Applying a direct random spreading sequence to a digital information sequence prior to multi-h modulation creates a new class of signals called spread-spectrum multi-h (SSMH) signaling. By spreading a known bandwidth efficient modulation scheme, the power spectral density is controlled so that the transmitted spectrum will have a wide, flat mainlobe and rapid sidelobe rolloff. Coincidentally, the power efficient modulation allows transmission at lower signal-to-noise ratios when the receiver knows a priori the spreading sequence and modulation index sequence. Optimal receiver structures are derived and numerically evaluated in an additive white Gaussian noise environment. It is shown that performance is dependent on the spreading sequence, the modulation indexes, and the possible phase states and can exceed direct sequence binary phase-shift keying by 1-2 dB at bit error rates of 10/sup -5/. Composite likelihood ratio analysis reveals a reduction of 50-70% in detectability of completely known SSMH signals vis-a-vis direct-sequence binary phase shift keying (DS/BPSK) at error rates of 10/sup -4/.>

The effectiveness of antenna and spread-spectrum diversity is evaluated for direct-sequence differential phase-shift-keyed signaling over a worst-case pulse-jammed multipath-fading channel. Two diversity combining techniques are proposed. One uses predetection selective combining followed by postdetection equal-gain combining, and the other uses only postdetection equal-gain combining. The use of coding with hard and soft decision decoding is also considered. It is shown that the proposed diversity techniques can provide a significant improvement in the performance of coded antijam systems operating over jammed multipath-fading channels.>

A statistical correlation study between 18 objective quality measures and a data base of subjective quality measures from the Paired Acceptability Rating Method (PARM) was done for nine communication systems, including waveform coders, channel vocoders, linear predictive coders, and adaptive predictive coders. The results of this study show which of the candidate objective measures are most effective in predicting the subjective results. The measure which was found to be most effective over all systems was a gain weighted L2spectral distance metric which had a correlation coefficient of -.83. Supported by DCA/DCEC via the RADC Post Doctoral Program.

The purpose of this study was to assess the effects of LPC/CVSD tandem connections and to investigate ways of improving performance. In the case of the CVSD-to-LPC connection, the CVSD quantizing noise severely affects the estimate of LPC coefficients, thereby distorting the spectral representation. An averaging technique was shown to reduce the average spectral distance from the noiseless case; however, no statistically significant perceptual advantage was measured. In the LPC-to-CVSD connection, it was found that modest improvements can be measured both in signal-to-noise ratio and perceptually, by modifying the phase in LPC synthetic speech before CVSD coding.