Spectrally Efficient Continuous Phase Modulation
Spectrally efficient continuous phase modulation (CPM) systems and generalizations of CPM for wireless links are considered. Continuous phase modulation defines constant envelope phase codes, which are advantageous when using non-linear amplifiers. The channel is an additive white Gaussian noise (AWGN) channel and the spectrum requirement is a mask, which puts stringent requirements on the spectral main lobe and the spectral side lobes.
The thesis covers complexity and power efficiency of spectrally efficient coded modulation systems with zero or a small envelope variation. Reduced state sequence detection and approximate metrics are used to simplify the maximum likelihood sequence detector. Modulation system parameters are numerically optimized in order to probe the full potential of the systems and we build an empirical model for the optimum parameters of CPM based on numerical results. The optimized schemes also improve the sensitivity for a detector based on reduced state sequence detection.
We study serially concatenated convolutionally encoded CPM for spectrally efficient systems, and we generalize multi-level CPM to a system called multi-pulse continuous phase modulation (mpCPM). We show that convolutional encoding does not improve the system, but mpCPM can gain up to 0.7 dB over CPM. We also illustrate that mpCPM can provide unequal error protection. Then, we develop a modulation system called constrained envelope continuous phase modulation (ceCPM) , which exhibit a small controlled envelope variation. We show design examples where the performance gain for ceCPM over CPM is up to 2.5 dB.
additive white Gaussian noise (AWGN)
trellis coded modulation (TCM)
continuous phase modulation (CPM)
maximum likelihood sequence detection (MLSD)