Methods and apparatus for audio signature generation and matching
View Patent ↗A method for generating a signature is disclosed. As part of the method, a first number of bits are identified in respective rows of an audio signature matrix that are determined to be the strongest bits in the row, bits of the audio signature matrix are replaced with respective cells having values depending on whether the respective bits are included in the first number of bits, a set of uniformly distributed numbers are generated within a range of numbered locations corresponding to cells of the audio signature matrix; numerical distances are determined, from respective numbers in the set of uniformly distributed numbers, to the numbered locations of the matrix, associated with nearest occurrences of a first value. A set of integers is generated based on the distances.
1. An apparatus comprising:
memory including machine executable instructions; and
a processor to execute the machine executable instructions to:
build an inverted index table to identify locations of bit values in a reference signature, a first group of the bit values corresponding to a first tuple group of the reference signature and a second group of the bit values corresponding to a second tuple group of the reference signature, the first tuple group including a first starting bit and a first ending bit, the second tuple group including a second starting bit and a second ending bit, the first ending bit of the first tuple group corresponding to the second starting bit of the second tuple group;
identify a location of a first tuple group of a query signature in the inverted index table, the location of the first tuple group in the inverted index table corresponding to a location of a first group of bit values in the inverted index table that match the first tuple group of the query signature;
identify a hit based on the location of the first tuple group of the query signature in the inverted index table;
perform comparison of bytes of the query signature to bytes of the reference signature for a first number of bytes based on the hit, respective ones of the bytes of the reference signature corresponding to respective reference packets and respective ones of the bytes of the query signature corresponding to respective query packets; and
determine the query signature and the reference signature are a potential match when the comparison satisfies a threshold.
2. The apparatus of claim 1 , wherein the processor is to identify the first tuple group of the query signature based on identification of four successive tuple groups, a fifth successive tuple group identified as the first tuple group of the query signature, respective ones of the four successive tuple groups defined by corresponding groups of bit values.
3. The apparatus of claim 1 , wherein the processor is to identify the hit based on the reference signature and the query signature having a matching group index.
4. The apparatus of claim 3 , wherein the matching group index is a tuple group at a same location in the reference signature and the query signature.
5. The apparatus of claim 1 , wherein the processor is to discard the hit when the hit does not correspond to a first permutation group, the first permutation group corresponding to the location of the first tuple group.
6. The apparatus of claim 1 , wherein the processor is to compare a first number of values of the reference signature to a second number of values of the query signature.
7. The apparatus of claim 1 , wherein the processor is to perform the comparison based on integer matching.
8. A method comprising:
building, by executing an instruction with a processor, an inverted index table to identify locations of bit values in a reference signature, a first group of the bit values corresponding to a first tuple group of the reference signature and a second group of the bit values corresponding to a second tuple group of the reference signature, the first tuple group including a first starting bit and a first ending bit, the second tuple group including a second starting bit and a second ending bit, the first ending bit of the first tuple group corresponding to the second starting bit of the second tuple group;
identifying, by executing an instruction with the processor, a location of a first tuple group of a query signature in the inverted index table, the location of the first tuple group in the inverted index table corresponding to a location of a first group of bit values in the inverted index table that match the first tuple group of the query signature;
identifying, by executing an instruction with the processor, a hit based on the location of the first tuple group of the query signature in the inverted index table;
performing, by executing an instruction with the processor, comparison of a first number of values of the query signature to a second number of values of the reference signature based on the hit; and
determining, by executing an instruction with the processor, the query signature and the reference signature are a potential match when the comparison satisfies a threshold.
9. The method of claim 8 , wherein the identifying of the first tuple group of the query signature is based on identification of four successive tuple groups, a fifth successive tuple group identified as the first tuple group of the query signature, respective ones of the four successive tuple groups defined by corresponding groups of bit values.
10. The method of claim 8 , wherein the identifying of the hit is based on the reference signature and the query signature having a matching group index.
11. The method of claim 10 , wherein the matching group index is a tuple group at a same location in the reference signature and the query signature.
12. The method of claim 8 , further including discarding the hit when the hit does not correspond to a first permutation group, the first permutation group corresponding to the location of the first tuple group.
13. The method of claim 8 , wherein the performing of the comparison is based on integer matching.
14. A non-transitory computer readable medium comprising instructions that, when executed, cause a processor to at least:
build an inverted index table to identify locations of bit values in a reference signature, a first group of the bit values corresponding to a first tuple group of the reference signature and a second group of the bit values corresponding to a second tuple group of the reference signature, the first tuple group including a first starting bit and a first ending bit, the second tuple group including a second starting bit and a second ending bit, the first ending bit of the first tuple group corresponding to the second starting bit of the second tuple group;
identify a location of a first tuple group of a query signature in the inverted index table, the location of the first tuple group in the inverted index table corresponding to a location of a first group of bit values in the inverted index table that match the first tuple group of the query signature;
identify a hit based on the location of the first tuple group of the query signature in the inverted index table;
perform comparison of bytes of the query signature to bytes of the reference signature for a first number of bytes based on the hit, respective ones of the bytes of the reference signature corresponding to respective reference packets and respective ones of the bytes of the query signature corresponding to respective query packets; and
determine the query signature and the reference signature are a potential match when the comparison satisfies a threshold.
15. The computer readable medium of claim 14 , wherein the instructions, when executed, cause the processor to identify the first tuple group of the query signature based on identification of four successive tuple groups, a fifth successive tuple group identified as the first tuple group of the query signature, respective ones of the four successive tuple groups defined by corresponding groups of bit values.
16. The computer readable medium of claim 14 , wherein the instructions, when executed, cause the processor to identify the hit based on the reference signature and the query signature having a matching group index.
17. The computer readable medium of claim 16 , wherein the matching group index is a tuple group at a same location in the reference signature and the query signature.
18. The apparatus of claim 1 , wherein the instructions, when executed, further cause the processor to discard the hit when the hit does not correspond to a first permutation group, the first permutation group corresponding to the location of the first tuple group.
19. The computer readable medium of claim 14 , wherein the instructions, when executed, cause the processor to compare a first number of values of the reference signature to a second number of values of the query signature.
20. The computer readable medium of claim 14 , wherein the instructions, when executed, cause the processor to perform the comparison based on integer matching.