4684.新型宽带面谐平衡混频器设计与性能的影响 毕业设计(论文)英文原文

《4684.新型宽带面谐平衡混频器设计与性能的影响 毕业设计(论文)英文原文》由会员分享，可在线阅读，更多相关《4684.新型宽带面谐平衡混频器设计与性能的影响 毕业设计(论文)英文原文（14页珍藏版）》请在装配图网上搜索。

1、 DESIGN AND PERFORMANCE OF A NOVEL BROADBAND UNIPLANAR BALANCEDSUBHARMONIC MIXERABSTRACT: A novel wideband uniplanar balanced subharmonic mixer is proposed and demonstrated. The mixer is based on the broadband tapered coplanar-waveguide (CPW) DC-block filter, and the in-phase and out-of-phase T-junc

2、tions are interconnected through a pair of quarter-wavelength slotlines. The fabricated subharmonic mixer has an RF bandwidth of more than one octave from 12 to 28GHz with a conversion loss ranging from 8.5 to 12.5 dB for LO signals from 6.1 to 14.1 GHz with a power level of 9.5 dBm. The measured LO

3、-to-RF, LO-to-IF, and RF-to-IF isolations are better than 15.5, 34, and 26.6 dB, respectively. . 2005 Wiley Periodicals, Inc.Microwave Opt Technol Lett 44: 508 510, 2005; Published online in Wiley InterScience. DOI 10.1002/mop.20681Key words: balanced subharmonic mixer; broadband uniplanar technolog

4、y; coplanar waveguide; slotline; microwave integrated circuits1. INTRODUCTIONUniplanar technology has received considerable attention with regard to low-cost and high-performance microwave integrated circuits (MICs) and microwave monolithic integrated circuits (MMICs), especially in the millimeter-w

5、ave range 13. This is due to the merits not requiring via holes for connecting circuit elements to the ground, elimination of the backside processes, simple realization of series and shunt circuit components, easy integration with solid-state devices and other lumped elements,and low sensitivity to

6、substrate thickness 4. Furthermore, this technology allows flexible circuit design using mixed transmission lines such as coplanar waveguide (CPW), slotline, and coplanar stripline. Subharmonic mixing is very useful at millimeter wave frequencies due to the availability of low-cost lower-frequency l

7、ocal oscillators (LOs). Balanced subharmonic mixers have the advantages of the inherent LO-to-RF isolation and the reduction of LO noise and spurious signals 5.Recently, a few balanced subharmonic mixers using a uniplanar structure have been developed 6, 7. A subharmonic diode mixer using a uniplana

8、r hybrid with a double-Y balun was proposed for low-cost broadband millimeter-wave transceiver design 6. The hybrid consists of a combination of a tapered slotline T-junction and a tapered CPW divider interconnected through a slotline ring. The conversion loss of the mixer is about 10 to 12 dB over

9、a 2636-GHz RF bandwidth. A balanced subharmonically pumped diode mixer based on the modified slotline-to-CPW junction was also proposed 7. In this design, the hybrid was realized by the combination of twin-spiral CPW-to-slotline transition and modified slotline-to-CPW junction. The conversion loss v

10、aries from 10 to 12 dB when the RF signal is swept from 3 to 4 GHz and the IF frequency is 0.1 GHz. While these mixers are suitable for the realization of a compact uniplanar transceiver, they show relatively narrow bandwidth. In this paper, the design and measurement results of a uniplanar single b

11、alanced subharmonic mixer with broader bandwidth than that of conventional ones are presented. The proposed subharmonic mixer using CPW and slotline has also features of low conversion loss, small size, the ease of fabrication, and a simple structure for integrated microwave circuits.Figure1 Configu

12、ration of the proposed uniplanar balanced subharmonicmixerFigure2 Structure of the broadband tapered CPW DC-block filter2. SUBHARMONIC MIXER DESIGNThe configuration of the proposed broadband uniplanar balanced subharmonic mixer is shown in Figure 1. The circuit consists of a broadband tapered CPW DC

13、-block filter, a CPW radial-line double stub, an in-phase CPW-slot T-junction, an out-of-phase tapered slotline T-junction, a half-wavelength circumference slotline ring, two anti-parallel Schottky diode pairs, and a CPW-to-slotline transition. Bonding wires are used at nonsymmetrical CPW discontinu

14、ities such as T-junctions to suppress the undesired coupled slotline mode by maintaining an equal potential on each side of the ground plane. The balanced subharmonic mixing is based on a uniplanar magic T, which is composed of an in-phase parallel T-junction and an out-of-phase series T-junction in

15、terconnected through a pair of quarter-wavelength slotlines at the LO center frequency. Since these T-junctions are frequency independent, the magic T has inherent LO-to-RF/IF isolation and broadband characteristics. The RF signal is applied to the diodes through the DC-block filter and the in-phase

16、 CPW-slot T-junction. For DC and IF blocking at the RF port, the tapered CPW DC-block filter proposed in 8 is used with a tapered CPW section. The centerspacing between the inner slots of the open-end series stub linearly varies for wide band operation, as shown in Figure 2. The bandwidth of the DC-

17、block filter increases as the spacing W between the slots in the CPW is increased and the characteristic impedance of the CPW is decreased. The spacing W of 0.5 mm with 30_ characteristic impedance is used and the DC-block is interconnected with a 50_ CPW through 3-mm tapered CPW sections. The LO si

18、gnal is fed to the diodes via a CPW-to-slotline transition and an out-of-phase slotline T-junction. The slotline T-junction splits the LO signal into two signals that are of equal amplitude and 180 out of phases on the slotlines. The tapered slotline provides a match from the 50_ slotline to the two

19、 diodes at the LO frequencies. The CPW-to-slotline transition, which employs a slotline radial stub, is used to transfer a 50_ CPW to a 50_ slotline. The radius of the radial stub is 2.9 mm and the radial stub angle is 90. The measurements of the back-to-back connected transitions, formed using a pa

20、ir of CPW-to-slotline transitions separated by 2-mm-long slotline, exhibit insertion loss of less than 0.8 dB and return loss of more than 15 dB over 515 GHz.Two anti-parallel Schottky diode pairs (Alpha DMK2308-000) areflip-chip mounted across the gaps of the slotline ring with a silver epoxy. The

21、IF signal is extracted via a low-pass filter that employs a CPW radial-line double stub in order to provide an open circuit at the RF frequencies. A commercial microwave software package (Agilent ADS) is used to simulate and optimize the uniplanar structures and the subharmonic mixer by employing a

22、harmonic balance analysis following the initial design.3. EXPERIMENTAL RESULTSTo verify the proposed design principle, a broadband tapered CPW DC-block filter and a uniplanar single balanced subharmonic mixer with a 16-GHz RF bandwidth centered at 20 GHz were fabricated and measured. The circuits we

23、re built on a 25-mil-thick alumina substrate with a dielectric constant of 9.9 using a standard thin-film process. The mixer dimensions are 0.635 mm. The manufactured DC-block filter is measured using an HP8510C network analyzer and a probe station with a line-reflect-match (LRM) calibration. Figure

24、 3 shows the measured and simulated S-parameters of the fabricated DC-block filter and good agreement between two results is observed. The measurements show insertion loss less than 1.5 dB and return loss better than 10 dB over 1230GHz.The entire subharmonic mixer circuit is measured with respect to

25、 conversion loss and the RF-to-IF and LO-to-RF/IF isolations. Figure 4 shows the measured conversion loss and RF-to-IF isolation versus RF frequency for the LO signals from 6.1 to 14.1 GHz with a power level of 9.5 dBm and fixed IF frequency of 0.2 GHz.Over the RF frequency of 1228 GHz, the conversi

26、on loss is from .5 to 12.5 dB and the RF-to-IF isolation is better than 26.6dB.The conversion loss of the subharmonic mixer as a function of the LO power is shown in Figure 5. With the RF power level fixed at _10 dBm at 20 GHz and the LO at 10.1 GHz, the conversion lossbegins to saturate at around 8

27、 dBm. The measured LO-to-RF and LO-to-IF isolations are shown in Figure 6. The LO-to-RF and LO-to-IF isolations are better than 15.5 and 34 dB for the LO frequency from 6.1 to 14.1 GHz, respectively.4. CONCLUSIONA broadband uniplanar balanced subharmonic mixer using CPW and slotline has been propose

28、d and demonstrated. The fabricated subharmonic mixer has an RF bandwidth of more than one octave from 12 to 28 GHz with a conversion loss ranging from 8.5 to 12.5dB for the LO signals from 6.1 to 14.1 GHz with a power level of 9.5 dBm. The measured LO-to-RF, LO-to-IF, and RF-to-IF isolations are bet

29、ter than 15.5, 34, and 26.6 dB, respectively. The measurements of the tapered CPW DC-block filter show insertion loss of less than 0.8 dB and return loss of more than 15 dB over 1230 GHz. The proposed subharmonic mixer has the features of low conversion loss, small size, ease of fabrication, and a s

30、imple structure for integrated microwave and millimeter-wave circuits.REFERENCES1. I.J. Chen, H. Wang, and P. Hsu, A V-band quasi-optical GaAs HEMT monolithic integrated antenna and receiver front end, IEEE Trans Microwave Theory Tech 51 (2003), 24612468.2. K. Hettak, G.Y. Delisle, and L. Talbi, A 3

31、8-GHz integrated uniplanar subsystem for high-speed wireless broad-band multimedia systems, IEEE Trans Microwave Theory Tech 47 (1999), 935942.3. V. Trifunovic and B. Jokanovic, Review of printed Marchand and double Y baluns: Characteristics and application, IEEE Trans Microwave Theory Tech 42 (1994

32、), 14541462.4. K.C. Gupta, R. Garg, and I.J. Bahl, Microstrip lines and slotlines, 2nd ed., Artech House, Norwood, MA, 1996.5. S.A. Maas, Microwave mixers, 2nd ed., Artech House, Norwood, MA,1992.6. H. Gu and K. Wu, A novel uniplanar balanced subharmonically pumped mixer for low-cost broadband milli

33、meter-wave transceiver design, IEEE MTT-S Int Symp Dig (2000), Boston, MA, 635638.7. C.H. Wang, H. Wang, and C.H. Chen, A full-wave analysis model for uniplanar circuits with lumped elements, IEEE Trans Microwave Theory Tech 51 (2003), 207215.8. H. Gu, and K. Wu, Broadband uniplanar building blocks

34、for monolithic and hybrid millimeter-wave integrated circuits, 30th Euro Microwave Conf, 2000, pp. 395398. 2005 Wiley Periodicals, Inc.All-Optical Mixer Based on Cross-AbsorptionModulation in Electroabsorption ModulatorJianjun Yu, Senior Member, IEEE, Zhensheng Jia, and Gee Kung Chang, Fellow, IEEEA

35、bstractWe have proposed and experimentally demonstrated a novel method to realize nonlinear optical mixing based on crossabsorption modulation in an electroabsorption modulator. Our experimental results show that the local oscillator power and optical filtering play an important role on the receiver

36、 sensitivity of the up-conversion signal; furthermore, the wavelength span of larger than 20 nm for up-conversion signal can be obtained.Index Terms：Cross-absorption modulation, nonlinear optical mixer, radio-over-fiber, wavelength conversion.1.INTRODUCTIONPEOPLE are paying more attention to the app

37、lication of radio-over-fiber (ROF) for broad-band wireless access systems recently which are capable of providing the anticipated demand for future broad-band interactive services. To reduce the complexity of the architecture and meet more end-users between central station and base station (BS) at t

38、he same time, a solution to seamlessly integrate the wavelength-division-multiplexing (WDM) or WDM passive optical network transport systems with ROF access system to take full advantage of its ultrabandwidth characteristics is desirable. For a successful implementation of WDM ROF systems, all-optic

39、al up-conversion for WDM signals is the key issue to be solved 13. Currently, all-optical up-conversion based on nonlinear effects in nonlinear fiber or waveguide device needs high optical power and is polarization sensitivity. Although the polarization sensitivity can be reduced by adding some opti

40、cal components, it will greatly increase the configuration complexity. While based on cross-gain modulation in semiconductor optical amplifier (SOA) to realize all-optical mixing, the modulation frequency of the SOA usually is narrow, which is very difficult to realize data signal to mix with high f

41、requency local oscillator (LO) signal. Among them, cross-absorption modulation (XAM) in an electroabsorption modulator (EAM) could be one of the most promising ways to realize all-optical signal up-conversion or mixing similar to its wavelength conversion principle at high bit rate 47; the main diff

42、erence from wavelength conversion is that the modulated data signal will be used to replace the CW lightwave. Comparing it with other existing all-optical mixing methods, this scheme has some unique advantages such as lowpower consumption, compact size, polarization insensitivity, easy integration w

43、ith other devices, and higher speed operation due to EAM inherent characteristics 7. Here, we report 2.5-Gb/s data signal mixed with a 40-GHz LO signal based on XAM in an EAM for the first time.We investigate the up-conversion performance at different LO power and different wavelength span. Vestigia

44、l sideband (VSB) is used to reduce the carrier-to-sideband signal ratio (CSR) 8 and increase the receiver sensitivity.Fig. 1. Experimental setup. LNM: LN-modulator. TL: tunable laser. OC: optical coupler. PS: phase shifter. PD: photodiode. EA: electrical amplifier. LPF: low-pass filter. RX: receiver

45、. TA: tunable attenuator. The preamplifier noise figure is 6 dB.II. EXPERIMENTAL SETUP AND RESULTSThe experimental setup is shown in Fig. 1. An EAM (Cy- Optics model: EAM 40) with 3-dB bandwidth of 32-GHz fiber-to-fiber insertion loss of 8 dB and polarization sensitivity lower than 1 dB was used to

46、realize data signal up-conversion. A 2.5-Gb/s data signal was generated from a tunable laser at 1560.4 nm modulated by a LiNbO MachZehnder modulator driven by 2.5-Gb/s pseudorandom bit sequence electrical signal with a word length of 2-1. To generate a 40-GHz optical LO signal, we used carrier-suppr

47、essed return-to-zero signal, which is realized by driving a dual-arm LiNbO MachZehnder modulator biased at V with two complementary 20-GHz sinusoidal waveforms. The carrier suppression ratio is larger than 25 dB, the repetitive frequency of the generated LO optical signal is 40 GHz, and the duty cyc

48、le of the LO is 0.6. The optical LO signal was amplified before the LO signal and the data signal were injected into the EAM. The up-conversion signal was separated from the LO signal by using a WDM filter. Then an erbium-doped fiber amplifier (EDFA) was used to boost optical power before a tunable

49、optical filter (TOF1) was employed to suppress amplified spontaneous emission (ASE) noise and realize VSB filtering. The up-conversion signal after TOF1 was preamplified by a regular EDFA with a gain of 30 dB at small signal. In this experiment, two types of TOF1 with different bandwidth, 0.5 or 1.4

50、 nm, were employed. After amplification, the amplified up-conversion signal was filtered by another TOF (TOF2) with a bandwidth of 1.4 nm before being opticalelectrical converted via a PIN photodiode with a 3-dB bandwidth of 60 GHz. The converted electrical signal was amplified by a narrow-band elec

51、trical amplifier with a bandwidth of 10 GHz centered at 40 GHz. The amplified electrical signal and waveform is shown in the inset in Fig. 1. It is seen that the 2.5-Gb/s data and 40-GHz LO are well mixed. An electrical LO signal at 40 GHz was generated by using a frequency multiplier from 10 to 40

52、GHz. For simplification, we do not use a phase-locked loop to synchronize the clock and the signal in this experiment. We used the electrical LO signal and a mixer to down-convert the electrical millimeter-wave signal. The down-converted 2.5-Gb/s signal was detected by a bit-error-rate (BER) tester.

53、 Eye diagrams were recorded by a high-speed oscilloscope. Note: the above experimental results were obtained when the bandwidth of TOF1 is 1.4 nm and the center wavelength of TOF1 is the same as that of the optical carrier at 1560.4 nm.When the dc bias on the EAM is -3 V and the data signal at 2.5 G

54、b/s into the EAM is- 4 dBm, the optical power and the CSR of the mixed optical signal after the EAM as a function of the input power of LO signal are shown in Fig. 2. The measured static transfer curve slope of the EAM at -3-V bias is 3.5 dB/V. Due to the absorption effect of the EAM, the data signa

55、l was almost absorbed when the LO signal is smaller than 10 dBm. When the LO signal is larger than 10 dBm, the LO signal was mixed with the data signal due to XAM in the EAM. When the power of the LO signal is 12.5 dBm, CSR is smallest. The smallest CSR is 18 dB, and this value can be reduced when t

56、he bandwidth of the EAM is wider, such as larger than 40 GHz. Fig. 2 also shows that CSR will be reduced with the LO signal enhancement when the LO signal is larger than 12.5 dBm. The reason is that the LO signal will saturate the EAM, and our experimental results show that CSR could be further redu

57、ced as long as we increase the dc bias on the EAM. But larger dc bias will reduce the power and optical signal-to-noise ratio of the up-conversion signal. Therefore, in this experiment, the maximum bias on the EAM is -3 V.Fig. 3. BER curves with different LO powers. Insert eye diagrams: (i) mixedopt

58、ical signal (100 ps/div); (ii) down-conversion signal received at the receiver after the low-pass filter (100 ps/div). Fig. 4. Optical spectra of a filtered ROF signal. Resolution: 0.01 nm. CW: center wavelength.Fig. 3 shows the BER curves. When the LO power is 11.2 dBm, the receiver sensitivity at

59、a BER of 10-9 is -14.2 dBm, and error-floor will appear at10-10. This error-floor is caused by the fact that the optical receiver was saturated when a large carrier signal at 2.5 Gb/s was received 8, 9. When the LO signal was increased to 12.2 dBm, the receiver sensitivity at a BER of 10-9 was -23 d

60、Bm. Over 10-dB receiver sensitivity was increased. This result shows that this scheme for optical nonlinear mixing is sensitivity to the LO power. But if the LO signal is generated at the remote BS 1, 2, and they are put together with the EAM, the power of the LO signal will always be stable. Even i

61、f the receiver sensitivity is sensitivity to the LO power, this scheme can still be effective. The horizontal axis in Fig. 3 is the received optical power injected into the optical preamplifier as is labeled in Fig. 1. The measured eye diagram after up-conversion is inserted in Fig. 3 as inset (i) w

62、hen the LO power is 12.2 dBm and regular filtering is used. Evidently, a 40-GHz optical sinusoidal waveform was mixed into the 2.5-Gb/s signal, indicating that the 2.5-Gb/s signal was carried on a 40-GHz LO signal. The eye diagram of the down-converted signal at 2.5 Gb/s is also inserted in Fig. 3 a

63、s inset (ii) when the LO power is 12.2 dBm and regular filtering is used; a wide open and clean eye diagram is obtained.The optical spectrum of the up-conversion signal is shown in Fig. 4 when the dc bias on the EAM, the LO power and data signal power is- 3 V, 12.2 dBm and -4 dBm, respectively. This

64、 optical spectrum shows the CSR of the mixed signals is 19 dB and the mixed signals will occupy 80-GHz bandwidth to keep the three frequency tones, separated by 40 GHz when the LO power is 12.2 dBm. A high CSR will lead to weak receiver sensitivity, while a wide bandwidth will reduce spectral effici

65、ency f the WDM system. In this experiment, VSB filtering method was used to enhance the receiver sensitivity and simultaneously educe the occupied bandwidth. We replaced TOF1 by a new OF with a bandwidth of 0.5 nm. Tuning the center wavelength f TOF1, VSB filtering could be performed. We measured the receiver sensitivity and CSR of the mixed signals at different enter wavelength of TOF1, and the measured results are shown n Fig. 5. When the center wavelength of the TOF is 1560.4 nm, which is the same wavelength as the data signal,