Keywords

29.1 Introduction

Ultra-wideband impulse radio has been receiving considerable interest of late as a highly promising solution for its applications in short-range, high-capacity wire-less communication systems and sensor networks because of advantages such as high data rate, low power consumption, and immunity to multipath fading. The Federal Communications Commission (FCC) of the U.S. has approved the unlicensed use of the UWB from 3.1 to 10.6 GHz with a power density lower than −41.3 dBm/MHz [FCC, part 15] [1–3]. In order to adapt to long distance transmission, a new type of UWB transmission system which use low loss optical fiber is proposed (UWB-over-fiber). The generation of UWB signal is the key technology in UWB-over-fiber system.

Recently the generation of UWB signal in the optical domain has attracted great interest by offering the advantages of low-loss and long-distance transmission of optical fiber. Many approaches to generating UWB signals in the optical domain have been reported [4–11]. An optical UWB pulse generator based on the optical phase modulation (PM) to intensity modulation (IM) conversion by use of an electrooptic phase modulator (EOPM) and FBG serving as an optical frequency discriminator was proposed and experimentally demonstrated [4]. Yao propose a scheme to generate UWB monocycle signals based on XPM of a semiconductor optical amplifier (SOA) [5]. Wang proposed UWB signals based on cross-gain modulation (XGM) of the SOA [6]. Dong proposed and demonstrated a simple scheme to generate UWB monocycle pulses based on the gain saturation effect of the SOA [7]. A method to generate UWB pulses based on chirp-to-intensity conversion using a distributed feedback (DFB) laser whose driving current is modulated by the electrical data signal was proposed [8]. Lin proposed UWB monocycle signals generated by using a gain-switched Fabry–Perot laser diode and a microwave differentiator [9]. An approach to generate UWB pulses based on an SOA and a electro-absorption modulator (EAM) [10]. A novel to generate UWB pulses by using a polarization modulator (PolM) and an FBG is proposed and experimentally demonstrated [11].

In this paper we propose a scheme to generate UWB monocycle pulse based on XPM of the DSF. According to the method, the phase modulation is achieved by simply using XPM of the DSF, without using electrooptic phase modulator. It significantly reduces the complexity of the system. By locating the cross-phase modulated probe at opposite linear slopes of the FBG reflection spectrum, UWB pulses with inverted polarity were obtained. This feature makes the implementation of the pulse polarity modulation (PPM) scheme possible in the optical domain, by simply switching the wavelength of the probe. The FBG, serving as a frequency discriminator, also serves as an optical bandpass filter to remove the residual pump and the amplified spontaneous emission (ASE) noise from the EDFA. It reduces the interference and improves stability of the system.

29.2 Operating Principle

The principle of UWB monocycle pulse generation based on XPM of the DSF is presented in Fig. 29.1. The pump light which emitted by LD1 propagates through PC is modulated by electrical Gaussian pulse in AM. To control the pulse width, a optical bandpass filter is incorporated after the AM. The pump light and a continuous-wave (CW) probe which transmitted by LD2 are applied to the EDFA to adjust the optical power via a 3 dB coupler. The two lights are then injected into a DSF that serves as the nonlinear medium, to achieve optical XPM. The phase modulated probe is then converted to an intensity-modulated pulse at an FBG. When the cross-phase modulated probe is located at the right linear slope of the FBG reflection spectrum, equivalent to first-order derivative of Gaussian pulse, the phase-to-intensity conversion is achieved, then the positive polarity UWB monocycle pulse can be obtained. If the cross-phase modulated probe is located at the left linear slope of the FBG reflection spectrum, the negative polarity UWB monocycle pulse will be got. This feature makes the implementation of the pulse polarity modulation (PPM) scheme possible in the optical domain. The FBG also serves as an optical bandpass filter to remove the residual pump and the amplified spontaneous emission (ASE) noise from the EDFA. It reduces the interference and improves stability of the system.

Fig. 29.1
figure 1

Principle of UWB monocycle pulse generation based on XPM of the DSF (LD1 laser diode1, LD2 laser diode2, PC polarization controller, AM amplitude modulation, OBF optical bandpass filter, OC optical coupler, EDFA erbium doped fiber amplifier, DSF dispersion shifted fiber, FBG fiber Bragg grating, PD photo diode, AMP amplifier, OSC oscilloscope)

Full size image

29.3 Simulation Design

The Simulation of UWB monocycle pulse generation based on XPM of the DSF is presented in Fig. 29.2. Simulation parameters set as follows: LD1 generate CW beams at 1,561.5 nm. Pseudo-random sequence is defined as “0000010000000000” (one “1”per 16 bits) at 13.5Gbit/s, the equivalent repetition rate is 0.84 GHz. The input signal pulse width is 0.50 bit. The central wavelength of the OBF is tuned at 1,561.5 nm, and its 3 dB bandwidth is 0.32 nm. The DSF has a chromatic dispersion of 5.6 ps/nm/km at 1,556 nm, a nonlinear refractive index of, and an effective mode-field area (MFA) of. The central wavelength of the FBG is tuned at 1,556.6 nm, its 3 dB bandwidth is 0.35 nm and the reflectivity of the FBG is 66 %.

Fig. 29.2
figure 2

Simulation of UWB monocycle pulse generation based on XPM of the DSF (BPG bit sequence pulse generator, GPG Gaussian pulse generator, LD1 laser diode1, LD2 laser diode2, PC polarization controller, AM amplitude modulation, OBF optical bandpass filter, OC optical coupler, EDFA erbium doped fiber amplifier, DSF dispersion shifted fiber, FBG fiber Bragg grating, PD photo diode, AMP amplifier)

Full size image

29.4 The Main Factors Affect the Generated UWB Signal

29.4.1 The Influence of the Signal Polarity

There are two opposite polarity in UWB monocycle pulses. The polarity of the UWB signal depends on the transfer function of the frequency discriminator and the wavelength of the probe. When the cross-phase modulated probe is located at the right linear slope of the FBG reflection spectrum, an ideal positive polarity UWB monocycle pulse can be got. The waveform and electrical spectra are presented in Fig. 29.3. The upper FWHM of the generated positive polarity UWB monocycle pulse is equal to 40 ps, lower FWHM is equal to 48 ps, the central frequency is 7 GHz, the fractional bandwidth is 143 %, Close to the FCC’s UWB template waveform and spectrum specifications. If the cross-phase modulated probe is located at the left linear slope of the FBG reflection spectrum, a negative polarity UWB monocycle pulse can be obtained. The waveform and electrical spectra are presented in Fig. 29.4. The upper FWHM of the generated negative polarity UWB monocycle pulse is equal to 49 ps, lower FWHM is equal to 43 ps, the central frequency is 6.95 GHz, the fractional bandwidth is 145 %, Close to the FCC’s UWB template waveform and spectrum specifications.

Fig. 29.3
figure 3

Waveform and spectra of the UWB positive pulse

Full size image
Fig. 29.4
figure 4

Waveform and spectra of the UWB negative pulse

Full size image

From the Fig. 29.3, 29.4, we can find that if the cross-phase modulated probe at opposite linear slopes of the FBG reflection spectrum, the UWB monocycle pulses with inverted polarity will be obtained. This feature makes the implementation of the pulse polarity modulation scheme possible in the optical domain, by simply switching the wavelength of the probe. As shown in Figs. 29.3 and 29.4, the waveforms of the generated UWB monocycle pulses show a little asymmetry. For the UWB monocycle pulses with inverted polarity, a slight asymmetry is observed, which is due mainly to the self phase modulation (SPM) and cross-gain modulation (XGM) in the DSF, which leads to the pulse broadening and distortion.

29.4.2 The Influence of the Input Pulse Width

The input signal width is an important parameter to affect the waveform of the generated UWB monocycle pulse signal. Keeping the other parameters fixed, the input signal pulse width is set at 0.50, 0.75 and 1.00 bit respectively. As shown in Fig. 29.5a, c and e, when the input signal pulse width is increasing, the generated UWB monocycle pulse signal upper/lower pulse width is increasing, the waveform distortion of the UWB signal is more serious. As shown in Fig. 29.5b, d and f, when the input signal pulse width is increasing, high frequency components of the RF signal power is becoming lower, the bandwidth of the generated UWB monocycle pulse signal is becoming narrower. When the input signal pulse width is set to 0.50, 0.75 and 1.00 bit, the center frequencies of the relative UWB pulse signals are 7, 5.35 and 4 Ghz, and the relative bandwidths are 143, 166 and 150 %, they all meet the criterion of FCC.

Fig. 29.5
figure 5

Waveforms and RF spectrum for generated monocycle pulse in cases of different input signal width

Full size image

The results show that this scheme has good tolerance to the input signal pulse width. However, when the pulse width increases to a certain level, the waveform distortion of the UWB signal becomes serious. This is because as the pulse width increases, the XGM and SPF effect in DSF will further increase, the waveform distortion of the UWB signal will become more serious. Therefore, when the input pulse width is set at 0.50 bit, an ideal UWB monocycle pulse signal can be got.

29.4.3 The Influence of the Reflectivity of FBG

The FBG is serving as a frequency discriminator, the reflectivity of the FBG is an important parameter to affect the waveform of the generated UWB monocycle pulse signal. Keeping the other parameters fixed, the reflectivity of the FBG is set at 30, 50, 66 and 80 % respectively. As shown in Fig. 29.6a, b, c and d, the generated UWB monocycle pulse signal upper/lower FWHMs are 33 and 43 ps, 37 and 46 ps, 40 and 48 ps, 36 and 51 ps respectively. As shown in Fig. 29.6, when the FBG reflectivity increases, the upper amplitude of the generated UWB monocycle pulse signal will decrease, the lower amplitude will increase; the upper FWHM of the generated UWB monocycle pulse signal will first increase to a certain level and then decrease, the lower FWHM will increase.

Fig. 29.6
figure 6

Waveforms for generated monocycle pulse in cases of different FBG reflectivity

Full size image

The results show that when the reflectivity of the FBG is set at 66 %, the lower amplitude is close to upper amplitude, the difference between upper amplitude and lower amplitude is smaller, the generated UWB monocycle pulse signal shows good symmetry. Therefore, when the reflectivity of the FBG is set at 66 %, an ideal UWB monocycle pulse signal can be got.

29.4.4 The Influence of Using Different Frequency Discriminators

In theory, any filtering device can be used as the frequency discriminator, but using different frequency discriminators will influence the performance of the generated UWB monocycle pulse signal. In this scheme, we adopt the commonly used commercial optical filters as frequency discriminators. Keeping the other parameters fixed, optical Gaussian band-pass filter, DWDM and FBG are used as the frequency discriminators respectively. The central wavelength of the DWDM channel which is used to frequency discriminate is tuned at 1,556.6 nm and 3 dB DWDM channel bandwidth is 0.35 nm. The central wavelength of the FBG and optical Gaussian band-pass filter are also tuned at 1,556.6 nm and their 3 dB bandwidths are 0.35 nm. In the simulation, optical Gaussian band-pass filter, DWDM and FBG are used as the frequency discriminators to frequency discriminate respectively. As shown in Fig. 29.7a, b and c, the generated UWB monocycle pulse signal upper/lower FWHMs are 39 and 42 ps, 38 and 40 ps, 40 and 48 ps respectively. We can find that when optical Gaussian band-pass filter and DWDM are used as the frequency discriminators, the generated UWB monocycle pulse signals are almost the same; when FBG is used, the waveform of the generated UWB monocycle pulse shows a little different, for the waveform of the generated UWB pulse signal can be adjusted by changing the reflectivity of FBG.

Fig. 29.7
figure 7

Waveforms for generated monocycle pulse in cases of different frequency discriminators

Full size image

The results show that optical Gaussian band-pass filter, DWDM and FBG can be used as the frequency discriminators, achieving the conversion from PM to IM. FBG has more advantages due to the flexibility to adjust the waveform of the generated UWB pulse signal by changing the reflectivity of FBG.

29.5 Conclusion

A method of optical ultra-wide-band pulse generation based on XPM in a dispersion shifted fiber is proposed. According to the method, the phase modulation is achieved by simply using XPM of the DSF, without using electrooptic phase modulator. It significantly reduces the complexity of the system. By locating the cross-phase modulated probe at opposite linear slopes of the FBG reflection spectrum, UWB pulses with inverted polarity were obtained. This feature makes the implementation of the pulse polarity modulation (PPM) scheme possible in the optical domain, by simply switching the wavelength of the probe. The FBG, serving as a frequency discriminator, also serves as an optical bandpass filter to remove the residual pump and the amplified spontaneous emission (ASE) noise from the EDFA. It reduces the interference and improves stability of the system. By using the software of Optisystem, the impacts of the input signal pulse width, the reflectivity of FBG and using different frequency discriminators on the generated UWB monocycle pulse are numerically simulated and studied. The results show that this scheme has good tolerance to the input signal pulse width. Optical Gaussian band-pass filters, wavelength division multiplexer and FBG can be used as the frequency discriminators. FBG has more advantages due to the flexibility to adjust the waveform of the generated UWB pulse signal by changing the reflectivity of FBG. The XGM and SPF effect in DSF can influence the simulation results, and how to minimize the influence and waveform distortion is the focus of our future research.