578368 玖、發明說明 (發明說明應敘明：發明所屬之技術領域、先前技術、內容、實施方式及圖式簡單說明） 【發明所屬之技術領域】 本發明係有關於一種新的多頻段電子電路（a multi-band electronic circuit )及其 設計方法。主要係於雙極（場效）電晶體之集極（汲極）與基極（閘極）間加入一電容 藉由該電谷加入，使從該電晶體基極（閘極）端看入之輸入阻抗與電性連接於基極（閘 極）之電感共振於所欲最大頻段與最小頻段之間，而達成各頻段之阻抗匹配。 & 【先前技術】 無線通訊產業已演進至多種標準/多種服務之境地，例如無線區域網路（Wireiess578368 发明 Description of the invention (The description of the invention should state: the technical field to which the invention belongs, the prior art, the content, the embodiments and the simple description of the drawings) [Technical field to which the invention belongs] The present invention relates to a new multi-band electronic circuit (A multi-band electronic circuit) and its design method. It is mainly connected with a capacitor between the collector (drain) and base (gate) of a bipolar (field-effect) transistor. The capacitor is added through the valley to make it look from the base (gate) of the transistor. The input impedance and the inductance electrically connected to the base (gate) resonate between the desired maximum frequency band and the minimum frequency band to achieve impedance matching of each frequency band. & [Previous Technology] The wireless communications industry has evolved to a variety of standards / multiple services, such as wireless LANs (Wireiess
Local Area Network，WLAN)使用 2.4 GHz，5·2 GHz，5·7 GHz 頻段、GSM 行動電 話使用 0.9 GHz，1·8 GHz，1.9 GHz 頻段、而全球定位系統（Gi〇ba丨 P〇siti〇n System， GPS)使用L5 GHz頻段。因此最好能將多種標準整合在同一收發機晶片中，亦即 要能設計製作出多頻段收發機。設計多頻段收發機最主要的挑戰，在於增進通訊收 發機的功能之同時，能使用最少額外之電路。 S知0又δ十多頻段收發機中的低雜訊放大器之策略是，針對某一頻段就設計符合該頻 段的低雜訊放大器。換言之，要設計能使用0.9 GHz，1.8GHz，1.9 GHz頻段之三頻收發 _ 機，就須設卞二組低雜訊放大器以因應三種不同頻率。因此在設計低雜訊放大器時，與 其相關的增益、雜訊指數（NoiseFigure)、輸入阻抗及輸出阻抗，都是對某一特定頻段 來做設計。如此一來，多頻段收發機之整個電路的面積及功率消耗，都要比單頻段收發 機大許多。以第一圖所示習知整合多頻段應用之超外差式（superheter〇dyne)接收機 B 為例：從天線100、頻段選擇濾波器1〇1、低雜訊放大器103、鏡像消除濾波器1〇4到頻 道選擇滤波器107 ’為應用頻段一之獨立接收路徑。從天線、頻段選擇濾波器η❶、 低雜訊放大器112、鏡像消除濾波器113到頻道選擇濾波器116，為應用頻段二之獨立 接收路徑。從天線118、頻段選擇濾波器119、低雜訊放大器121、鏡像消除濾波器122 • 到頻道選擇濾波器125，為應用頻段三之獨立接收路徑。 以應用頻段一之獨立接收路徑來做說明，訊號由天線10〇接收進來之後，先經 過頻段選擇濾波器101來濾除應用頻段一以外之頻段，然後再經由下一級之低雜訊 放大器103來放大訊號且減低雜訊的增加。再接下來由鏡像消除濾波器1〇4來消除 鏡像頻率處的雜訊，經降頻後，由頻道選擇濾波器107挑選應用頻段一中的某一頻 • 道。接下來是應用頻段一、應用頻段二及應用頻段三共用之電路部份，訊號在確認 為某一應用頻段之後，再降頻並利用類比·數位轉換器128來將訊號數位化，最後由 數位訊號處理129來處理已數位化之訊號。 由以上之敘述可知，在整合多頻段應用之接收機時，傳統的做法是將各頻段應 彙 用電路分別設計，再全部放在一起。而接收機中的關鍵電路低雜訊放大器，也須要 針對不同頻段而設計。這樣一來整個電路的面積及功率消耗勢必大大增加。在以往 所發表的論文中，對於整合多頻段應用的電路，都是採用這樣子的做法（亦即，使 用不同低雜訊放大器來處理不同頻段），可參照： • ▲一、Τ· Antes 氏和 C· Conkling 氏在 1996 年十二月於 Miciwave RF 上發表 之論文·· “RF chip set fits multimode cellular/PCS handsets，，，。 二、S· Wu氏和Β· Razavi氏在1998年十二月於IEEE jSSC上發表之論文： ]續次頁（發明說明頁不敷使用時’請註記並使用續頁 7 578368 __ w 發明說明續貝 “A 900-MHz/1.8-GHz CMOS receiver for dual-band applications，”。 三、R· Magoon 氏，I· Koullias 氏，L. Steigerwald 氏，W· Domino 氏，Ν· Vakillian 氏，Ε· ^ Ngompe 氏，Μ· Damgaard 氏，Κ· Lewis，和 A· Molna 氏在 2001 年二月於 ISSCC Digest of Technical papers 上發表之論文：“A triple-band 900/1800/19⑽ MHz low-power image-reject front-end for GSM，” o 四、K. L· Fong 氏在 1999 年二月於 ISSCC Digest of Technical papers 上發 P 表之論文：“Dual-band high-linearity variable-gain hm-noise amplifiers for wireless applications，”。 最近 H· Hashemi 氏和 A· Hajimiri 氏在 2002 年一月於 IEEE Transactions on Microwave Theory and Techniques 上發表之論文：“Concurrent Multiband _ Low_N〇ise Amplifiers-Theoiy，Design，and Applications，’’ 乃使用同一低雜訊放大器 來處理多頻段之訊號。此種多頻段的低雜訊放大器由於可以使用同一低雜訊放大器 滿足不同頻段的要求，所以在多頻段應用的整合上，可以簡化收發機的設計（不須 要設計多個不同的低雜訊放大器）。這樣一來也可以縮小整個系統電路的面積並減少 消耗功率，而面積的縮小及消耗功率的減少，對於電路的商品化是非常有利的。 • Η· Hashemi氏和A· Hajimiri氏所提出之低雜訊放大器的設計方法不同於傳統 低雜訊放大器之設計方法。關於習知低雜訊放大器的設計方法，請參照第二圖。其 乃利用源極電感207產生輸入阻抗匹配所需之電阻（通常為50歐姆），再利用電^ 201，使其與看入閘極端之總輸入電容達成共振於所欲頻段。輸出端處則使用電感 ft 204和電容208所構成的共振腔，選擇出所欲之頻段。 而關於上述H· Hashemi氏和A· Hajimiri氏所提出的多頻段低雜訊放大器的設計方 法，請參照第三圖。在輸入端處，除了使用習知可產生輸入阻抗匹配所需之電阻（通常 為50歐姆）的電感310及可達成共振於所欲頻段之電感304外，其又增設了並聯組合 _ 之電感及電容洲2。目的在於增加另一共振頻率，達成多頻段輸入匹配之功能。在 輸出端處，除了使用習知由電感312及電容313所組成之並聯共振腔外，亦增設了串聯 組合之電感307及電谷306。目的也在於增加另一共振頻率，達成選擇所欲多頻段之功 能。簡言之H· Hashemi氏和A· Hajimiri氏乃以增加電感及電容之數量來達成多頻段應 用之功能。這樣子的設計方法有不少缺點。 ’ ® 严先，此設計一共用了五個電感（即電感301、電感304、電感307、電感310 和電感312,其中電感301和電感304為晶片外的電感）和三個電容（包括電容3〇2、 電容306和電容313,其中電容302為晶片外的電容），比起傳統低雜訊放大器的設 a十（明參考第一圖’具二個晶片上的電感：201、204、207和一個晶片上的電容： • 208)要多了兩個電感和兩個電容。由於電感、電容數目的增加，甚至使用到晶片外 的電感、電容（比在晶片上的電感、電容面積要大很多）。整個電路的面積變得很大，Local Area Network (WLAN) uses 2.4 GHz, 5 · 2 GHz, 5 · 7 GHz frequency bands, GSM mobile phones use 0.9 GHz, 1.8 GHz, 1.9 GHz frequency bands, and the global positioning system (Gioba 丨 P〇siti〇) n System, GPS) uses the L5 GHz frequency band. Therefore, it is better to integrate multiple standards in the same transceiver chip, that is, to design and manufacture a multi-band transceiver. The main challenge in designing a multi-band transceiver is to improve the capabilities of the communications transceiver while using a minimum of extra circuitry. It is known that the low-noise amplifier strategy in more than ten-band transceivers is to design a low-noise amplifier that meets that frequency band for a certain frequency band. In other words, to design a tri-band transceiver that can use the 0.9 GHz, 1.8 GHz, and 1.9 GHz frequency bands, two sets of low-noise amplifiers must be set up to respond to three different frequencies. Therefore, when designing a low-noise amplifier, the related gain, noise figure, input impedance, and output impedance are all designed for a specific frequency band. As a result, the area and power consumption of the entire circuit of the multi-band transceiver is much larger than that of the single-band transceiver. Take the conventional superheterodyne receiver B integrated with multi-band applications shown in the first figure as an example: antenna 100, frequency band selection filter 101, low noise amplifier 103, and image cancellation filter 104 to the channel selection filter 107 'is an independent receiving path of the application frequency band 1. The antenna, the band selection filter η❶, the low noise amplifier 112, the image removal filter 113, and the channel selection filter 116 are independent receiving paths for the application band two. From antenna 118, band selection filter 119, low-noise amplifier 121, image removal filter 122 • to channel selection filter 125, it is the independent receiving path for application band three. Take the independent receiving path of the application frequency band one for illustration. After the signal is received by the antenna 100, it first passes the frequency band selection filter 101 to filter out the frequency band other than the application frequency band 1, and then passes through the low-noise amplifier 103 of the next stage. Amplify the signal and reduce the increase in noise. Next, the image cancellation filter 104 removes the noise at the image frequency. After the frequency is reduced, the channel selection filter 107 selects a certain channel in the application frequency band 1. The next part is the circuit common to the application frequency band 1, the application frequency band 2, and the application frequency band 3. After confirming the signal as an application frequency band, the frequency is reduced and the analog / digital converter 128 is used to digitize the signal. The signal processing 129 processes a digitized signal. As can be seen from the above description, when integrating receivers for multi-band applications, the traditional approach is to design the application circuits for each frequency band separately and put them all together. The low-noise amplifier for the key circuits in the receiver must also be designed for different frequency bands. As a result, the area and power consumption of the entire circuit is bound to increase significantly. In the previous published papers, for integrated circuits of multi-band applications, this method is used (that is, different low-noise amplifiers are used to handle different frequency bands), which can be referred to: • ▲ 一 、 Τ · Antes And C. Conkling ’s paper published on Miciwave RF in December 1996. “RF chip set fits multimode cellular / PCS handsets,”, 2. S. Wu and B. Razavi in December 1998 Papers published on IEEE jSSC:] Continued page (When the description page of the invention is insufficient, please note and use the next page 7 578368 __ w Description of the invention continued "A 900-MHz / 1.8-GHz CMOS receiver for dual- band applications, "III. R. Magoon, I. Koullias, L. Steigerwald, W. Domino, N. Vakillian, E. ^ Ngompe, M. Damgaard, K. Lewis, and A. Molna's paper published in ISSCC Digest of Technical papers in February 2001: "A triple-band 900/1800 / 19⑽ MHz low-power image-reject front-end for GSM," o K. L · Fong ISSCC D in February 1999 Igest of Technical papers published the paper of the P form: "Dual-band high-linearity variable-gain hm-noise amplifiers for wireless applications," recently H · Hashemi and A · Hajimiri's in January 2002 in IEEE on A paper published on Microwave Theory and Techniques: "Concurrent Multiband _ Low_Noise Amplifiers-Theoiy, Design, and Applications," uses the same low-noise amplifier to process multi-band signals. This multi-band low-noise amplifier Because the same low-noise amplifier can be used to meet the requirements of different frequency bands, the integration of multi-band applications can simplify the design of the transceiver (there is no need to design multiple different low-noise amplifiers). In this way, the area of the entire system circuit can be reduced and the power consumption can be reduced. The reduction in area and power consumption is very advantageous for the commercialization of the circuit. • The design method of low noise amplifier proposed by Η Hashemi and A. Hajimiri is different from that of traditional low noise amplifier. For the design method of the conventional low noise amplifier, please refer to the second figure. It uses the source inductor 207 to generate the resistance (usually 50 ohms) required for input impedance matching, and then uses electricity 201 to make it resonate with the total input capacitance at the gate extremes at the desired frequency band. At the output end, a resonant cavity formed by an inductor ft 204 and a capacitor 208 is used to select a desired frequency band. For the design method of the multi-band low noise amplifier proposed by H. Hashemi and A. Hajimiri, please refer to the third figure. At the input, in addition to the conventional inductor 310 that can generate the resistance (usually 50 ohms) required for input impedance matching and the inductor 304 that can reach the desired frequency band, it also adds a parallel combination of the inductor and Capacitor continent 2. The purpose is to add another resonance frequency to achieve the function of multi-band input matching. At the output end, in addition to the conventional parallel resonant cavity composed of the inductor 312 and the capacitor 313, a series combination inductor 307 and an electric valley 306 are also added. The purpose is also to increase another resonance frequency to achieve the function of selecting the desired multiple frequency bands. In short, H. Hashemi's and A. Hajimiri's use multi-band applications by increasing the number of inductors and capacitors. There are many disadvantages to this design approach. '® Strictly, this design shares five inductors (ie inductor 301, inductor 304, inductor 307, inductor 310, and inductor 312, where inductor 301 and inductor 304 are off-chip inductors) and three capacitors (including capacitor 3 〇2, capacitor 306 and capacitor 313, of which capacitor 302 is off-chip capacitor), compared to the traditional low-noise amplifier design a (refer to the first picture 'with two chip inductors: 201, 204, 207 And capacitors on one chip: • 208) requires two more inductors and two capacitors. Due to the increase in the number of inductors and capacitors, even inductors and capacitors outside the chip are used (much larger than the area of the inductors and capacitors on the chip). The area of the entire circuit becomes very large,
而且沒有辦法將整個設計整合於同一晶片上。晶片外之電感及電容須額外之打線及 配線，增加成本且降低可靠度，這對於積體電路的量產和商品化是相當不利的、在 5又®十低雜訊放大器的時候，通常會儘量減少電感的使用，一來是因為電感所佔面積 很大’一來疋在晶片上的電感其品質因子Fact〇r)不高，會造成雜訊指數 的劣化。所以在設計低雜訊放大器時，一般是要儘量避免使用電感。而H 氏和A· Hajimiri氏所提出的方法卻是增加電感的使用。 因此非常需要有-種不增加電感使用數量且不需額外打線，但仍能處頻段的放 馨 大器。 續次頁（發明說明頁不敷使用時，請註記並使用續頁 8 •578368And there is no way to integrate the entire design on the same chip. The inductors and capacitors outside the chip must have additional wiring and wiring, which increases costs and reduces reliability. This is very detrimental to the mass production and commercialization of integrated circuits. In the case of 5 and 10 low-noise amplifiers, usually Minimize the use of inductors. One reason is that the inductor occupies a large area. At the same time, the quality factor Fact0) of the inductor on the chip is not high, which will cause the deterioration of the noise index. Therefore, when designing a low noise amplifier, it is generally necessary to avoid the use of inductors as much as possible. The method proposed by H and A. Hajimiri is to increase the use of inductance. Therefore, it is very necessary to have an amplifier that does not increase the number of inductors and does not require additional wiring, but can still be in the frequency band. Continued pages (Note when the invention description page is not enough, please note and use the continuation page 8 • 578368
明說明續頁 【內容】 、咅点之目的在提供一種多頻段放大器及其設計方法，僅使用單一放大器即可 違成^頻段之輸人阻抗匹配，而且不增加電感使用數量，也不需額外打線。 感使用數量，也不要額外打線、配線，本發明提出於放大器中雙極電晶 茨：質接面雙極電晶體（Bip〇lar juncti〇n Transist〇r〇r Heter〇juncti〇n 职时 的Λ?與集-極間再電性連接一電容器；或者放大器中場效電晶想的閘極與 k4效二If器，使得看入輸入端（雙極電晶趙或異質接面雙極電晶體為i 極％效電曰曰體為閘極）之總輸入電容與連接於基極（閘極）之電感雖共振於所欲最大 頻段與最小頻段之間，但仍使所欲各頻段之輸入折返損耗（i叩utretunU〇ss，^ )小 於^l^dB而達成多頻段輸入阻抗匹配之功能。以雙極電晶體為例，看入基極端之總輸 極-射極、電容與米勒電容（Mmer capaeit觀e，其乃由基極與集極間^電 體增錢造成）。於此基極端電性連接—電感11，則看人基極端之總輸入電容 感可共振於所欲之頻率。由於本發明乃於基極與集極間（或閘極與汲極間）電性 Siif性元件，藉㈣勒效應，電晶體基極（或閘極）會看到此電容被放大，因此基 極興集極間（或閘極與汲極間）電性連接電容性元件之電容值雖小，即可達成大幅产丘 振頻率之改變。所以相對於習知技藝，本創作不需增加電感，雖需增加電容，但 很小即可（即所增面積亦小）。又本創作不要額外打線、配線。 為讓本發明之上述和其他目的，特徵，和優點能更明顯易懂，下文特舉較佳 施例，並配合所附圖，作詳細說明如下： 【實施方式】 兹 第一實施例 參閱第四圖，其乃本創作具多頻段處理功能之第一實施例的電路圖。在此電路 中我們雖使用雙極電晶體，但使用場效電晶體也可以。第一電阻4〇7與第二電阻412 均為2〇0歐姆；第三電阻410為600歐姆；直流阻隔/交流耦合電容409為3pF ;該 第一义晶體408與第二電晶體413射極面積均為12· 18平方微米。製程採TSMCO. 35um SiGe BiCMOS製程。在此多頻段低雜訊放大器中，我們將一電容值〇 2pF的電容器 415電性連接於放大器中第一級電晶體的基極端與集極端之間，使得看入基極端之 總輸入電容與連接於基極之電感404雖共振於所欲最大頻段（5·7 GHz)與最小頻段 (2·4 GHz)之間，但仍使各頻段之輸入折返損耗（input retuni 1〇ss，|%丨| )小於， dB而達成多頻段輸入阻抗匹配之功能。在輸出端414部份，我們使用了回授電阻4i〇 達成輸出阻抗匹配。在不需輸出阻抗匹配的情況下，可不用回授電阻41〇達成輸出 阻抗匹配。電阻407及電阻412為分別為第一級電晶體及第二級電晶體之負載。本 實施例雖用電阻為負載，視需要使用電感或電容負載亦是可以的。重點是輸入端能 達成多頻段阻抗匹配。由於我們只使用了一個電感404，而且是製作在晶片上的電 感，因此不但整個電路可以完全在單一晶片上實現，而且電路的面積非常小。這對 於商品化非常有利。 有關此雙頻段低雜訊放大器在增益上的表現，請參照第五圖。此多頻段低雜訊 放大器在2.4 GHz、5.2 GHz及5.7 GHz的增益（散射參數中S21來表示）分別達到 續次頁（發明說明頁不敷使用時，請註記並使用續頁 9 發明說明續頁 工25dB、17.5dB及16dB«此多頻段低雜訊放大器對於輸入阻抗的匹配程度（通 辛以散射參數中輸入折返損耗input return loss Sii來表示），在2·4 GHz和2.5 GHz 之間皆低於-12·5 dB以下（愈低愈好）；5.15 GHz和5·35 GHz之間皆低於-14 dB以 下（愈低愈好）；在5.725GHz和5.825GHz之間皆低於-UdB以下（愈低愈好）。 有關此多頻段低雜訊放大器在雜訊指數上的表現，請參照第六圖。在2/GHz“、5.2/、 5.7GHz的雜訊指數分別為2·2、2·8、3·1 dB(愈低愈好）。一般對K802Jla&802.llb 無線區域網路（WLAN)之應用而言，低雜訊放大器之雜訊指數只要低於5仙即可， 輸入（輸出）折返損耗小於-10 dB即可。因此我們可以說，根據本創作之實施例： 2·4/5·2/5.7 GHz多頻段低雜訊放大器，其有關於增益、雜訊指數、輸入阻抗匹配程 度上的表現’在2·4 GHz、5.2 GHz和5·7 GHz三個頻段下都有相當好的實施結果。 相較於習知的多頻段低雜訊放大器，本創作僅使用單一放大器即可達成多種頻段之 輸入阻抗匹配，既不增加電感數量，也不會大幅增大所佔面積，而且不需額外打線。 又，如要使輸入折返損耗inputreturnlossSn更低，可使第一級雙極電晶體之射極 不直接接地，而是射極接上一個電感之一端，電感另一端再接地。 第二實施例 第七圖乃本創作具2.4/5.2GHz多頻段處理功能之第二實施例的電路圖。第二 實施例在證明根據本創作之精神，不僅可以使用電阻性負載，亦可以使用電感或電 容性負載，甚至能具有鏡頻抑制之功能。參照第七圖，此電路基本上亦由兩級^射 極電路所組成。只不過本第二實施例之第二級共射極電路乃疊接在第一級共射以電 路之上。第二實施例第一級共射極電路基本上與第一實施例之第一級共射極電路相 同。第二實施例第二級共射極電路基本上亦與第一實施例之第二級共射極電路相 同，只不過此時於第二實施例第二級共射極電路中電晶體5〇6的射極端接上電感5〇8 與電容507組成之共振腔，共振頻率選於鏡像頻率以抑制鏡像頻率訊號。此多頻段 低雜訊放大器中，我們將一電容值〇.〇6PF的電容器513電性連接於放大器中第一級 電晶體的基極端與集極端之間，使得看入基極端之總輸入電容與連接於基極之電感 514雖共振於所欲最大頻段（5.2GHz)與最小頻段（2.4GHz)之間，'但仍使各頻 段之輸入折返損耗（input return l〇ss，|心| )小於-10dB而達成多頻段輸入阻抗匹 配之功能。電阻511為第一級電晶體512之負載。第二級共射極電路之負載採用與 H· Hashemi氏和A· Hajimiri氏相同之雙共振頻率共振腔，由電感5〇3、電感5〇2、' 電容504、電容501所組成。電容509為兩級間直流阻隔/交流耦合電容。電阻5〇5 提供第二級電晶體506之輸入偏壓電流。電容510為旁路電容，交流接地用。 有關此雙頻段低雜訊放大器在增益上的表現，請參照第八圖。此多頻段低雜訊 放大器在2.4 GHz、5.2 GHz的增益（散射參數中Sr來表示）分別達到了 Η及 10dB。此多頻段低雜訊放大器對於輸入阻抗的匹配程度（通常以散射參數中輸入折 返損耗input return loss Sn來表示），在2.4 GHz和2.5 GHz之間皆低於-u dB以下 (愈低愈好）’ 5·15 GHz和5·35 GHz之間皆低於-12 dB以下（愈低愈好）。有關此 多頻段低雜訊放大器在雜訊指數上的表現，請參照第九圖。在、5.2GHz的 雜訊指數分別為2· 4、3·3 dB (愈低愈好）。一般對於8〇2.lla及8〇2.1lb無線區域網 路（WLAN)之應用而言，低雜訊放大器之雜訊指數只要低於5 dB即可了輸入（輸 出）折返損耗小於-1〇 dB即可。因此我們可以說，根據本創作之實施例：2·4/5 2 GHz 多頻段低雜訊放大器，其有關於增益、雜訊指數、輸入阻抗匹配程度上的表現，在 續次頁（發明說明頁不敷使用時，請註記並使用續頁 發明說明續頁 2.4 GHz、5.2 GHz兩個頻段下都有相當好的實施結果^相較於習知的多頻段低雜訊 放大器，本創作僅使用單一放大器即可達成多種頻段之輸入阻抗匹配，既不增加電 感數量，也不會大幅增大所佔面積，而且不需額外打線。又，如要使輸入折返損耗 inputreturnlossSn更低，可使第一級雙極電晶體之射極不直接接地，而是射極接 上一個電感之一端，電感另一端再接地。 综上所述’當知本案所創作之多頻段同時共存電子電路已具有產業利用性、新穎性 與進步性，符合發明專利要件。惟以上所述者，僅為本創作之一較佳實施例而已，並非 用來限定本創作實施之範圍。即凡依本創作申請專利範圍所做的均等變化與修飾，皆為 本創作專利範圍所涵蓋。 【圖式簡單說明】 各圖意義如下： 第一圖為習知為了多頻段應用所採之多頻段晶片整合方法 第二圖為習知低雜訊放大器之電路圖 第三圖為H· Hashemi氏和A· Hajhniri氏所發表之多頻段低雜訊放大器的電路 圖 第四圖為本創作第一實施例（2.4/ 5·2/ 5·7 GHz多頻段低雜訊放大器）的電路圖 第五圖為本創作實施例（2.4/5.2/5.7GHz多頻段低雜訊放大器)功率增益及輸 入折返損耗對頻率的特性 第六圖本創作實施例（2·4/5.2/5.7 GHz多頻段低雜訊放大器)雜訊指數對頻率 的特性 第七圖為本創作第二實施例（2.4/5.2 GHz多頻段低雜訊放大器）的電路圖 第八圖為本創作實施例（2.4/5.2GHz多頻段低雜訊放大器)功率增益及輸入折 返損耗對頻率的特性 第九圖為本創作實施例（2.4/ 5·2/ GHz多頻段低雜訊放大器)雜訊指數對頻率的 特性 102帶通濾波器 105帶通濾波器 108帶通濾波器 111帶通濾波器 圖式中之參照號數 100天線 103低雜訊放大器 106本地振盪訊號 109天線 101頻段選擇濾波器 104鏡像消除濾波器 107頻道選擇濾波器 110頻段選擇濾波器 D續次頁（發明說明頁不敷使用時，請註記並使用續頁 578368 發明說明續頁 112低雜訊放大器 113鏡像消除濾波器 114帶通濾波器 115本地振盪訊號 116頻道選擇濾波器 117帶通濾波器 118天線 119頻段選擇濾波器 120帶通濾波器 121低雜訊放大器 122鏡像消除濾波器 123帶通濾波器 124本地振盪訊號 125頻道選擇濾波器 126帶通濾波器 127中頻訊號 128類比-數位轉換器 129數位訊號處理 200輸入端 201電感 202偏壓 203電壓源 204電感 205電晶體 206電晶體 207電感 208電容 209輸出端 300輸入端 301電感 Ψ 302電容 303偏壓 304打線電感 305襯墊 306電容 307電感 308場效電晶體 309場效電晶體 310電感 » 312電感 313電容 314輸出端 400輸入端 401電流源 402電容 404電感 405集極電流 406電壓源 » 407電阻 408電晶體 409電容 410電阻 411電源 412電阻 413電晶體 414輸出端 415電容 501電容 502電感 503電感 504電容 505電阻 506電晶體 507電容 508電感 509電容 510電容 511電阻 512電晶體 513電容 514電感The content of the continuation page [Content] and the purpose of the point is to provide a multi-band amplifier and its design method, using only a single amplifier can violate the input impedance matching of the ^ band, and does not increase the number of inductors, and does not require additional Hit the line. Sense the number of use, and do not need additional wiring or wiring, the present invention proposes a bipolar transistor in the amplifier: the quality interface bipolar transistor (Bip〇lar juncti〇n Transist〇r〇r Heter〇juncti〇n A capacitor is re-electrically connected between Λ? And the collector; or the gate of the field effect transistor in the amplifier and the k4 effect two If device, so that the input terminal (bipolar transistor or heterojunction bipolar capacitor) The total input capacitance of the crystal is i pole% efficiency (the body is a gate) and the inductance connected to the base (gate) resonates between the desired maximum frequency band and the minimum frequency band, but still makes the desired frequency bands The input foldback loss (i 叩 utretunU0ss, ^) is less than ^ l ^ dB to achieve the function of multi-band input impedance matching. Taking a bipolar transistor as an example, look at the total input-emitter, capacitor, and meter of the base terminal. Le capacitor (Mmer capaeit concept, which is caused by the base and collector ^ electric body increase money). At this base extreme electrical connection-inductor 11, the total input capacitance of the human base can be resonant in the sense Desired frequency. Since the present invention is electrically between the base and the collector (or between the gate and the drain) Sii f-type element, by virtue of the Haller effect, the capacitor base (or gate) of the transistor will see that this capacitor is amplified, so the capacitor is electrically connected to the capacitor between the base and the collector (or between the gate and the drain). Although the value is small, you can achieve a large change in the frequency of the hillock vibration. Therefore, compared to the conventional technique, this creation does not need to increase the inductance. Although it needs to increase the capacitance, it can be small (that is, the increased area is also small). Do not make extra wiring or wiring for the creation. In order to make the above and other objects, features, and advantages of the present invention more comprehensible, the following describes the preferred embodiments in detail with the accompanying drawings, as follows: [Embodiment] The first embodiment is referred to the fourth figure, which is a circuit diagram of the first embodiment of the present invention with multi-band processing function. Although we use bipolar transistors in this circuit, field effect transistors can also be used. First The resistors 407 and 412 are both 200 ohms; the third resistor 410 is 600 ohms; the DC blocking / AC coupling capacitor 409 is 3 pF; the emitter areas of the first sense crystal 408 and the second transistor 413 are both It is 12 · 18 square microns. The manufacturing process is TSMC O. 35um SiGe BiCMOS process. In this multi-band low-noise amplifier, we electrically connect a capacitor 415 with a capacitance value of 2 pF between the base terminal and the collector terminal of the first-stage transistor in the amplifier. The total input capacitance of the base terminal and the inductor 404 connected to the base terminal resonate between the desired maximum frequency band (5 · 7 GHz) and the minimum frequency band (2 · 4 GHz), but still make the input foldback loss of each frequency band (input retuni 1〇ss, |% 丨 |) is less than, dB to achieve the function of multi-band input impedance matching. At the output terminal 414, we use a feedback resistor 4i〇 to achieve output impedance matching. In the case of no output impedance matching, the output impedance matching can be achieved without the feedback resistor 41. The resistors 407 and 412 are the loads of the first-stage transistor and the second-stage transistor, respectively. Although a resistance is used as a load in this embodiment, an inductive or capacitive load may be used as required. The important point is that the input can achieve multi-band impedance matching. Since we only use one inductor 404, and it is an inductor made on a chip, not only the entire circuit can be completely implemented on a single chip, but the circuit area is very small. This is very beneficial for commercialization. For the performance of this dual-band low-noise amplifier in gain, please refer to the fifth figure. The gain of this multi-band low-noise amplifier at 2.4 GHz, 5.2 GHz, and 5.7 GHz (indicated by S21 in the scattering parameters) has reached the next page (when the description page of the invention is not enough, please note and use the continued page 9 Invention description continued Pager 25dB, 17.5dB and 16dB «The matching degree of the input impedance of this multi-band low-noise amplifier (Tongxin is expressed by the input return loss Sii in the scattering parameters), between 2.4 GHz and 2.5 GHz Both are below -12 · 5 dB (lower is better); between 5.15 GHz and 5.35 GHz are below -14 dB (lower is better); between 5.725GHz and 5.825GHz Below -UdB (the lower the better). For the performance of this multi-band low-noise amplifier on the noise index, please refer to the sixth figure. The noise indexes at 2 / GHz, 5.2 /, and 5.7GHz are 2 · 2, 2 · 8, 3 · 1 dB (lower is better). Generally, for the application of K802Jla & 802.llb wireless local area network (WLAN), the noise index of the low noise amplifier is only less than 5 cents. That is, the input (output) foldback loss is less than -10 dB. So we can say that according to the implementation of this creation : 2 · 4/5 · 2 / 5.7 GHz multi-band low-noise amplifier, which has performance on gain, noise index, and input impedance matching level 'in three of 2.4 GHz, 5.2 GHz, and 5.7 GHz All bands have fairly good implementation results. Compared to the conventional multi-band low-noise amplifier, this creation uses only a single amplifier to achieve input impedance matching in multiple frequency bands, neither increasing the number of inductors nor greatly increasing It has a large area and does not require additional wiring. In addition, if the input return loss is to be lower, the emitter of the first-stage bipolar transistor can not be directly grounded, but the emitter can be connected to one end of an inductor. The other end of the inductor is grounded again. The seventh diagram of the second embodiment is a circuit diagram of the second embodiment of the present invention with a 2.4 / 5.2 GHz multi-band processing function. The second embodiment proves that according to the spirit of this creation, not only resistive The load can also use an inductive or capacitive load, and can even have the function of image frequency suppression. Referring to the seventh figure, this circuit is basically also composed of a two-stage ^ emitter circuit. The stage common emitter circuit is superimposed on the first stage common emitter circuit. The second embodiment first stage common emitter circuit is basically the same as the first stage common emitter circuit of the first embodiment. Second implementation The second-stage common-emitter circuit of the example is basically the same as the second-stage common-emitter circuit of the first embodiment, except that the emission of the transistor 506 in the second-stage common-emitter circuit of the second embodiment is now At the extreme end, a resonance cavity composed of an inductor 508 and a capacitor 507 is connected. The resonance frequency is selected at the image frequency to suppress the image frequency signal. In this multi-band low-noise amplifier, we electrically connect a capacitor 513 with a capacitance of 0.06PF between the base terminal and the collector terminal of the first-stage transistor in the amplifier, so that the total input capacitance of the base terminal can be seen. Although the inductor 514 connected to the base resonates between the desired maximum frequency band (5.2GHz) and the minimum frequency band (2.4GHz), it still causes the input return loss of each frequency band (input return l0ss, | heart |) Less than -10dB to achieve multi-band input impedance matching function. The resistor 511 is a load of the first-stage transistor 512. The load of the second-stage common-emitter circuit uses the same double-resonance frequency resonant cavity as H. Hashemi's and A. Hajimiri's, and is composed of inductor 503, inductor 502, capacitor 504, and capacitor 501. The capacitor 509 is a DC blocking / AC coupling capacitor between the two stages. The resistor 505 provides the input bias current of the second transistor 506. The capacitor 510 is a bypass capacitor for AC ground. For the performance of this dual-band low-noise amplifier in gain, please refer to Figure 8. The gain of this multi-band low-noise amplifier at 2.4 GHz and 5.2 GHz (indicated by Sr in the scattering parameters) has reached Η and 10 dB, respectively. The matching degree of the input impedance of this multi-band low-noise amplifier (usually expressed by input return loss Sn in the scattering parameters) is below -u dB between 2.4 GHz and 2.5 GHz (the lower the better ) 'Both below 5.15 GHz and 5.35 GHz are below -12 dB (the lower the better). For the performance of this multi-band low noise amplifier on the noise index, please refer to the ninth figure. The noise index at 5.2GHz is 2.4, 3.3 dB (the lower the better). Generally, for the applications of 802.lla and 802.1lb wireless local area network (WLAN), the noise index of the low-noise amplifier is less than 5 dB, and the input (output) return loss is less than -1. dB is enough. Therefore, we can say that according to the embodiment of this creation: 2 · 4/5 2 GHz multi-band low-noise amplifier, which has performance on gain, noise index, and input impedance matching degree. When the page is not enough, please note and use the continuation of the invention description. The continuation pages have good results in both 2.4 GHz and 5.2 GHz bands. ^ Compared to the conventional multi-band low-noise amplifier, this creation uses only A single amplifier can achieve input impedance matching for multiple frequency bands, neither increasing the number of inductors, nor increasing the occupied area significantly, and no additional wiring is required. In addition, if the input return loss is to be lower, the first The emitter of the bipolar transistor is not directly grounded, but the emitter is connected to one end of an inductor, and the other end of the inductor is then grounded. In summary, 'When the multi-band coexistence electronic circuit created in this case is known, it has industrial use. Nature, novelty, and progress are in line with the requirements of the invention patent. However, the above is only one of the preferred embodiments of this creation, and is not intended to limit the scope of implementation of this creation. That is to say, all equal changes and modifications made according to the scope of the patent application for this creation are covered by the scope of the invention patent. [Simplified illustration of the diagram] The meaning of each diagram is as follows: Frequency band chip integration method. The second figure is the circuit diagram of the conventional low noise amplifier. The third diagram is the circuit diagram of the multi-band low noise amplifier published by H. Hashemi and A. Hajhniri. The fourth diagram is the first embodiment of the creation. (2.4 / 5 · 2/5 · 7 GHz multi-band low-noise amplifier) Circuit diagram The fifth figure is the creative example (2.4 / 5.2 / 5.7GHz multi-band low-noise amplifier) power gain and input return loss vs. frequency Characteristics of the sixth picture This creative embodiment (2 · 4 / 5.2 / 5.7 GHz multi-band low-noise amplifier) noise index versus frequency characteristics The seventh picture is the second embodiment of the creative (2.4 / 5.2 GHz multi-band low-noise amplifier) Noise amplifier) Circuit diagram The eighth diagram is the author ’s example (2.4 / 5.2GHz multi-band low-noise amplifier) power gain and input foldback loss vs. frequency characteristics The ninth diagram is the author ’s embodiment (2.4 / 5 · 2 / GHz multi-band low noise amplifier Characteristics of noise index versus frequency 102 band-pass filter 105 band-pass filter 108 band-pass filter 111 band-pass filter reference number 100 antenna 103 low-noise amplifier 106 local oscillation signal 109 antenna 101 Band selection filter 104 Mirror elimination filter 107 Channel selection filter 110 Band selection filter D Continued page (When the description page of the invention is insufficient, please note and use the next page 578368 Invention description continued page 112 Low noise amplifier 113 Mirror Elimination filter 114 Bandpass filter 115 Local oscillation signal 116 Channel selection filter 117 Bandpass filter 118 Antenna 119 Band selection filter 120 Bandpass filter 121 Low noise amplifier 122 Mirror elimination filter 123 Bandpass filter 124 Local oscillation signal 125 channel selection filter 126 band-pass filter 127 intermediate frequency signal 128 analog-to-digital converter 129 digital signal processing 200 input 201 inductive 202 bias 203 voltage source 204 inductor 205 transistor 206 transistor 207 inductor 208 capacitor 209 Output 300 Input 301 Inductance Ψ 302 Capacitor 303 Bias 304 Wire Inductor 305 Pad 306 Capacitor 307 inductor 308 field effect transistor 309 field effect transistor 310 inductor »312 inductor 313 capacitor 314 output terminal 400 input 401 current source 402 capacitor 404 inductor 405 collector current 406 voltage source» 407 resistor 408 transistor 409 capacitor 410 resistor 411 Power supply 412 resistor 413 transistor 414 output 415 capacitor 501 capacitor 502 inductor 503 inductor 504 capacitor 505 resistor 506 transistor 507 capacitor 508 inductor 509 capacitor 510 capacitor 511 resistor 512 transistor 513 capacitor 514 inductor