Abstract: Random data is derived from the value of the elapsed time over which a loop is counted on a multitasking computer. The random data can be manipulated and whitened to produce random numbers of arbitrary length that pass both chi-squared and compression tests for randomness. The present invention does not rely on devices such as noise generators, leaky capacitors, gas discharge tubes, or keyboard strokes to generate random data, and reliably generates random data at a substantially higher rate than other systems and methods that can be implemented using only a general purpose computer executing instructions stored in computer readable memory. The present invention is advantageously portable among UNIX-based platforms, and operates without occupying UNIX timers or signals.

Abstract: A system and method for determining if a received identifier could be equivalent to one of a set of predetermined identifiers. Predetermined identifiers are encoded into an output. The coded output is stored on a node in a network. An identifier is received over the network and is tested against the coded output. The test reveals if the received identifier could be equivalent to one of a set of predetermined identifiers encoded into the output. An action can be taken that depends upon the outcome of the test. The present invention is useful in efficiently routing messages in a network, determining the authorization of user requests, and other applications.