Triangular modulation technique for digital data transmission

Abstract

This paper presents a novel technique for digital data modulation and demodulation called triangular modulation (TM). The modulation technique was developed primarily to maximize the amount of data sent over a limited bandwidth channel while still maintaining very good noise rejection and signal distortion performance. The modulation technique involves breaking digital data into a series of parallel words. Each word is then represented by one half period of a triangular waveform whose slope is proportional to the value of the parallel word it represents. The demodulation technique for this uniquely defined waveform involves first digitizing the waveform at a higher constant sampling rate. A linear regression algorithm using the method of least squares is then used to compute the slope of the digitized waveform to a very high precision. This process is repeated for each rising and falling edge of the triangular modulated waveform. All encoded data is extracted by precise slope computation since each slope uniquely defines the encoded data word it represents. The ability of the demodulation algorithm to compute the exact slope of the modulated waveform determines how many bits can be represented by the modulated waveform. Transmission channel bandwidth limitations determine the allowable range of slopes used. Several simulations are performed to provide a sample of how the modulation method will perform in various real world environments. The paper also discusses several application areas where the modulation technique will provide superior results over other modulation methods.