Address Scrambling Methods

Address scrambling maps one IP address to another IP address. There are a number of methods to scramble addresses.

1.

the sequential numbering method maps each IP address occurrence to a sequential number. Although this method is easy to understand, it is difficult to preserve other meanings of addresses.

2.

the hash method maps an IP address to another IP address using a hash function in order to provide random mapping. It is also possible to preserve the common address prefix between 2 addresses by maintaining an ordered tree of addresses similar to a routing table. In this method, if 2 IP addresses have a common address prefix, they are mapped to addresses with a common address prefix of the same length. Note that, although it preserves routing information, this method has a risk of being reverse-engineered. For example, one can use a well-known server's address as a clue to de-scramble the address prefix [Ylo96]. The impact of this threat, however, depends on the importance of hiding the network topology.

There are several choices regarding address consistency between two or more data sets.

1.

all occurrences of an address are to be mapped to a single address within a data set.

2.

all occurrences of an address are to be mapped to a single address across different data sets.

Longer consistency is convenient for analysis but it also makes reverse-engineering easier.