Recognition Strategies: Integrity Detection

This chapter discusses integrity detection as a recognition strategy. Integrity is discussed in several contexts. The use of checksums is discussed, with stress on the fact that checksums provide safety but not security. Several increasingly complicated algorithms are discussed to introduce the concepts of integrity detection: parity bits, repetition codes, and Hamming codes. Pointers to other encodings are provided, however, an exhaustive discussion of all encoding types is not presented. The example of checksum function in UPCs is developed to demonstrate the concepts. Then cryptographic integrity detection and digital signatures using hash functions is discussed. RSA and DSS are mentioned, however, the algorithms are not developed in detail. The general form of a signature algorithm is discussed, as are the basic desired properties of a hash function, such as collision resistance and pre-image resistance. After a sidebar on Richard Hamming, rule-based methods are introduced. Rule-based methods focus on database integrity. Rules in the context of entity integrity, referential integrity, and domain integrity are discussed; input sanitization is also introduced. The final integrity check method discussed is content comparisons, as epitomized by the Unix diff tool. A diff example is given. The chapter is tied together by a final example of integrity detection in the operation of the GPS.