The ratio between the discrete and continuous maxima of signals used in multicarrier communications is analyzed. An exact expression, up to an additive term, is obtained for the considered ratio when the number of uniformly distributed samples is equal to the number of carriers. A more accurate upper estimate on the continuous maximum from discrete samples of the signal and its derivative is suggested. Finally, a tight upper bound is given for the value of the continuous maximum via discrete values of samples when the number of samples is greater than the number of carriers.

This paper address the problem of classification of BPSK-modulated multicarrier signals with the peak-to-average power ratio (PAPR) exceeding 3/4 of the maximal possible value. By providing a simple characterization of such signals, an upper bound on their number are derived and propose a simple sufficient condition for the signals to have low PAPR. Based on this, a simple method for generation of signals with guaranteed PAPR gain of more than 2.5 dB using only two redundant tones are described.

We address the problem of classification of binary phase-shift keying-modulated multicarrier signals with the peak-to-average power ratio (PAPR) exceeding 3/4 of the maximal possible value. By providing a simple characterization of such signals, we derive an upper bound on their number, and propose a simple sufficient condition for the signals to have low PAPR. Based on this, we describe a simple method for generation of signals with guaranteed PAPR gain of more than 2.5 dB using only two redundant tones.