Chaotic signals for digital communication

Ghobad Heidari-Bateni, Purdue University


A new class of signals for digital communication that shows promise in becoming a very desirable waveform in digital communication systems is introduced. These signals originate from the theory of chaos in non-linear dynamical systems. The distinct random-like behavior of these signals in a completely deterministic setting has proven to be very useful in digital communications. The correlation properties of these signals are shown to be identical to those of the random noise signals. Two applications of these signals are found and investigated. The first one has to do with a non-coherent multiple-access communication system that takes advantage of the noise-like appearance of the chaotic signals to hide the modulation in noise. The second application involves the use of chaotic sequences as spectral spreading sequences in direct-sequence spread-spectrum (DS/SS) communication systems. The error probabilities of such systems are investigated and shown to be, for all practical purposes, identical to the conventional DS/SS systems that use binary signature sequences. One of the advantages of the use of chaotic sequences in spreading the DS/SS waveforms is that there is a greater number of them available to use. Also, their generation is very simple, requiring very limited amount of stored information. Furthermore, the security of transmission via Chaotic DS/SS waveforms is shown to improve tremendously compared to the Binary DS/SS systems, both in civilian and in military applications. The interception methods conventionally used against DS/SS systems are examined. Some of these methods are very effective on Binary DS/SS systems, while the proposed Chaotic system is shown to be very resistant against them.




McGillem, Purdue University.

Subject Area

Electrical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server