適用於超寬頻通訊系統與新型感知通訊之射頻前端電路設計

Abstract

寬頻系統是現在通訊系統的趨勢，本次為期三年的研究計畫將提出符合超寬頻(UWB)通訊與感知通訊(Cognitvie Radio)的射頻前端電路，分述如下。 一．超寬頻(UWB)系統為操作在3.1~10.6 GHz 的頻段中，即約有7.5 GHz 的頻寬。由於具有相當大的通道容量，所以UWB 在無線個人網路的應用上，能夠達到高速資料傳輸的需求(＞100Mb/sec)，因此這會是一個在應用上相當具有吸引力的通訊系統。在目前提出的超寬頻系統， 架構主要可分為： (a) DS-CDMA 和(b)MB-OFDM。其中MB-OFDM 的架構是將頻譜分成14 個次頻帶，而利用時間交錯方式輪流用多個次頻帶傳送符碼的主要優點，在於UWB 系統可以傳送同樣的平均功率並且瞬間處理頻寬較小(528 MHz)，如此可以增加頻譜的使用彈性和其他無線通訊系統的相容性，並同時降低功率消耗及成本。所以MB-OFDM UWB將為本計劃所探討的架構。 二．另外，依據聯邦通訊委員會（FCC）[0-1]，隨著時間與位置的變化，以分配的頻譜的使用率介於15%至85%之間，如圖0-1．隨著通訊服務的巨幅增加，現今有限的頻譜已經不敷使用．所以能夠更有效運用頻譜的感知通訊技術(cognitiv radio)[0-2],[0-3] ,在最近被提出以有效解決上述問題。 圖0-1 Spectrum utilization 此動態頻譜存取的技術(Dynamic Spectrum Access Networks)，也就是感知通訊（cognitive radio），會透過異質網路(heterogeneous wireless architectures)與動態的頻譜存取技術來提供移動中的使用者（mobile users）很寬頻的服務．而所謂動態頻譜技術即是在不影響已分配頻段(licensed band)的使用者前提下，機會性的在此頻譜作存取的動作．如圖0-2所示，也就是在不同時間下，系統會在寬頻的頻譜內偵測到低使用率的頻段進行利用。 圖0-2 動態存取示意圖 由於感知通訊（cognitive radio）的收發機架構裡基頻( baseband )的電路架構與現在的收發基架構類似，所以其收發機重點在於前端電路的設計，因為前端電路須能在寬 頻下將操作頻段操作至任何所需要的頻譜位置．所以感知通訊（Cognitive Radio）的系統會需要寬頻低雜訊放大器與寬頻混波器。 第ㄧ年研究計畫：以0.18um CMOS製程設計多頻帶之頻率合成器(Multi-Band Frequency Synthesizer) 應用在MB-OFDM UWB傳輸架構的頻率合成器，必須具有非常快速的切換時間(＜9.5ns)，以及在3~10 GHz 附近都能有好的相位雜訊表現(＜-86.5 dBc/Hz @ 1MHz) [1-1]，所以傳統頻率合成器必須加以修改以符合此系統之需求。目前在已提出的架構中，頻率合成器的實現可分為三種形式： Type 1－利用多個頻率合成器並聯在一起，每個頻率合成器各自產生所需要的頻 率[1-2],[1-3]。 Type 2－利用一個包含多頻帶壓控震盪器的頻率合成器來產生每個頻帶需要的 頻率[1-4],[1-5]。 Type 3(第二年研究計畫提案:)－頻率合成器產生某一特定頻率後，再利用多個混波器和濾波器來產生各個頻率[1-6~1-9]。 由於Type1中的架構有著較高的面積和功率消耗，因此本計畫中將不考慮Type 1 這種架構，只對Type 2 和Type 3 分別進行探討分析其優缺點。 在Type 2 的架構下，頻率合成器幾乎可以用傳統的鎖相迴路來實現，不過 在壓控震盪器的部份必須能是多頻帶的輸出，而且為了達到9.5 ns 內的切換時 間，必須得有很高的參考頻率以及用到兩個鎖相迴路來提供夠快的鎖定時間。不 過這架構的最大優點在於不需要利用混波器來產生多頻帶輸出，可以節省面積和 功率，並且消除混波器產生的spurious noise。由於[1-4]中提出的架構依然有著功率消耗過大的問題，因此本實驗室對此缺點提出改良的架構，如Fig.0-3所示，並將發表在Progress in Electromagnetics Research Symposium( PIERS2007 ) [1-5]使得整個電路更能有效率利用每個block。 Figure.0-3 第ㄧ年的計畫中,基於在Type 3 的架構下，由頻率合成器產生一特定頻率後，再利用混波器和濾波器產生各個頻率，如Fig.0-4所示。由8448 MHz VCO 設計為起點，利用除二電路與混波器分別達成6336 MHz、4224 MHz 與2112 MHz 的頻率輸出，再經過Switch決定某一特定頻率輸出，而外加較低頻訊號(264 MHz、792 MHz、1320 MHz 與1848 MHz)也由另一Switch 決定另外一特定頻率輸出，最後經由混波器與High-pass filter 即可得到其中一個頻帶的LO 信號，Fig.0-5所示。此架構優點在於只需一個頻率合成器，即可達到全頻帶輸出，因此有更低的功率消耗。並且頻率切換時間只跟Switch 有關，在現今製程中的選擇器切換時間大都小於1 ns的情況下，此頻率合成器不會有來不及完成頻率切換的問題。 Figure.0-4 Figure.0-5 第二年研究計畫：微機電製程研究與實現微機電開關以符合未來寬頻感知通訊( Cognitive Radio)系統的需求。 MEMS Antenna:傳統單極天線如果要增加頻寬需要擴大天線的寬度。在1990年時，Nair提出一寬頻微帶天線[2-4] ，其利用耦合形式產生電容效應以達到寬頻的效果，但如果將這類方法運用在單極天線上，會因天線本體下方無接地面而大大減少了耦合效應，因此我們再結合MEMS技術設計3-D架構之天線，此技術主要將天線元件間蝕刻以產生的凹型槽孔，並在深凹槽內電鍍銅金屬，以增加元件與元件間的耦合效應，如此不僅能避免增加天線的面積，亦能提升天線的阻抗頻寬． 第三年研究計畫：以0.18um CMOS製程並以新穎的阻抗匹配方式設計 寬頻低雜訊放大器(Broadband LNA)與寬頻混波器(Broadband Mixer)以符合未來寬頻感知通訊(Cognitive Radio)系統的需求。 1.寬頻低雜訊放大器（Broadband LNA）:在3-15GHz提供20dB的順向增益 (Gain),-20dB的返回損失(Return Loss), 與3dB的雜訊指數(NF)． 2.寬頻混波器(Broadband Mixer):在3-15GHz提供10dB轉換增益(Conversion Gain)，IIP3約3dB，功率消耗(Power)約10mW，雜訊指數(NF)約7dB． 在這第三年的計畫，文獻[3-6]中，提出利用電容與電阻性的負載，如下圖所示， 可以達成寬頻的匹配，而不需要在輸入端加上複雜的被動匹配網路，因此可在寬頻的狀況下達到很低的雜訊指數。但此方式的匹配往往需要大面積的電晶體且在汲極端會需要串聯大電感作為直流路徑同時作為Rf Choke，基於此原因我們使用Current reuse組態以降低功率損耗，並避免使用大電感。所以我們提出如圖3-28的架構：適當設計負載的被動網路，預期可達到寬頻輸入阻抗匹配，同時達到寬頻低雜訊的要求。 圖3-28 第三年計畫LNA 圖4-9為我們第三年計畫的第二個電路，我們亦將此概念應用在混波器的轉倒級(conductance stage)，所以混波器可省略複雜的被動匹配網路，使得寬頻狀況下達到低雜訊指數，同時利用此Current reuse架構降低功率損耗與增加線性度，並利用源極退化( Soucre Degeneration)的觀念進一步增加線性度。 圖4-9. 第三年計畫Mixer [0-1] FCC, ET Docket No 03-222 Notice of proposed rule making and order, December 2003. [0-2] DARPA XG WG, The XG Architectural Framework V1.0, 2003. [0-3] DARPA XG WG, The XG Vision RFC V1.0, 2003. [1-1] C. Mishra, A. Valdes-Garcia, F. Bahmani, A. Batra, E. Sanchez-Sinencio, J. Silva-Martinez,“Frequency Planning and Synthesizer Architectures for Multiband OFDM UWB Radios,” IEEE Transactions on Microwave Theory and Techniques : Accepted for future publication, Volume PP, Issue 99, 2005 [1-2] B. Razavi, T. Aytur, F.-R. Yang, R.-H. Yan, H.-C. Kang, C.-C. Hsu, and C.-C. Lee, “A 0.13 um CMOS UWB transceiver,” IEEE Int. Solid-State Circuits Conf. Tech. Dig., Feb. 2005. [1-3] A. Medi and W. Namgoong, “A fully integrated multi-output CMOS frequency synthesizer for channelized receiver,” Proc. IEEE Int. System-on-Chip Conf., Sep. 2003 [1-4] T. Geum-Young, H. Seok-Bong, K. Tae Young, C. Byoung Gun, P. Seong Su, “A 6.3-9-GHz CMOS fast settling PLL for MB-OFDM UWB applications,” IEEE J. Solid-State Circuits, Volume 40, Issue 8, Aug. 2005 [1-5] Shih-Hao Tarng, Yu-Ching Tsai and Christina. F. Jou, “A Fully-Integrated,Low Power, Fast-Locking, Integer-N Frequency Synthesizer for MB-OFDM UWB System,” to be published inPIERS 2007 [1-6] D. Leenaerts, R. van de Beek, G. van der Weide, J. Bergervoet, K. S. Harish, H. Waite, Y. Zhang, C. Razzell, and R. Roovers, “A SiGe BiCMOS 1 ns fast hopping frequency synthesizer for UWB radio,” IEEE Int. Solid-State Circuits Conf. Tech. Dig., Feb. 2005. [1-7] J. Lee and D. Chiu, “A 7-band 3–8 GHz frequency synthesizer with 1 ns band-switching time in 0.18 um CMOS technology,” IEEE Int. Solid-State Circuits Conf. Tech. Dig., Feb. 2005. [1-8] Lawrence Williams, Daniel Wu, Eldon Staggs, Albert Yen, “Ultra-Wideband Radio Design for Multi-band OFDM 480 Mb/s Wireless USB,” DesignCon 2005. [1-9] Sandner, C., Wiesbauer, A., “ A 3GHz to 7GHz fast-hopping frequency synthesizer for UWB,” Ultra Wideband Systems, 2004. Joint with Conference on Ultrawideband Systems and Technologies. Joint UWBST & IWUWBS. 2004 International Workshop on 18-21 May 2004 Page(s):405 - 409 Digital Object Identifier 10.1109/UWBST.2004.1321005 [2-4] C. K. Aanadan, P. Mohanan, and K. G. Nair, “Broad-band gap coupled microstrip antenna,” IEEE Transactions on Antenna and Propagation, Vol. 38, No. 10, Oct. 1990. [3-6] Wide-Band Matched LNA Design Using Transistor＇s Intrinsic Gate–Drain Capacitor

Wideband system is a trend of communication system, this research by three years will find out a solution of RF front end matching UWB and Cognitvie Radio, it is described as following. 1) Ultra Wide Band (UWB) system is operated between 3.1 to 10.6 GHz, i.e. bandwidth with 7.5 GHz. The application of wireless personal network with UWB can fit the requirment of high speed data transmission (> 100Mb/sec) since the channel capacity is so large, this will be very attractive for the application of a communication system. The main research of UWB system can be divided into two parts : (a) DS-CDMA and (b) MB-OFDM. MB-OFDM will be discussed in this research because of its advantages that MB-OFDM divides the main band into 14 sub-bands, using time-interlace to transmit the data on these sub-bands, and MB-OFDM can transmit the same average power but with the smaller immediate-processing bandwidth (528 MHz) than the DS-CDMA. These increase the flexibility of the bandwidth’s usage and the compatibility of wireless communication system, decreases the cost and power consumption. 2) In addition, a large portion of the assigned spectrum is used sporadically as illustrated in Fig. 0-1, where the signal strength distribution over a large portion of the wireless spectrum is shown. The spectrum usage is concentrated on certain portions of the spectrum while a significant amount of the spectrum remains unutilized. According to Federal Communications Commission (FCC), temporal and geographical variations in the utilization of the assigned spectrum range from 15% to 85%. Although the fixed spectrum assignment policy generally served well in the past, there is a dramatic increase in the access to the limited spectrum for mobile services in therecent years. Fig.0-1 Spectrum utilization The Dynamic Spectrum Access Networks, as well as cognitive radio, will serve the mobile users with wideband via heterogeneous wireless architectures and dynamic spectrum access techniques, but this dynamic spectrum access is working under the condition that it won’t interferer the users of the licensed band over the broad bandwidth, as Fig. 0-2. Fig.0-2 Spectrum Hole concept The novel characteristic of cognitive radio transceiver is a wideband sensing capability of the RF front-end. This function is mainly related to RF hardware technologies such as wideband antenna, low noise amplifier, and adaptive filter. RF hardware for the cognitive radio should be capable of tuning to any part of a large range of frequency spectrum. Also such spectrum sensing enables real-time measurements of spectrum information from radio environment. In the second year and third year of this proposal, we propose a novel broadband MEMS Antenna and broadband LNA and broadband Mixer. The First Year Proposal: Multi-Band Frequency Synthesizer for UWB synthesizer Frequency synthesizer for MB-OFDM UWB has two specifications: very fast locking time (<9.5 ns) and moderate phase noise in the whole bandwidth (< -86.5 dBc/Hz @ 1 MHz). So the traditional frequency synthesizer need some modification to meet these specifications. Three types of frequency synthesizer for UWB have been proposed. Type 1－Combination of several frequency synthesizers which have different bands from each other. Type 2－Using a frequency synthesizer including a multi-band VCO to generate all specified frequencies. Type 3( The first-year proposal )－Using a frequency synthesizer and several mixers to generate all specified frequencies. Due to larger area and higher power consumption in Type 1, it will be discarded in this project. Although Type 2 have many advantages, it still comsumes large power. In view of the above disadvantage, we propose a synthesizer, as shown in Fig.0-3, and will be published in Progress in Electromagnetics Research Symposium ( PIERS2007 ). Only Type 3 will be researched and implemented in this project. Figure.0-3 A 12 bands synthesizer for UWB communication operating from 3 to 10 GHz is presented. The architecture consists of a wideband quadrature voltage-controlled oscillator (QVCO), 2-stage dividers, switch buffer and only one quadrature single-sideband (SSB) mixer. This work will be designed with 0.18-μm CMOS technology, the expected QVCO’s measured phase noise will be less than -98 dBc/Hz at 1 MHz offset, and the expected output powers of twelve bands will be better than 35 dB sideband rejection while consuming 60.76mW of the core circuit and 52.93mW of the buffer from a 1.8-V supply. The expected switching time for hopping frequency will be about 1ns. The Second Year Proposal: MEMS Process and Broadband Antenna In general, the impedance bandwidth of traditional monopole antenna can be enhanced by increasing the width of monopole antenna. In 1990, a broad-band gap coupled microstrip antenna was proposed by Nair [4], the gap coupled microstrip antenna improved the capacitance of the microstrip patch antenna to increase the impedance bandwidth. If the method is utilized on the monopole antenna, the coupling effect can be decreased greatly. In this project, the microelectromechanical system (MEMS) technology would be used to design the 3-D antenna, the MEMS technology would etch the substrate between the elements of the antenna to make the U-shape slot, as shown in Fig.2-9, and plate with copper at the U-shape slot. So the coupling effective and the impedance bandwidth will be increased. Fig.2-9 The Third Year Proposal: 0.18um CMOS Broadband LNA and Broad Mixer for the New Cognitive Radio It was found that the capacitive and resistive loading will achieve input wideband input matching. But it will consume large power. In the third year proposal, a novel CMOS Broadband LNA will be propose by applying current reuse topology and a passive loading, as shown in Fig.3-28 . The expected power consumption will be small and achieves Broadband simultaneous noise and impedance matching. Fig.3-28 0.18um CMOS Broadband Mixer for the new Cognitive Radio The second circuit of our third year proposal is shown in Fig.4-9. we also apply this concept in the transconductance stage of the mixer, so the complicated input matching networks can be ignored and achieve the low noise figure over the broad bantwidth. At the meanwhile, by applying current reuse and source degeneration topology, the linearity can be further increased. Fig.4-9. 第三年計畫Mixer [0-1] FCC, ET Docket No 03-222 Notice of proposed rule making and order, December 2003. [0-2] DARPA XG WG, The XG Architectural Framework V1.0, 2003. [0-3] DARPA XG WG, The XG Vision RFC V1.0, 2003. [1-1] C. Mishra, A. Valdes-Garcia, F. Bahmani, A. Batra, E. Sanchez-Sinencio, J. Silva-Martinez,“Frequency Planning and Synthesizer Architectures for Multiband OFDM UWB Radios,” IEEE Transactions on Microwave Theory and Techniques : Accepted for future publication, Volume PP, Issue 99, 2005 [1-2] B. Razavi, T. Aytur, F.-R. Yang, R.-H. Yan, H.-C. Kang, C.-C. Hsu, and C.-C. Lee, “A 0.13 um CMOS UWB transceiver,” IEEE Int. Solid-State Circuits Conf. Tech. Dig., Feb. 2005. [1-3] A. Medi and W. Namgoong, “A fully integrated multi-output CMOS frequency synthesizer for channelized receiver,” Proc. IEEE Int. System-on-Chip Conf., Sep. 2003 [1-4] T. Geum-Young, H. Seok-Bong, K. Tae Young, C. Byoung Gun, P. Seong Su, “A 6.3-9-GHz CMOS fast settling PLL for MB-OFDM UWB applications,” IEEE J. Solid-State Circuits, Volume 40, Issue 8, Aug. 2005 [1-5] Shih-Hao Tarng, Yu-Ching Tsai and Christina. F. Jou, “A Fully-Integrated,Low Power, Fast-Locking, Integer-N Frequency Synthesizer for MB-OFDM UWB System,” to be published inPIERS 2007 [1-6] D. Leenaerts, R. van de Beek, G. van der Weide, J. Bergervoet, K. S. Harish, H. Waite, Y. Zhang, C. Razzell, and R. Roovers, “A SiGe BiCMOS 1 ns fast hopping frequency synthesizer for UWB radio,” IEEE Int. Solid-State Circuits Conf. Tech. Dig., Feb. 2005. [1-7] J. Lee and D. Chiu, “A 7-band 3–8 GHz frequency synthesizer with 1 ns band-switching time in 0.18 um CMOS technology,” IEEE Int. Solid-State Circuits Conf. Tech. Dig., Feb. 2005. [1-8] Lawrence Williams, Daniel Wu, Eldon Staggs, Albert Yen, “Ultra-Wideband Radio Design for Multi-band OFDM 480 Mb/s Wireless USB,” DesignCon 2005. [1-9] Sandner, C., Wiesbauer, A., “ A 3GHz to 7GHz fast-hopping frequency synthesizer for UWB,” Ultra Wideband Systems, 2004. Joint with Conference on Ultrawideband Systems and Technologies. Joint UWBST & IWUWBS. 2004 International Workshop on 18-21 May 2004 Page(s):405 - 409 Digital Object Identifier 10.1109/UWBST.2004.1321005 [2-4] C. K. Aanadan, P. Mohanan, and K. G. Nair, “Broad-band gap coupled microstrip antenna,” IEEE Transactions on Antenna and Propagation, Vol. 38, No. 10, Oct. 1990. [3-6] Wide-Band Matched LNA Design Using Transistor’s Intrinsic Gate–Drain Capacitor