Spread-spectrum waveforms simplifying transform domain signal processing

In the paper a novel class of spread-spectrum waveforms is proposed which is based on spreading operations in the frequency domain. As a main advantage over conventional spread-spectrum waveforms, the proposed waveforms are especially well suited for transform domain correlation at low expense, because their Fourier spectra may be constructed from simple binary pseudonoise sequences. It will be shown that SNR degradation due to time and amplitude limiting requirements in practical spread-spectrum communication systems can be limited to about 1.2 dB by properly choosing the system parameters. The paper is primarily intended to initiate the development of new techniques for obtaining spread-spectrum waveforms in the frequency domain, and does not purport to give a complete recipe for design.