?
The Evaluation of Gain of Statistical Modulation Method on the Example of QAM16 for Input Data with Exponential Distribution
This paper concerns new effective method of joint data coding/modulation which may improve energy-efficiency and energy savings of modern wireless transmission systems. The method require a priori knowledge of probability distribution of input data to map them to the modulation symbols in the most efficient way.
The key idea of the proposed methods of Statistical Modulation is to map the most frequent input values into the modulation symbols with the lowest energy. To estimate the benefit we apply the approach to well-known Quadrature Amplitude Modulation (QAM): the most frequent input symbols are mapped to the most frequent QAM constellation pints. As the result, an average energy needed for data transmission is much smaller that allows increasing the distance between QAM constellation points for the same average energy. Therefore better Bit-Error-Rate (BER) is achievable for the same Signal-To-Noise-Ratio (SNR) in comparison with the standard QAM that does not utilizes the probabilities of input symbols.
In our research we have compared new SQAM and traditional QAM modulation (which does not utilizes the probabilities of input symbols) for the case of exponential distribution of input symbols. Our experiments and theoretical calculations shows that SQM for exponential input provides up to 3 dB gain in BER-SNR.
This method may be applied to improve BER-SNR and reduce the power consumption of the whole transmission system. The list of potential application areas includes M2M communications, IIoT, mobile networks and other scenarios that are critical to power consumption, battery life and latency.