Table-Connected Tokenization
AssigneeProtegrity Corporation
InventorsUlf Mattsson, Yigal Rozenberg, Vichai Levy
Life of this patent
Estimating the term…
Abstract
A tokenization system tokenizes sensitive data to prevent unauthorized entities from accessing the sensitive data. The tokenization system accesses sensitive data, and retrieves an initialization vector (IV) from an IV table using a first portion of the sensitive data. A second portion of the sensitive data is modified using the accessed initialization vector. A token table is selected from a set of token tables using a third portion of the sensitive data. The modified second portion of data is used to query the selected token table, and a token associated with the value of the modified second portion of data is accessed. The second portion of the sensitive data is replaced with the accessed token to form tokenized data.
Claims20
A method for improving the security of data in a tokenization environment, comprising:
- receiving data to be tokenized;
- selecting, by a hardware processor, a token table from a set of token tables using a first portion of the received data, the selected token table mapping each of a plurality of input values to a different token value;
- and
- after selecting the token table, transforming the received data to produce tokenized data by:
- modifying, by the hardware processor, a second portion of the received data different than the first portion of the received data based on a value of a third portion of the received data different than the first portion and the second portion of the received data;
- querying, by the hardware processor, the token table with the modified second portion to identify a token value mapped to a value of the modified second portion;
- and
- replacing, by the hardware processor, the second portion of the received data with the identified token value to produce the tokenized data.
A tokenization system for improving the security of data in a tokenization environment, comprising:
- a non-transitory computer-readable storage medium storing executable instructions that, when executed by a processor, perform steps comprising:
- receiving data to be tokenized;
- selecting a token table from a set of token tables using a first portion of the received data, the selected token table mapping each of a plurality of input values to a different token value;
- and
- after selecting the token table, transforming the received data to produce tokenized data by:
- modifying a second portion of the received data different than the first portion of the received data based on a value of a third portion of the received data different than the first portion and the second portion of the received data;
- querying the token table with the modified second portion to identify a token value mapped to a value of the modified second portion;
- and
- replacing the second portion of the received data with the identified token value to produce the tokenized data;
- and
- a hardware processor configured to execute the instructions.
A non-transitory computer-readable storage medium storing executable computer instructions that when executed by a hardware processor perform steps for improving the security of data in a tokenization environment, comprising:
- receiving data to be tokenized;
- selecting a token table from a set of token tables using a first portion of the received data, the selected token table mapping each of a plurality of input values to a different token value;
- and
- after selecting the token table, transforming the received data to produce tokenized data by:
- modifying a second portion of the received data different than the first portion of the received data based on a value of a third portion of the received data different than the first portion and the second portion of the received data;
- querying the token table with the modified second portion to identify a token value mapped to a value of the modified second portion;
- and
- replacing the second portion of the received data with the identified token value to produce the tokenized data.
Dependencies are inferred from the claim text.