or the run length compressing method after applying BW transform to the test sequence one or more times (hereinafter referred to as BWT run length compressing method), or one of the other compressing methods such as an LZ compressing method, a Huffman compressing method, an arithmetic coding compressing method to compress the test sequence.

33. The test pattern compressing method according to claim 32, further including the steps of:

applying BW transform to the test sequence;

measuring the number of data changes of the data transformed by BW transform;

calculating a compression rate from the number of data changes;

obtaining the maximum value of compression rate and the number of application times of BW transform for maximizing the compression rate;

comparing the maximum value of compression rate with a threshold value of compression rate; and

selecting in accordance with the comparison result whether the test sequence is to be compressed by the run length compressing method, the BWT run length compressing method, or one of the other compressing methods.

31. A test pattern compressing method includes the steps of:

dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test; and

selecting appropriate compressing methods in parallel in accordance with respective data structures of the divided test sequences to compress the respective test sequences in the basis of respective pins.

34. A test pattern expanding apparatus comprising:

compressed data dividing means for dividing an inputted data into compressed data each being for each pin of an integrated circuit under test; and

a plurality of test sequence expanding means each for selecting an expanding method in accordance with a flag indicating a compressing method to apply the expanding method to each compressed data.

35. The test pattern expanding apparatus according to claim 34, wherein

said plurality of test sequence expanding means are provided for each pin of the integrated circuit under test;

said compressed data dividing means supplies each of the divided compressed data to corresponding one of said test sequence expanding means; and

said plurality of test sequence expanding means operate in parallel.

36. The test pattern expanding apparatus according to claim 34 or 35, further including:

compressing method determining means for determining, by the flag of the compressed data, whether the compressing method is a run length compressing method, a BWT run length compressing method, or one of the other compressing methods;

run length expanding means for expanding the compressed data using the run length expanding method;

repetitive inverse BW transform means for applying inverse BW transform one or more times to a data expanded by said run length expanding apparatus, and

the other expanding means for expanding the compressed data using the other expanding methods.

37. A test pattern expanding method including the steps of:

dividing an inputted compressed data into compressed data each being for each pin of an integrated circuit under test; and

expanding each divided compressed data by an expanding method corresponding to the compressing method specified by a flag of the compressed data.

38. A test pattern expanding method according to claim 37, wherein said expanding step is performed such that the compressed data are expanded in parallel in the basis of respective pins.

39. A The test pattern expanding method according to claim 37 or 38, wherein

the expanding method used in said expanding step is a run length expanding method or an expanding method in which inverse BW transform is applied to the data expanded by the run length expanding method.

40. The test pattern expanding method according to claim 39, wherein said expanding step further includes expanding method other than said run length expanding method and said expanding method in which inverse BW transform is applied to the data expanded by the run length expanding method.

41. A test pattern compressing and expanding apparatus comprising:

a plurality of compressing means each having a different compressing method from one another;

dividing means for dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test;

compressing method determining means for selecting suitable compressing means out of a plurality of compressing means in accordance with a data structure of each test sequence to compress the corresponding test sequence in the basis of a pin;

compressed data dividing means for dividing an inputted data into compressed data each being for each pin of an integrated circuit under test; and

a plurality of test sequence expanding means each for selecting an expanding method in accordance with a flag indicating a compressing method to apply the expanding method to the corresponding compressed data.

46. An automatic semiconductor integrated circuit test system for automatically testing a semiconductor integrated circuit comprising:

means for compressing a test pattern;

a large capacity storage apparatus for storing compressed test pattern data;

a pattern generator for storing the test pattern data and for generating a test pattern; and

a test pattern compressing and expanding apparatus recited in claim 41.

42. A test pattern compressing and expanding method including the steps of:

dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test;

selecting an appropriate compressing method in accordance with a data structure of each divided test sequence data to compress the test sequence in the basis of a pin;

dividing an inputted compressed data into compressed data each being for each pin of an integrated circuit under test; and

expanding each divided compressed data by an expanding method corresponding to the compressing method specified by a flag of the compressed data.

43. A test pattern compressing and expanding method including the steps of:

dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test;

selecting appropriate compressing methods in parallel in accordance with respective data structures of the divided test sequences to compress the respective test sequences in the basis of respective pins;

dividing an inputted compressed data into compressed data each being for each pin of an integrated circuit under test; and

expanding each divided compressed data by an expanding method corresponding to the compressing method specified by a flag of the compressed data.

50. A test pattern compressing apparatus comprising:

a plurality of compressing means each having a different compressing method from one another;

dividing means for dividing an inputted test pattern into test sequences each being for each of pins of an integrated circuit under test; and

compressing method determining means for selecting suitable compressing means out of the plurality of compressing means in accordance with difference in statistical characteristics of data of each test sequence to compress the corresponding test sequence in the basis of each pin.

51. A data expanding method for expanding an inputted compressed data which has a uniform data type and is composed of a plurality of compressed unit data sequences, the respective compressed unit data sequences being compressed by different compressing methods according to difference at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted compressed data into the respective compressed unit data sequences, which are in turn distributed to a plurality of blocks in accordance with their compressing methods used for compression of the respective compressed unit data sequences, so that each block contains the compressed unit data sequences compressed by the same compressing method which is different from that of the other block, wherein the inputted compressed data is divided into compressed unit data sequences directly compressed by a run length compressing method, compressed unit data sequences compressed by the run length compressing method after application of BW transform, and compressed unit data sequences compressed by the other compressing methods; and

(b) expanding the compressed unit data sequences contained in the respective blocks by using different expanding methods each corresponding to the compressing method of each block, wherein a run length expanding method is applied to the compressed unit data sequences directly compressed by the run length compressing method to output the expanded data, the run length expanding method is applied to the compressed unit data sequences compressed by the run length compressing method after application of BW transform and thereafter inverse BW transform is applied to the expanded data the number of times that BW transform was applied thereto at the compression time to output the expanded data, and an expanding method corresponding to one of the other compressing methods is applied to each of the compressed unit data sequences compressed by the other compressing methods to output the expanded data.

52. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and selectively distributing the respective divided unit data sequences in accordance with difference in their data structure into a plurality of compressing blocks, wherein step (a) includes selectively distributing the respective divided unit data sequences in accordance with difference in their data structures into plurality of blocks, wherein the inputted data is a test pattern which is applicable to a semiconductor integrated circuit for its testing, and the unit data sequences are test sequences of the test pattern, each said test sequence being applicable to each pin of the semiconductor integrated circuit under test;

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.

53. A data compressing and expanding method for compressing an inputted data to a compressed data and for expanding the compressed data to an expanded data, wherein the inputted data has a uniform data type and is composed of a plurality of unit data sequences and the respective unit data sequences are different at least either in their data structures or statistical characteristics of data, said method comprising the steps of:

(a) dividing the inputted data to the respective unit data sequences and selectively distributing the respective divided unit data sequences in accordance with difference in their data structure into a plurality of compressing blocks, wherein step (a) includes selectively distributing the respective divided unit data sequences in accordance with difference in their statistical characteristic of data into a plurality of blocks, wherein the inputted data is a test pattern which is applicable to a semiconductor integrated circuit for its testing, and the unit data sequences are test sequences of the test pattern, each said test sequence being applicable to each pin of the semiconductor integrated circuit under test;

(b) compressing the respective divided unit data sequences distributed to said plurality of compressing blocks using different compressing methods which are each suitable for each block to thereby obtain a compressed data, wherein step (b) includes compressing the respective unit data sequences distributed into said plurality of blocks using different compressing methods each of which is suitable for each block;

(c) dividing the compressed data to respective compressed unit data sequences and distributing the thus divided respective compressed unit data sequences in accordance with their compressing methods used to a plurality of expanding blocks in accordance with difference in the compressing methods used; and

(d) expanding the respective compressed unit data sequences contained in each of said plurality of the expanding blocks by using a suitable one expanding method which is selected for each said expanding block from different expanding methods so as to correspond to the compressing method used to the compressed unit data sequence.