Research Background and Thesis Main Contents (Optical UWB Signals Generation Based on Lithium Niobate Electro-optic Modulator)
With the continuous development of communication technology, people’s requirements for communication are also rising. Wireless ultra wide band signals (UWB signals) get a lot of attention because of its high data transmission rate. In order to overcome the limitation of wireless ultra wide band signals transmission distance, UWBoF technology is proposed in recent paper, which is short for Ultra Wide Band Signal over Fiber. However, current research of generating a UWB waveform in optical circuits is more or less difficult and requires complicated devices. In my work, I propose a new method of generating doublet waveform by using a lithium niobate modulator with biased sinusoidal electric voltages applied on it. This method is mathematically proved feasible and experimental results also help to verify the correctness of my method.
The following electrical input vs optical intensity output picture shows the meaning of so-called “nonlinear region of a modulator” and the “linear region of a modulator”. Normally linear region is chosen to modulate electrical signals on the optical wave. In my work, nonlinear region can be used to generate doublet waveform by biased sinusoidal input.
The following pictures show the standard doublet UWB signal and its frequency domain distribution compared with the Frequency Mask (in fact, this waveform is the most ideal waveform act as a UWB signal). The same with other researches, the aim of my thesis is to generate this signal, but with theoretically high quality and high feasibility by a whole new method.
An Innovative New Method to Generate the Doublet UWB And Steps of It
With reference of the function of an input signal and the corresponding output signal in a lithium niobate electro-optic intensity modulator, an interesting fact is proposed by myself, which can be described as follows (pictures show the schematic diagram of the physical structure and the mathematical function ):
If the input signal satisfies: (1) Its waveform is symmetric. (2) Its intensity increases on the left side of the symmetry axis and then decreases on the right side. (3) Its intensity converges to a constant at infinity of the time axis. Then just by modifying the value of input intensity from the source and the value of bias voltage from the modulator, the output waveform in which there exists three extreme points can be generated. As the waveform of a doublet UWB signal also has three extreme points, these input electrical waveforms are potential of generating doublet signals. To make things easy, sinusoidal signal in one period is tested in the lab.
As the result that the waveform has three extreme points is not a sufficient condition to consider it as a UWB signal but rather a necessary condition, it is necessary to create a benchmark to further describe the doublet signal and find an approach to satisfy the benchmark for those output waveforms.
The benchmark is chosen to be the ratio of h1/h2, which is a constant for any doublet UWB signals shown in the following left picture. Luckily, this ratio can be achieved by applying a specific biased DC voltage to the input sinusoidal signal. Therefore, a sinusoidal input signal with a biased voltage generates a doublet UWB signal and input frequency of the sinusoidal signal is tunable to further control the width of the signal.
The following picture compares the output waveform and the standard doublet signal. From the result, there are some mismatch at the start and end of the signal, which may degrade the characteristics of the spectrum. However, considering that even the standard doublet cannot fully satisfy the FCC mask when the power gets the maximum, the degradation is tolerable.
Generate Signals in A Laboratory
Narrow band laser is chosen as a carrier wave to prevent chromatic dispersion and a function generator is chosen to create a periodical sinusoidal input signal. According to the conclusions above, the output should be periodical doublet signals.
The output waveform and its theoretical result shows in the following part. If noise is gotten rid of, they are quite consistent. In my paper, different DC voltages are applied to the sinusoidal input wave and the shape of the output changes correspondingly. These experiments give some clues of how to tune the DC voltage to find the proper value to generate a doublet.
