JP2004252386A - Optical millimeter wave / microwave signal generation method and apparatus - Google Patents

Abstract

A two-mode optical millimeter-wave / microwave QPSK signal generation method that does not generate signal interference components, a two-mode optical millimeter-wave / microwave BPSK signal generation method that does not cause spectrum expansion, and an optional modulation format To provide an optical millimeter-wave / microwave signal generation method.
A light from a single wavelength light source is split by a branching optical waveguide, and the split first light and second light are input to an optical modulator and an optical modulator, respectively. The first RF signal 13 and the second RF signal 15 are input to the optical modulator 4 and the optical modulator 5, respectively, and are subjected to optical DSB-SC modulation, and the optical DSB-SC modulated first light and second light are input. The phase of π / 2 is added by the π / 2 fixed phase shifter 6 and added. In the heterodyne received light output of the added first light and the light by the second photodiode 9, the interference component output of the first RF signal and the second RF signal is suppressed.
[Selection diagram] Fig. 1

Description

式中の変調度αは、光位相変調器半波長電圧ＶπおよびＲＦ電圧の振幅Ａ_ＲＦと以下の関係で結ばれている。
【数２】

また、ベッセル関数Ｊ_ｎ （ｘ）は、図３３に示される。
半波長電圧Ｖπは光位相変調器透過光が位相変調π／２を受ける電圧である。Ｖπは変調効率を示すデバイスパラメータであり、低い電圧であることが望まれる。現在は５Ｖ_ｐ−ｐ が一般的であるが近年の研究によりこの値は低下している。ここでは、一定のＤＣ（直流）バイアスで一定の位相を印加する光位相変調器を固定位相器と呼ぶ。また、同一の光位相変調器で、ＲＦ信号による光位相変調と固定位相の付加を行う場合があるが、ここでは区別しない。同一の光位相変調器で行う場合には、電気回路においてＤＣ信号とＲＦ信号を合成するバイアスＴを利用する。
【００１１】
次に、この光位相変調器の集積により構成される１電極ＭＺＭの出力光および光ＳＳＢ変調器の出力光を説明する。ＭＺＭは２つの光位相変調器と１つの固定位相器とから構成されており、ＤＳＢ−ＳＣ法を行う１電極ＭＺＭ構成は図３７に示されている。１電極ＭＺＭ変調器は、光位相変調器間の位相差、および入力するＲＦ信号の位相差が重要なパラメータとなる。前者は電極より注入する電界のＤＣ成分により制御可能であり、後者は電気回路構成を変えることで制御可能である。なお、通常の１電極ＭＺＭは、内部の２つの光位相変調器に対し互いに逆相となるＲＦ信号を入力する、すなわち後者パラメータはπに固定される。
【００１２】
１電極ＭＺＭの出力光のＤＣバイアス依存性を説明する。
ここでＤＣバイアスにより与えられる位相差がθ_ＤＣである時の１電極ＭＺＭ の出力電界Ｅ_ｏｕｔ を表す。なお、ここでは簡単のためチャープ、挿入損失は含めない。
【数３】

ここで、θ_ＤＣがπである場合、上式は以下のように展開される。
【数４】

これより１電極ＭＺＭの出力光が奇数次側波帯からなり、搬送波および偶数次側波帯が抑圧されることが分かる。この変調法は搬送波抑圧両側波帯変調（ＤＳＢ−ＳＣ）（非特許文献３参照）と呼ばれ、ミリ波・マイクロ波信号生成の際に多用される。その特徴は後述する。ＤＣバイアスにより制御する位相項が０である場合は、以下のように展開される。
【数５】

このバイアス下では偶数次側波帯が出力される。なお、ここでαが２．４程度となるＲＦ強度を入力すると、四逓倍機能を有する光搬送波成分が抑圧され２モード光ミリ波・マイクロ波信号が生成される。
以上のように、光位相変調器から構成されるＭＺＭ等の光変調器においては、ＤＣバイアスにより制御する被変調光間の位相差が、光スペクトル制御上重要なパラメータとなる。
【００１３】
前述のように定量的に光スペクトルを把握することが可能であるが、より理解を容易にするために、位相および周波数の包括的表現が可能なスペクトル模式図を説明する。前提として式１で表される光位相変調器出力光電界を、周波数と位相に注目しつつ展開する。
【数６】

ここで、入力ＲＦ信号位相に対応した光電界は図３４（ａ）で表現可能である。図において、斜軸は周波数軸を示し、軸回転角は位相を表す。また、光搬送波の有する位相を同図における基準位相とし、これを波線横軸としている。図中、矢印長は電界強度を示すが、これは各次数におけるベッセル関数に対応する。
【００１４】
図３４、図３５、図３６は、それぞれＲＦ信号の位相変化、すなわち時間変化に対応する。各図中の（ｂ）〜（ｊ）は以下の通りである。（ｂ）は、入力ＲＦ信号を表現するフェーザである。図中Ｉ（同相：Ｉｎ ｐｈａｓｅ）軸成分に対応する実電界が変調器にかかる。（ｃ）は、このＲＦ電界から位相変調を受けた光を表現したフェーザである。このフェーザに表現される電界を分解したものを（ｄ）〜（ｆ）に示す。（ｄ）は定常成分（搬送波成分）であり、（ｅ）はＱ（直交：Ｑｕａｄｒａｔｕｒｅ）軸方向変動成分、（ｆ）はＩ軸方向変動成分である。これは、式１における、
【数７】

に対応する。ただし、ここでは３次以上の高次成分を除いている。また、（ｅ）と（ｇ）および（ｈ）の関係は、
【数８】

に対応し、（ｆ）と（ｉ）および（ｊ）の関係は、
【数９】

に対応する。すなわち、（ｃ）に示される位相変調が、（ｄ）、（ｇ）、（ｈ）、（ｉ）、（ｊ）のフェーザに示される位相を有する周波数成分に分解される。これら各周波数成分の有する位相を同一軸上に並べたものが（ａ）である。故に、以降では（ａ）の表現形式で各スペクトル成分の位相を表現する。
【００１５】
この表現形式を用いて、ＤＳＢ−ＳＣ変調の原理を詳細に説明する。
ＤＳＢ−ＳＣ法を行うＭＺＭ構成を図３７に示す。１電極ＭＺＭ変調器は、内部の２つの位相変調器に逆相の電界を供給する。ＤＣバイアスは一方の伝搬光にのみ位相変化を与える。図中の（ａ）〜（ｃ）に対応する光スペクトル模式図を図３８に示す。図３８では、図３７の（ａ）位置の透過光搬送波のもつ位相および入力ＲＦ電界の位相を基準としている。図３８（ｃ）より変調器出力光が奇数次成分を有することが分かり、これは式５と対応する。
なお、図３８（ｃ）で上下１次変調成分間ビートの持つ位相と、１次変調成分と３次変調成分間ビートが持つ位相とが互いにπ異なることが図示されているが、これは分散の影響を受けないｂａｃｋ−ｔｏ−ｂａｃｋ伝送において出力ＲＦ強度が最小となることを示している。実用面ではこの高次側波帯の影響を低減する為に入力ＲＦ強度を低下させ、図３３のＪ_３ 成分が小さい範囲において駆動することが必要となる。これにより、図３８（ｃ）における３つのミリ波・マイクロ波帯ビート成分を近似的に１つとしている。
このようにＤＳＢ−ＳＣ法は、２モード光スペクトル伝送であるため、有効なミリ波・マイクロ波帯ビート成分数は１である。故に、ヘテロダイン受光により得られるミリ波・マイクロ波信号は光ファイバ分散の影響を受けにくい。
【００１６】
ＤＳＢ−ＳＣ法を用いた光ミリ波・マイクロ波信号生成の特徴は要約すると主に以下のようになる。
（ａ） 周波数逓倍機能：
図３８（ｃ）からもわかるように、１次の上下側波帯成分のビートを出力信号とするので、必要とされる出力信号の周波数の半分の周波数で光位相変調器を変調すればよく、ＲＦ信号生成回路が簡便になる。
（ｂ） 高分散耐性：
図３８（ｃ）からもわかるように、入力ＲＦ信号の強度を下げ３次変調成分を低減することにより、１次の上下側波帯成分のみが光ファイバー中に伝送されるので、ビート信号は１種類に限られ、従って光ファイバの分散ペナルティが発生しない。
ＤＳＢ−ＳＣ法を用いた光ミリ波・マイクロ波信号生成は、１次の上下側波帯成分を光ファイバー中に伝送することから、２モード光ミリ波・マイクロ波信号生成方法とも呼ばれる。
２モード光ミリ波・マイクロ波信号生成方法は上記のように優れた特徴を有することから、２モード光ミリ波・マイクロ波信号生成方法を用いたＰＳＫ（Ｐｈａｓｅ Ｓｈｉｆｔ Ｋｅｙｉｎｇ）変調方法、すなわち、２モード光ミリ波・マイクロ波ＰＳＫ信号生成法が研究（非特許文献４、５参照）されている。
【００１７】
次に、従来の２モード光ミリ波・マイクロ波信号生成法の課題を説明する。
従来の２モード光ミリ波・マイクロ波ＰＳＫ信号生成の原理を説明する。
ＤＳＢ−ＳＣ動作により生成される電界は、高次変調成分を除くと式５より、次式で表される。なお、この式は、入力ＲＦ信号強度が低い場合に正しい。
【数１０】

これをフォトダイオード（ＰＤ）でヘテロダイン受光することで得られる光電流の目的とするミリ波・マイクロ波信号成分は、
【数１１】

となる。これにより入力信号の二倍波が生成されるが、位相偏移も同時に二倍となる。このために、例えば、入力信号としてＰＳＫ信号を用いた場合に、ＰＤ受光後にＰＳＫ信号が得られない。そこで、目的とするＰＳＫ信号の１／周波数逓倍数の位相偏移を有する信号を生成する。これは、図３９に示す回路で実現される。光スペクトル模式図は図４０で表現される。
【００１８】
しかしながら、この従来方法には、以下に示す課題がある。
（ａ）入力信号として独立なＢＰＳＫ信号２つを重畳する場合、すなわち、ＱＰＳＫ信号を生成する場合、２つのＰＳＫ信号の混信成分が生成されてしまい、直交するＢＰＳＫ信号の加算によるＱＰＳＫ信号の生成が不可能である。
（ｂ）ＢＰＳＫ信号のシンボル１とシンボル０間の過渡遷移過程に伴うスペクトル拡大が生じ、周波数利用効率の低下、及び符合誤り率の上昇が生じる。
（ｃ）変調器入力端において生成する信号の変調フォーマットと、ヘテロダイン受光後に得られる信号の変調フォーマットが異なるために、任意の変調フォーマットを利用することが不可能となる。
【００１９】
【非特許文献１】
Ｕ．Ｇｌｉｅｓｅ，Ｓ．Ｎｏｒｓｋｏｖ ａｎｄ Ｔ．Ｎ．Ｎｉｅｌｓｅｎ，“Ｃｈｒｏｍａｔｉｃ Ｄｉｓｐｅｒｓｉｏｎ ｉｎ Ｆｉｂｅｒ−ＯｐｔｉｃＭｉｃｒｏｗａｖｅ ａｎｄ Ｍｉｌｌｉｍｅｔｅｒ−Ｗａｖｅ Ｌｉｎｋｓ”，ＩＥＥＥ Ｔｒａｎｓ．Ｍｉｃｒｏｗａｖｅ Ｔｈｅｏｒｙ Ｔｅｃｈ．，Ｖｏｌ．４４，ｐｐ．１７１６−１７２４，１９９６．
【非特許文献２】
Ｕ．Ｇｌｉｅｓｅ，Ｔ．Ｎ．Ｎｉｅｌｓｅｎ，Ｓ．Ｎｏｒｓｋｏｖ ａｎｄ Ｋ．Ｅ．Ｓｔｕｂｋｊａｒ，“Ｍｕｌｔｉｆｕｎｃｔｉｏｎａｌ ｆｉｂｅｒ−ｏｐｔｉｃ ｍｉｃｒｏｗａｖｅ ｌｉｎｋｓ ｂａｓｅｄ ｏｎ ｒｅｍｏｔｅ ｈｅｔｅｒｏｄｙｎｅ ｄｅｔｅｃｔｉｏｎ”，ＩＥＥＥ Ｔｒａｎｓ． Ｍｉｃｒｏｗａｖｅ Ｔｈｅｏｒｙ Ｔｅｃｈ．，Ｖｏｌ．４６，ｐｐ．４５８−４６８，１９９８．
【非特許文献３】
Ｊ．Ｊ．Ｏ’Ｒｅｉｌｌｙ，Ｐ．Ｍ．Ｌａｎｅ，Ｒ．Ｈｅｉｄｅｍａｎｎ ａｎｄ Ｒ．Ｈｏｆｓｔｅｔｔｅｒ，“Ｏｐｔｉｃａｌ Ｇｅｎｅｒａｔｉｏｎ ｏｆ Ｖｅｒｙ Ｎａｒｒｏｗ Ｌｉｎｅｗｉｄｔｈ Ｍｉｌｌｉｍｅｔｒｅ Ｗａｖｅ Ｓｉｇｎａｌｓ”，Ｅｌｅｃｔｒｏｎ．Ｌｅｔｔ．，Ｖｏｌ．２８，ｐｐ．２３０９−２３１１，１９９２．
【非特許文献４】
Ｙ．Ｏｚｅｋｉ，Ｍ．Ｋｉｓｈｉ ａｎｄ Ｍ．Ｔｓｕｃｈｉｙａ，“Ｄｕａｌ−Ｍｏｄｅ Ｏｐｔｉｃａｌ ＭＭ−Ｗａｖｅ ＰＳＫ Ｓｉｇｎａｌ Ｇｅｎｅｒａｔｉｏｎ Ｔｅｃｈｎｉｑｕｅ ｆｏｒ ６０ ＧＨｚ−Ｂａｎｄ Ｆｉｂｅｒ−Ｒａｄｉｏ Ｓｙｓｔｅｍｓ”， Ｔｅｃｈｎｉｃａｌ Ｄｉｇｅｓｔ ｏｆ ｔｈｅ ２ｎｄ Ｋｏｒｅａ−Ｊａｐａｎ Ｊｏｉｎｔ Ｗｏｒｋｓｈｏｐ ｏｎ Ｍｉｃｒｏｗａｖｅ−Ｐｈｏｔｏｎｉｃｓ，ＩＶ−３，ｐｐ．１２５−１２８，Ｓｅｏｕｌ，Ｆｅｂｒｕａｒｙ ２００１．
【非特許文献５】
小関泰之，岸眞人，土屋昌弘「マッハツェンダー型光変調器による６０ＧＨｚ帯２モード光ミリ波・マイクロ波ＰＳＫ信号生成法」２００１年電子情報通信学会総合大会，立命館大学．
【非特許文献６】
Ｍ．Ｉｚｕｔｓｕ，Ｓ．Ｓｈｉｋａｍａ，ａｎｄ Ｔ．Ｓｕｅｔａ，“Ｉｎｔｅｇｒａｔｅｄ Ｏｐｔｉｃａｌ ＳＳＢ Ｍｏｄｕｌａｔｏｒ／ Ｆｒｅｑｕｅｎｃｙ Ｓｈｉｆｔｅｒ”，ＩＥＥＥ Ｊ．Ｑｕａｎｔｕｍ Ｅｌｅｃｔｒｏｎ．，Ｖｏｌ．ＱＥ−１７，ｐｐ．２２２５−２２２７，１９８１．
【非特許文献７】
Ｓ．Ｓｈｉｍｏｔｓｕ，Ｓ．Ｏｉｋａｗａ，Ｔ．Ｓａｉｔｏｕ，Ｎ．Ｍｉｔｓｕｇｉ，Ｋ．Ｋｕｂｏｄｅｒａ，Ｔ．Ｋａｗａｎｉｓｈｉ ａｎｄ Ｍ．Ｉｚｕｔｓｕ，“Ｓｉｎｇｌｅ Ｓｉｄｅ−Ｂａｎｄ Ｍｏｄｕｌａｔｉｏｎ Ｐｅｒｆｏｒｍａｎｃｅ ｏｆ ａ ＬｉＮｂＯ_３ Ｉｎｔｅｇｒａｔｅｄ Ｍｏｄｕｌａｔｏｒ Ｃｏｎｓｉｓｔｉｎｇ ｏｆ Ｆｏｕｒ−Ｐｈａｓｅ Ｍｏｄｕｌａｔｏｒ Ｗａｖｅｇｕｉｄｅｓ”，ＩＥＥＥ Ｐｈｏｔｏｎｉｃｓ Ｔｅｃｈ． Ｌｅｔｔ．，Ｖｏｌ．１３，ｐｐ．３６４−３６６，２００１．
【非特許文献８】
Ａ．Ｎａｒａｓｉｍｈａ，Ｘ．Ｊ．Ｍｅｎｇ，Ｍ．Ｃ．Ｗｕ ａｎｄ Ｅ．Ｙａｂｌｏｎｏｖｉｔｃｈ，“Ｔａｎｄｅｍ ｓｉｎｇｌｅ ｓｉｄｅｂａｎｄ ｍｏｄｕｌａｔｉｏｎ ｓｃｈｅｍｅ ｆｏｒ ｄｏｕｂｌｉｎｇ ｓｐｅｃｔｒａｌ ｅｆｆｉｃｉｅｎｃｙ ｏｆ ａｎａｌｏｇｕｅ ｆｉｂｒｅ ｌｉｎｋｓ”，Ｅｌｅｃｔｒｏｎｉｃｓ Ｌｅｔｔｅｒｓ，Ｖｏｌ．３６，ｐｐ．１１３５−１１３６，２０００．
【００２０】
本発明は上記課題に鑑み、信号の混信成分が生じない２モード光ミリ波・マイクロ波ＱＰＳＫ信号生成方法を提供することを第１の目的とする。
また、スペクトル拡大が生じない２モード光ミリ波・マイクロ波ＢＰＳＫ信号生成方法を提供することを第２の目的とする。
さらに任意の変調フォーマットを利用することができる光ミリ波・マイクロ波信号生成方法を提供することを第３の目的とする。
【００２１】
【課題を解決するための手段】
上記目的を達成するため、本発明の光ミリ波・マイクロ波信号生成方法は、単一波長光を第１の光と第２の光に分割し、分割した第１の光と第２の光をそれぞれ、第１の光変調器と第２の光変調器に入力し、第１の光変調器と第２の光変調器にそれぞれ、第１のＲＦ信号と第２のＲＦ信号を入力して第１の光と第２の光をそれぞれＤＳＢ（ｄｏｕｂｌｅ ｓｉｄｅ ｂａｎｄ）変調し、ＤＳＢ変調された第１の光と第２の光の間に固定位相π／２を付加して加算することからなり、加算された第１の光と第２の光のヘテロダイン受光出力において、第１のＲＦ信号と第２のＲＦ信号の混信成分が抑圧されることを特徴とする。
この方法によれば、複数のミリ波・マイクロ波帯信号を混信無く伝送することができる。
【００２２】
また、単一波長光の分割、加算手段、光変調手段、及び固定位相π／２の付加手段は、電気光学結晶上に集積して形成した、分岐光導波路、光導波路に電界を印加して光導波路を伝搬する光に位相を付加する複数の光位相変調器で構成した光ＳＳＢ変調器、及び光導波路に一定電界を印加するπ／２固定位相器であればよい。
または、単一波長光の分割、加算手段、光変調手段、及び固定位相π／２の付加手段は、電気光学結晶上に集積して形成した、分岐光導波路、光導波路に電界を印加して光導波路を伝搬する光に位相を付加する複数の光位相変調器で構成した２電極ＭＺＭ光変調器光、及び光導波路に一定電界を印加するπ／２固定位相器であり、かつ、第１の光と第２の光の加算後に、搬送波を除去するための光帯域阻止フィルターを有していればよい。
上記の方法によれば、ヘテロダイン受光出力は、第１のＲＦ信号及び第２のＲＦ信号それぞれの、上下側波帯の１次変調成分間のヘテロダイン出力であり、第１のＲＦ信号及び第２のＲＦ信号それぞれの周波数の２倍の周波数を持ち、かつ、第１のＲＦ信号及び第２のＲＦ信号それぞれの入力位相の２倍の位相を有している。
また、ヘテロダイン受光出力は、第１のＲＦ信号及び第２のＲＦ信号それぞれの、上下側波帯の２次変調成分間のヘテロダイン出力であっても良く、第１のＲＦ信号及び第２のＲＦ信号それぞれの周波数の４倍の周波数を持ち、かつ、第１のＲＦ信号及び第２のＲＦ信号それぞれの入力位相の４倍の位相を有していても良い。
【００２３】
また、上記の２倍の周波数及び位相が得られる方法において、第１のＲＦ信号と第２のＲＦ信号を位相位相偏移量がπ／２である互いに独立なＮＡ−ＢＰＳＫ（ｎａｒｒｏｗ ａｎｇｌｅ ｂｉｎａｒｙ ｐｈａｓｅ ｓｈｉｆｔ ｋｅｙｉｇ）信号とし、かつ第１のＲＦ信号と第２のＲＦ信号が互いにπ／４の位相差を有するように遅延することにより、ヘテロダイン受光において発生する２つのＢＰＳＫ信号間にπ／２の位相差が生じ、この２つのＢＰＳＫ信号が加算されてＱＰＳＫ（ｑｕａｄｒａｔｕｒｅ ｐｈａｓｅ ｓｈｉｆｔ ｋｅｙｉｎｇ）信号が得られることを特徴とする。
また、上記の４倍の周波数及び位相が得られる方法において、第１のＲＦ信号と第２のＲＦ信号を位相位相偏移量がπ／４である互いに独立なＮＡ−ＢＰＳＫ信号とし、かつ第１のＲＦ信号と第２のＲＦ信号が互いにπ／８の位相差を有するように遅延することにより、ヘテロダイン受光において発生する２つのＢＰＳＫ信号間にπ／２の位相差が生じ、この２つのＢＰＳＫ信号が加算されてＱＰＳＫ信号が得られることを特徴とする。
【００２４】
また、上記の２倍の周波数及び位相が得られる方法において、第１のＲＦ信号と第２のＲＦ信号を位相偏移量がπ／２であり、かつ、互いに逆の位相変化をするＮＡ−ＢＰＳＫ信号とし、かつ、第１のＲＦ信号と第２のＲＦ信号が互いにπ／２の位相差を有するように遅延することにより、ヘテロダイン受光において発生する２つのＢＰＳＫ信号が、上記信号のシンボル間の過渡遷移状態において互いに不要成分を相殺し合い、スペクトル広がりを防ぐことを特徴とする。
また、上記の４倍の周波数及び位相が得られる方法において、第１のＲＦ信号と第２のＲＦ信号を位相位相偏移量がπ／４であり、かつ、互いに逆の位相変化をするＮＡ−ＢＰＳＫ信号とし、かつ、第１のＲＦ信号と第２のＲＦ信号が互いにπ／４の位相差を有するように遅延することにより、ヘテロダイン受光において発生する２つのＢＰＳＫ信号が、上記信号のシンボル間の過渡遷移状態において互いに不要成分を相殺し合い、スペクトル広がりを防ぐことを特徴とする。
【００２５】
また、第１のＲＦ信号と第２のＲＦ信号の周波数を互いに異ならせることにより、ヘテロダイン受光において発生する２つ信号の周波数が異なり、この２つの信号間の混信成分が抑圧されているため、副搬送波多重が可能であることを特徴とする。
【００２６】
また、第１のＲＦ信号を、変調を受けていない第３のＲＦ信号と任意変調方式により変調された第５のＲＦ信号に９０°位相変化を与えて得られる第６のＲＦ信号との加算により生成し、かつ、第２のＲＦ信号を、第３のＲＦ信号に９０°位相変化を与えて得られる第４のＲＦ信号と第５のＲＦ信号との加算により生成し、生成した２つの信号をそれぞれの光変調器に入力することにより、ヘテロダイン受光において、第３のＲＦ信号の周波数と第５のＲＦ信号の周波数との和の周波数を有し、かつ、第５のＲＦ信号に対して与えられた変調方式による変調を受けた信号を得ることを特徴とする。
【００２７】
【発明の実施の形態】
以下、図面に基づき本発明の実施の形態を詳細に説明する。なお、実質的に同一の部材には同一の符号を付して説明する。
初めに、本発明の第１の実施の形態を説明する。
図１は、光変調器としてＯＳＳＢＭ（光ＳＳＢ変調器）を用いた場合の本発明の光ミリ波・マイクロ波信号生成方法の構成を示す図である。単一波長レーザ光源等である単一波長光源１からの光を電気光学結晶２上に形成された分岐光導波路３に入力する。分岐光導波路３で分岐された光の一方を、分岐光導波路３の一方の分岐３ａを介して光変調器４に入力する。分岐された光の他方を、分岐光導波路３の他方の分岐３ｂを介し、かつ、π／２固定位相器６を介して光変調器５に入力する。光変調器４で光位相変調された光、及び光変調器５で光位相変調された光を分岐光導波路７の一方の分岐７ａ及び他方の分岐７ｂを介して、分岐光導波路７で加算し、伝送用の光ファイバ８に入力し、光ファイバ８の出力端に設けたフォトダイオード等である光検出器９に入力する。
光変調器４、及び５はＭＺＭであり、ＭＺＭを２つ用いる構成はＯＳＳＢＭ（光ＳＳＢ変調器）と呼ばれる。光変調器４は、分岐光導波路１０の分岐１０ａに光位相変調器１１ａを有しており、分岐光導波路１０の分岐１０ｂにπ固定位相器１２と光位相変調器１１ｂを直列に有している。光変調器５は光変調器４と同一の構成である。
【００２８】
この構成で動作させるには、ＲＦ入力信号１３を、ＲＦ用π位相器１４を用いて、光変調器４の光位相変調器１１ａ，１１ｂに互いに逆相で入力する。同様に、ＲＦ入力信号１５を、ＲＦ用π位相器１４を用いて、光変調器５の光位相変調器１１ａ，１１ｂに互いに逆相で入力する。
【００２９】
図２は光変調器としてＤＥ−ＭＺＭ（２電極マッハツェンダー光変調器）を用いた場合の本発明の光ミリ波・マイクロ波信号生成方法の構成を示す図である。この構成は、図１の変調器４，５に光位相変調器１１ａ，１１ｂを使用したものであり、ＤＥ−ＭＺＭと呼ばれる。光分岐導波路３の一方の分岐３ａに光位相変調器１１ａを有し、他の分岐３ｂに光位相変調器１１ａとπ／２固定位相器６を有している。光変調器４，５により光変調された光が光分岐導波路６により加算された後に、搬送波（光源１の光）を除去するための光帯域阻止フィルター１６が設けられている。光帯域阻止フィルター１６は、例えばファイバブラッググレーティング（ＦＢＧ）である。他の構成は図１と同じである。
この構成で動作させるには、ＲＦ入力信号１３及び１５をそれぞれ、光変調器４及び５に入力する。
【００３０】
図１または図２の構成によれば、以下のように作用する。
光源１から出力される光の光搬送波周波数をΩとし、ＲＦ信号１３とＲＦ信号１５の周波数をω_１ とし、ＲＦ信号１３とＲＦ信号１５の電界強度、及び位相をそれぞれ、Ｅ_１ ，Ｅ_２ 及びφ_１ ，φ_２ とすれば、光変調器４，５から出力される上下１次側波帯はそれぞれ、
Ｅ_１ ｅｘｐｊ｛（Ω−ω_１ ）ｔ−φ_１ ｝、
Ｅ_２ ｅｘｐｊ｛（Ω＋ω_２ ）ｔ＋φ_２ ｝
と表される。上記一方の１次側波帯にπ／２固定位相器６によってπ／２の位相を付加して加算し、加算された上下１次側波帯をフォトダイオード８でヘテロダイン受光することによって得られる光電流ｉ_ｐ は次式となる。
【数１２】

この式から明らかなように、ＲＦ信号１３とＲＦ信号１５を混信無く取り出すことができる。
【００３１】
次に、本発明の第１の目的である、信号の混信成分が生じない２モード光ミリ波・マイクロ波ＱＰＳＫ信号生成方法を説明する。
本発明の２モード光ミリ波・マイクロ波ＱＰＳＫ信号生成方法で用いるＱＰＳＫ信号について説明する。ＱＰＳＫ信号は直交位相偏移変調（Ｑｕａｄｒａｔｕｒｅ Ｐｈａｓｅ Ｓｈｉｆｔ Ｋｅｙｉｎｇ）と四相位相偏移変調（Ｑｕａｔｅｒｎａｒｙ Ｐｈａｓｅ Ｓｈｉｆｔ Ｋｅｙｉｎｇ）と呼ばれる２つがある。前者は、直交なＢＰＳＫの加算により実現される。この直交位相偏移変調信号の生成回路構成を図３（ａ）に示し、フェーザを同図（ｂ）のＩ／Ｑダイアグラムに示す。一方、後者は位相変調器を用いて生成される。その構成例を図４（ａ）に示し、フェーザを同図（ｂ）に示す。両者の違いは、シンボル間の遷移過程にある。詳細は後述するが、シンボル間の遷移過程における位相変動成分がスペクトル拡がりを誘起する。四相位相偏移変調は、特にスペクトル拡がりが大きい。このため、本発明では、直交位相偏移変調を用いる。
【００３２】
図５は、本発明の２モード光ミリ波・マイクロ波ＱＰＳＫ信号生成方法の原理・構成を説明する図である。
集積型光変調器は、図１または図２で説明したように、ＯＳＳＢＭ（光ＳＳＢ変調器；非特許文献６，７参照）、またはＤＥ−ＭＺＭ（２電極ＭＺＭ）を用いる。ただし、ＤＥ−ＭＺＭを用いる場合には、ＤＳＢ−ＳＣ動作を得られないため、搬送波を抑圧する為に、ファイバブラッググレーティングのような光帯域阻止フィルタ（ＢＥＦ：ｂａｎｄ ｅｌｉｍｉｎａｔｉｏｎ ｆｉｌｔｅｒ）を用いる。
入力ＲＦ信号は、周波数・位相の識別子ｉを用いてそれぞれω_ｉ 、φ_ｉ と表現されるＲＦ信号であるとし、この図においては、ω_１ ＝ω_２ 、φ_１ ＝（２ｍ−１）π／４、φ_２ ＝ｍπ／２とし、ｍ及びｎは、０または１の値を持つデジタル信号とする。入力端において２つのＢＰＳＫ信号間に４５°の位相差を設ける。また、２つの変調器間に設ける固定位相差をφ_ｂｉａｓとする。
【００３３】
この構成によれば、変調器の非線形性により２つのＢＰＳＫ信号間位相差は２倍され、直交なＢＰＳＫ信号が出力端において得られる。光搬送波周波数Ωに対し、上下１次側波帯は、
Ｅ_ｉ ｅｘｐｊ｛（Ω−ω_ｉ ）ｔ−φ_ｉ ｝、
Ｅ_ｉ ｅｘｐｊ｛（Ω＋ω_ｉ ）ｔ＋φ_ｉ ｝
と表される。
入力ＲＦ信号と光スペクトルの対応は次式となる。
【数１３】

この光スペクトルをヘテロダイン受光することによって得られる光電流ｉ_ｐ は次式となる。
【数１４】

この式にはデータ信号を含む位相φ_１ ，φ_２ をともに有する混信成分が生成されており、従来方法で述べた２つのＰＳＫ信号の混信成分に対応する。入力ＮＡ−ＢＰＳＫ（Ｎａｒｒｏｗ Ａｎｇｌｅ Ｂｉｎａｒｙ Ｐｈａｓｅ Ｓｈｉｆｔ Ｋｅｙｉｎｇ）信号により生成される２つのＢＰＳＫ信号が出力端において混信する可能性があることを、再度スペクトル模式図を用いて示す。
【００３４】
図６に、φ_１ およびφ_２ に情報をもつ２つのＲＦ信号、ＲＦ１およびＲＦ２を各変調器に入力した際の光スペクトル模式図を示す。
ＲＦ１およびＲＦ２は、それぞれ上側波帯および下側波帯を生成するが、これらをＵＳＢ１、ＬＳＢ１およびＵＳＢ２、ＬＳＢ２とする。ＬＳＢ１とＵＳＢ１のヘテロダイン受光、およびＬＳＢ２とＵＳＢ２のヘテロダイン受光により生成されるＲＦ信号は２φ_１ および２φ_２ の位相情報を有する。しかしながら、ここでＵＳＢ１とＬＳＢ２、およびＵＳＢ２とＬＳＢ１のヘテロダイン受光により２φ_１ ＋２φ_２ の位相を有するＲＦ信号も同時に出力される。
【００３５】
本発明の方法は、φ_ｂｉａｓをπ／２にし、この混信成分を除去することを特徴とする。すなわち、この２つの混信成分を互いに逆相にして相殺する。すなわち、ＤＣバイアスにより制御される２導波路間位相差φ_ｂｉａｓをπ／２とすることで、混信成分のうち一方がπ／２の進み位相を持ち、他方がπ／２の遅れ位相を持たせる。これにより、両成分が互いに逆相となる。この時の変調器内部における光スペクトルの位相関係は図７に示される関係となる。式１４において、φ_ｂｉａｓ＝π／２とすれば、ヘテロダイン受光することによって得られる光電流ｉ_ｐ は式１２となる。これにより、光変調器への入力信号が出力端において混信無く加算され、２つのＢＰＳＫ信号を混信無く取り出すことができる。
なお、式１２の導出でω_１ ＝ω_２ としたが、ω_１ ≠ω_２ であっても良く、この場合に、ヘテロダイン受光することによって得られる光電流ｉ_ｐ は次式となる。
【数１５】

すなわち、ＢＰＳＫ信号を副搬送波多重（ＳＣＭ：ｓｕｂ−ｃａｒｒｉｅｒ ｍｕｌｔｉｐｌｅｘ）しても混信成分が生じない。一般にＭＺＭ変調器を用いる際には、変調器の非線形性のためＳＣＭ時には不要成分発生が不可避であるが、この方式であれば理論的に不要成分の発生が抑圧できる。
【００３６】
次に、第１の実施形態の実施例を説明する。
６０ＧＨｚ帯のＲＦ周波数で、光位相変調器としてＯＳＳＢＭ、及びＤＥ−ＭＺＭを用いる方法それぞれについて実施した。
図８は、ＯＳＳＢＭを用いる実験系を示す図である。ＯＳＳＢＭは、ＬｉＮｂＯ_３ ｘ−ｃｕｔ２電極の変調器である。ＰＰＧ（ｐｕｌｓｅ ｐａｔｔｅｒｎ ｇｅｎｅｒａｔｏｒ）より出力された１５６Ｍｂｐｓ擬似ランダム（ＰＲＢＳ：ｐｓｅｕｄｏ ｒａｎｄｏｍ ｂｉｔ ｓｅｑｕｅｎｃｅ）信号をもとに生成した１ＧＨｚ帯ＮＡ−ＢＰＳＫ信号を、３０ＧＨｚ帯にアップコンバートし、透過帯域幅６００ＭＨｚＢＰＦを用いてイメージ成分を除いた。２ＮＡ−ＢＰＳＫ信号の生成は、この信号を２分岐し、一方を１ビット長の時間遅延を加えることで実現した。２信号間には、３０ＧＨｚ成分のπ／４にあたる遅延を加えている。光源としては波長可変レーザを用い、この出力光を偏波コントローラ（ＰＣ）を介してＯＳＳＢＭに入力した。本方式による変調が与えられた出力光を、ＥＤＦＡ（Ｅｒｂｉｕｍ Ｄｏｐｅｄ Ｆｉｂｅｒ Ａｍｐ）及びＢＰＦを介してＰＤ受光した。ここでは受光素子として進行波型フォトダイオードを用い、受光後のミリ波・マイクロ波帯信号を増幅した。これを２ＧＨｚ帯にダウンコンバートし、直交復調によりＩ／Ｑダイアグラムを得た。
【００３７】
以下に測定結果を示す。ＢＰＦ出力光スペクトルを図９に示す。
（ａ）は、ＮＡ−ＢＰＳＫ信号による変調を受けた２モードであるのに対し、（ｂ）は無変調ＲＦ信号による変調を受けた２モードである。いずれにも上下側波帯間に強度差が存在している。前者の強度比は５．９ｄＢであり、後者のそれは７．２ｄＢである。この強度差はＤＣバイアスにより設定したφ_ｂｉａｓ＝９０°の位相差と、２入力ＲＦ信号間に設けた４５°位相差により発生している。すなわち、前者は２信号間の位相差を両側波帯に対して９０°に設定するのに対し、後者は２信号間の位相差を上側波帯に対して−４５°、下側波帯に対して＋４５°に設定する。それ故、上側波帯（短波長側）では強めあう干渉がおき、下側波帯（長波長側）では弱めあう干渉がおきている。シングルトーン変調により得られた光２モード間には７．２ｄＢであり、これは理論値、
【数１６】

とほぼ一致する。
この２モード光スペクトルをＰＤでヘテロダイン受光して得られたＲＦ信号のスペクトルを図１０に示す。６０ＧＨｚ帯の信号生成は確認されたが、スペクトル上に周期的な歪みが見られた。これは、ＯＳＳＢＭ内の２つの導波路間において定常的な伝搬時間差があるからと考えられる。この時間差は実験系を簡素化するために、単一ＰＰＧより生成された１つのＮＡ−ＢＰＳＫ信号から２つのＮＡ−ＢＰＳＫ信号を作成したことによる。すなわち、ここで１ビット程度の時間差を電気回路に加えたが為に、光回路においてその逆数の１５０ＭＨｚ程度の周期を有するマッハツェンダフィルタとなったと考えられる。これは、異なるデジタルデータを有するＮＡ−ＢＰＳＫ信号ならば発生しない。
この６０ＧＨｚ帯ＱＰＳＫ信号を２ＧＨｚ帯にダウンコンバートして得られたフェーザを図１１（ｂ）に示す。同図（ａ）はＯＳＳＢＭへの入力ＮＡ−ＢＰＳＫ信号である。これらの図からＱＰＳＫ信号の生成が確認されたことは明らかであり、本発明が実証された。
【００３８】
次に、光位相変調器としてＤＥ−ＭＺＭを用いた実施例を説明する。
本発明の方法は、光位相変調により生成された上下側波帯のビートを得るものであり、ＯＳＳＢＭを用いなくともＱＰＳＫ信号を得ることは可能である。ＯＳＳＢＭの利用目的は、同変調器内部のＭＺＭに対し搬送波抑圧動作を与えることにある。ゆえに、ＤＥ−ＭＺＭによる変調と光帯域阻止フィルタによる搬送波除去によっても可能である。これにより特殊な変調器であるといえるＯＳＳＢＭを用いなくとも、商用のＤＥ−ＭＺＭでも可能となるため、利便性は高い。
【００３９】
実験系は、位相変調器の構成を除いて図８と同じである。位相変調器構成には、図１２中（ａ）と（ｂ）の違いが存在する。
本実施例で生成した２モード光ミリ波・マイクロ波信号の光スペクトルを図１３に示す。ＯＳＳＢＭにより生成した光スペクトルとの大きな違いは、２次変調成分が生成されていることである。これはＯＳＳＢＭでは搬送波を含む偶数次側波帯の抑圧が可能であるが、この方式ではそれが不可能であることによる。しかしながら、この２次変調成分と目的とするミリ波・マイクロ波周波数離調をなす成分である搬送波は抑圧されている。そのため、分散ペナルティへの影響はない。この２モード光ミリ波・マイクロ波信号を受光し、ダウンコンバートの後に得たフェーザ、ＢＥＲを図１４に示す。図から明らかなように、受光強度−１７ｄＢｍにおいてエラーフリー動作が実現しており、また、ＱＰＳＫ信号の生成が確認された。従って本発明が実証された。
【００４０】
次に、本発明の第２の実施形態を説明する。
この実施形態は、ＢＰＳＫ信号のシンボル１とシンボル０間の過渡遷移過程に伴うスペクトル拡大を抑圧する２モード光ミリ波・マイクロ波ＢＳＫ信号生成方法に関する。
はじめに、ＮＡ−ＢＰＳＫ信号のＭＺＭ入力により生成するＢＰＳＫ信号にはスペクトル拡大が生ずる要因について説明する。
ＮＡ−ＢＰＳＫ信号、およびこの二逓倍により生成されるＢＰＳＫ信号のフェーザを図１５に示す。ここで、シンボル１およびシンボル０間の過渡遷移過程に注目すると、生成されるＢＰＳＫ信号には位相変動成分がある。この成分は図１６に示すように、Ｉ軸並行な成分とＱ軸並行な成分とに分けられる。このうち後者がビットレートで以って変動する場合、前者はビットレートの２倍の速さで以って変動する。これが、スペクトル拡がりの原因となる。
【００４１】
図１７（ａ）は、１００Ｍｂｐｓで変調された基底帯域ＮＡ−ＢＰＳＫ信号を５０ＭＨｚのＬＰＦ（α＝０．２）透過後のスペクトルの計算結果である。同図（ｂ）は、これの二逓倍により得たＢＰＳＫ信号である。およそ１００ＭＨｚ程度の帯域幅が最大２００ ＭＨｚ程度まで拡がっており、搬送波が現れている。理想的なＢＰＳＫ信号には搬送波は存在しない。
【００４２】
ここで、前述のＩ軸平行成分とＱ軸平行成分の有するスペクトルを図１８に示す。この図からＩ軸平行成分がスペクトル拡大要因であることが明らかである。
【００４３】
次に、Ｉ軸平行成分によるスペクトル拡大を抑圧する方法について述べる。
図１９で示されているように、１つのＢＰＳＫ信号に対してＱ軸対称なＢＰＳＫ信号をＰＤ受光後に加算することで、Ｉ軸変動成分を抑圧する。ここでも用いる位相変調器としてはＯＳＳＢＭ、あるいはＤＥ−ＭＺＭである。後者では分散耐性を高めるために、ＦＢＧ等による搬送波除去が必要となる。
実施の形態１と同様に、ＯＳＳＢＭ、あるいはＤＥ−ＭＺＭに対して、２つのＮＡ−ＢＰＳＫ信号を入力する。しかし、ここでは実施の形態１と異なり、同一データを有する信号で、かつ互いに逆の位相変化を持つ信号を生成する。また、２つのＲＦ信号間の位相差は９０°とし、ＰＤによるヘテロダイン受光後に１８０°の位相差を得る。これによりＱ軸対称なＢＰＳＫ信号が得られる。
また、実施の形態１で説明したように、混信成分抑圧のためのＤＣバイアスφ_ｂｉａｓ＝π／２を印加する。ただし、ここでの混信成分は実施の形態１と異なり、異なるデータを持つ２信号の混信成分ではなく、同一データを持つ信号間の混信であり、これは出力ＰＳＫ信号に搬送波を持たせるものである。
この方法によれば、Ｉ軸対称なＢＰＳＫ信号とＱ軸対称なＢＰＳＫ信号の合成によって、シンボル１およびシンボル０間の過渡遷移過程におけるＩ軸平行成分がキャンセルされ、スペクトル拡大が抑止される。
【００４４】
次に、実施例を示す。
上記原理を確認するために、ＯＳＳＢＭを用いて１．５ＧＨｚ帯のＢＰＳＫ信号生成実験を行った。その実験系を図２０に示す。ＰＰＧより出力された２００ＭｂｐｓＰＲＢＳ信号をＬＰＦ１００ＭＨｚを介して７５０ＭＨｚ帯ＢＰＳＫ信号とする。ここで発生するサイドローブ成分を除去するために７５０ＭＨｚ帯で透過帯域幅３００ＭＨｚを有すＢＰＦを用いる。このＢＰＳＫ信号を増幅後、もとの７５０ＭＨｚシングルトーン信号と共に９０°ハイブリッド回路に入力することで、目的とする２つのＮＡ−ＢＰＳＫ信号が得られる。
ＤＦＢ−ＬＤ出力光に対して、この動作を施した後にＥＤＦＡ、ＢＰＦを介してＰＤ受光を行った。得られた１．５ＧＨｚ帯ＢＰＳＫ信号に対して直交復調を行い、Ｉ／Ｑダイアグラムを得た。得られたＲＦスペクトルを図２１に、Ｉ／Ｑダイアグラムを図２２に示す。図２１では（ａ）に示した従来法による帯域に比較して、（ｂ）に示した本発明の方法による占有ペクトル帯域は狭窄化しており、また搬送波成分も抑圧されていることがわかる。
【００４５】
図２２（ａ）および（ｂ）は従来法によるＢＰＳＫ信号である。ここでは、ＯＳＳＢＭへの入力ＲＦ信号の注入、非注入によって同フェーザを得た。同図（ｃ）は、（ａ）および（ｂ）に対応するＢＰＳＫ信号の加算により得られるＢＰＳＫ信号である。若干、シンボル間の軌跡が直線からずれているが、これは２ＢＰＳＫ信号間の強度の違いから生じていると考える。図２２（ｃ）から、本発明の原理が実証された。
【００４６】
次に、本発明の第３の実施の形態を説明する。
この実施形態は、信号の変調フォーマットによらずに、変調器入力端における変調方式と同一の変調方式を有す信号をヘテロダイン受光後に得ることが可能な光ミリ波信号生成方法に関する。
図２３は本発明の光ミリ波・マイクロ波信号生成方法の構成を示す図である。発振器出力信号を２分岐し、一方に任意フォーマットの変調を行う。これとシングルトーン信号を９０°ハイブリッド回路に入力する。９０°ハイブリッド回路は図２４に示すように、２入力端子（入力端子２１、入力端子２２）と２出力端子（出力端子２３、出力端子２４）を有すデバイスである。入力端子２１から入力された信号は２分岐され、分岐した一方の信号に９０°位相変化が与えられた後に出力端子２４に出力される。分岐した他方の信号は位相変化が与えられずに出力端子２３に出力される。入力端子２２から入力された信号も同様に２分岐され、一方に対して９０°位相変化が与えられた後に出力端子２３に出力され、他方に対しては位相変化が与えられずに出力端子２４に出力される。なお、第２の実施の形態では入力２信号間の強度・位相を調節し、ＮＡ−ＢＰＳＫ信号を得たが、本方式では９０°ハイブリッド回路を用いるので厳密な調整の必要はない。厳密な振幅位相調整を必要とする際には、デバイダ、位相シフタ、アッテネータ等を用いてこの機能を実現する。
【００４７】
なお、９０°ハイブリッドを介して２電極ＭＺＭにＲＦ信号を入力するものとして従来例（非特許文献８参照）があるが、この従来例は、同一光に対して異なる情報を有する２つのＲＦ信号を変調することを目的としているので、９０°ハイブリッド回路に入力する２信号はともに変調を受けており、本発明と異なる。本発明では、光搬送波と変調側波帯とのビート信号をヘテロダイン受光によって得るのではなく、上下側波帯のビート信号を得るものである。
【００４８】
第２の実施の形態のＢＰＳＫ変調法はこの方式の一条件下での変調法であると言える。そこで、第２の実施の形態におけるＮＡ−ＢＰＳＫ信号生成過程と光側波帯がシングルトーンとなる過程をここで示す。
図２５および図２６が、ＲＦ信号フェーザおよび上下側波帯の有すフェーザである。なお、（ａ）〜（ｉ）は図２３（ａ）〜（ｉ）に対応する。
図２６（ｉ）で上側波帯がシングルトーンであり、下側波帯がＢＰＳＫ変調を受けた光であることがわかる。このように図２３の構成をとることで、上下側波帯のうち一方はシングルトーンとなり、他方は変調をうけたスペクトルとなる。
これより、この変調法が２つのＳＳＢ変調からなる変調法と考えることができる。すなわち、ＳＳＢ変調器の通常利用法では、入力ＲＦ信号を、９０°ハイブリッド回路を介してＳＳＢ変調器に入力することにより、上側波帯あるいは下側波帯が生成される。その際、９０°進みか９０°遅れかにより、生成側波帯が上側波帯となるか下側波帯となるかが決まる。本方式では、シングルトーン信号と任意変調信号とを９０°ハイブリッド回路経由で光ＳＳＢ変調器に入力するが、その際に９０°ＲＦ位相差の進み、あるいは遅れが、シングルトーン信号と任意変調信号間で逆の関係にある。ゆえに、変調フォーマット無依存である。
しかしながら、ここで光変調器ＤＣバイアスのＲＦ強度依存性を考慮に入れると、ＱＰＳＫ信号生成時などにシンボルによってＲＦ強度が変化してしまう。例えば、図２５（ｃ）に対応するフェーザが、図２７（ａ）のようになった際にはＲＦ強度ゼロとなり、これによりＬＮ変調器のバイアスシフトが起こる可能性がある。そこで、図２７（ｂ）のようにシンボルを振幅のゼロ点からずらせばよい。これは、図２３（ａ）におけるＲＦ強度を（ｂ）におけるそれより大きくすることで実現可能である。
【００４９】
次に、第３の実施の形態の実施例を説明する。
上記原理を確認するために６０ＧＨｚ帯において実験を行った。この実験系を図２８に示す。
１５６ＭｂｐｓＰＲＢＳの２信号から７５０ＭＨｚ帯ＱＰＳＫ信号生成し、これを増幅およびフィルタの後、シングルトーンと共に９０°ハイブリッド回路に入力した。これにより得られるＩＦ帯２信号を２９．５ＧＨｚ帯にアップコンバートした後、増幅し、光ＳＳＢ変調器に入力した。この信号により変調を受けたＤＦＢ−ＬＤ出力光をＥＤＦＡ増幅、ＢＰＦの後に測定した光スペクトルを図２９に示す。搬送波が若干残っているが、これとミリ波・マイクロ波帯のビートを生成すべく２次側波帯は存在しないため、分散ペナルティへの影響は無視できる。また、光変調器入力時に２８ＧＨｚ帯イメージ成分除去を行っていないために上下側波帯はそれぞれ２以上の側波帯で構成されていると考えられるが、光スペクトラムアナライザの波長分解能が０．０１ｎｍ（１．２５ＧＨｚ程度）であるため、これは光スペクトル上では認識されていない。この２モード光ミリ波・マイクロ波信号をＵＴＣ−ＰＤによって受光した。５６ＧＨｚ帯イメージ成分等を除去する目的でＰＤ受光後に５９ＧＨｚ帯ＢＰＦを用いた。１．５ＧＨｚ帯へのダウンコンバートの後に得られたＩ／ＱダイアグラムおよびＩＦ帯スペクトルを図３０、図３１に示す。ここで、目的としたＱＰＳＫ信号が得られていることが確認できた。なお、この実験結果については変調動作およびノイズ除去等の最適化が不十分であることを付記する。
【００５０】
【発明の効果】
上記説明から理解されるように、本発明の光ミリ波・マイクロ波信号生成方法によれば、複数の信号を混信することなく伝送でき、例えばＱＰＳＫ信号を伝送できる。また、スペクトル拡大が生じることなくＢＰＳＫ信号を伝送できる。さらに、さらに任意の変調フォーマットを利用することができる。
本発明法を光ミリ波・マイクロ波アクセスシステムのダウンリンクにおける光ミリ波・マイクロ波信号生成に用いれば極めて有用である。
【図面の簡単な説明】
【図１】光変調器としてＯＳＳＢＭ（光ＳＳＢ変調器）を用いた場合の本発明の光ミリ波・マイクロ波信号生成方法の構成を示す図である。
【図２】光変調器としてＤＥ−ＭＺＭ（２電極マッハツェンダー光変調器）を用いた場合の本発明の光ミリ波・マイクロ波信号生成方法の構成を示す図である。
【図３】直交位相偏移変調よるＱＰＳＫ信号の生成回路構成、及びそのフェーザを示す図である。
【図４】四相位相偏移変調信号の生成回路構成、及びそのフェーザを示す図である。
【図５】本発明の第１の実施形態のＱＰＳＫ信号生成方法の構成を示す図である。
【図６】変調器内のスペクトルを模式的に示す図である。
【図７】ＤＣバイアスをπ／２としたときのスペクトルを模式的に示す図である。
【図８】光変調器にＯＳＳＢＭを用いた場合の実験系を示す図である。
【図９】実施例で得られたＢＰＦ出力光のスペクトルを示す図である。
【図１０】実施例で得られたＲＦ信号スペクトルを示す図である。
【図１１】ＯＳＳＢＭへの入力信号フェーザと出力信号フェーザを示す図である。
【図１２】ＯＳＳＢＭを用いた実験系とＤＥ−ＭＺＭを用いた実験系の違いを示す図である。
【図１３】ＤＥ−ＭＺＭを用いた場合の２モード光ミリ波・マイクロ波信号の光スペクトルを示す図である。
【図１４】実施例で得られたＱＰＳＫ信号の符合誤り率とフェーザを示す図である。
【図１５】ＮＡ−ＢＰＳＫ信号及び二逓倍されて生成するＢＰＳＫ信号のフェーザを示す図である。
【図１６】ＢＰＳＫ信号のシンボル間遷移過程の位相変動成分を示す図である。
【図１７】基底帯域ＮＡ−ＢＰＳＫ信号のスペクトルと、二逓倍されて得られたＢＰＳＫ信号のスペクトルを示す図である。
【図１８】二逓倍されて得られたＢＰＳＫ信号のスペクトルのＱ軸平行成分とＩ軸平行成分を示す図である。
【図１９】本発明の第２の実施形態であるＢＰＳＫ信号のスペクトル拡大を抑圧する方法を示す図である。
【図２０】実施例に用いた実験系を示す図である。
【図２１】実施例で得られたＲＦスペクトルを示す図である。
【図２２】実施例で得られたＩ／Ｑダイアグラムを示す図である。
【図２３】本発明の第３の実施形態である、任意の変調フォーマットで変調できる方法を示す図である。
【図２４】９０°ハイブリッド回路の動作を説明する図である。
【図２５】本発明の方法によるＢＰＳＫ信号生成の際の各部のフェーザを示す図である。
【図２６】本発明の方法によるＢＰＳＫ信号生成の際の各部のフェーザを示す図である。
【図２７】光変調器のＤＣバイアスのＲＦ強度依存性を顧慮したＱＰＳＫ信号フェーザを示す図である。
【図２８】実施例に用いた実験系を示す図である。
【図２９】実施例で得られたＲＦスペクトルを示す図である。
【図３０】実施例で得られたＩ／Ｑダイアグラムを示す図である。
【図３１】実施例で得られたＩＦ帯スペクトルを示す図である。
【図３２】ミリ波・マイクロ波帯周波数に相当する離調周波数を有する光スペクトルを示す概念図である。
【図３３】ベッセル関数を示す図である。
【図３４】位相及び周波数の包括的表現が可能なスペクトル模式図を説明する図である。
【図３５】位相及び周波数の包括的表現が可能なスペクトル模式図を説明する図である。
【図３６】位相及び周波数の包括的表現が可能なスペクトル模式図を説明する図である。
【図３７】光ＤＳＣ−ＳＣ法を行うＭＺＭの構成を示す図である。
【図３８】光ＤＳＣ−ＳＣ変調を受けた各部の光のスペクトル模式図を示す図である。
【図３９】ＮＡ−ＰＳＫ信号生成回路を示す図である。
【図４０】２モード光ミリ波・マイクロ波ＢＰＳＫ信号のスペクトル模式図
【符号の説明】
１ 単一波長光源（単一波長レーザーダイオード）
２ 電気光学結晶
３ 分岐光導波路
４ 光変調器
５ 光変調器
６ π／２固定位相器
７ 分岐光導波路
８ 光ファイバ
９ 光検出器（フォトダイオード）
１０ 分岐光導波路
１１ａ，１１ｂ 光位相変調器
１２ π固定位相器
１３ ＲＦ信号
１４ ＲＦ用π固定位相器
１５ ＲＦ信号
１６ 光帯域阻止フィルター（ファイバブラッググレーティング）[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method of transmitting a plurality of signals without interference, a method of transmitting a QPSK signal, a method of suppressing a spectrum spread caused by a PSK signal, and a method of suppressing a spectrum spread regardless of a signal format in an optical millimeter-wave / microwave signal transmission system. And a device therefor.
[0002]
[Prior art]
First, the background art of the present invention will be described.
In recent years, broadband wireless systems have been remarkable. In order to develop this flow in the future, it is desired to use a millimeter-wave / microwave (mm-W) band requiring abundant frequency resources. However, since this millimeter-wave / microwave band signal has a large propagation loss in the air and in the coaxial, the application of Radio-on-Fiber technology using an optical fiber and the use of many picocell-configured base station systems are required. desired.
For practical use of this system, it is important to reduce the dispersion penalty in the optical fiber and to simplify the base station. It is also desired that the central office reduce the driving frequency. As an optical millimeter-wave / microwave signal generation method satisfying these, there is a carrier suppression double sideband modulation method. In addition, when considering the use of arbitrary modulation formats such as BPSK signal and QAM signal, improvement of spectrum use efficiency, use of a single optical modulator, etc., candidates for the millimeter wave / microwave band signal generation method are extremely extreme. Is narrowed down to This is because the above-described modulation method is a modulation method that actively uses nonlinearity. The only candidate is the M-ZM input of the NA-PSK signal, but the presence of carrier and spectrum spread, the difficulty of using 16QAM signal, etc.
[0003]
Millimeter-wave / microwave-band frequencies have a wide range of frequencies as well as high frequencies of 30 to 300 GHz. Therefore, its attractiveness as a candidate for broadband wireless is great. This is because, according to Shannon's theorem, the communication capacity C is expressed as C = Wlog (1 + S / N) using the used bandwidth W and the signal-to-noise ratio S / N. However, a wired cable that distributes millimeter-wave / microwave signals from a central station to a base station poses a problem.
In general, candidates for a wired cable used as a backbone of wireless communication include a telephone line, a UTP (Unshielded Twisted Pair) cable such as an Ethernet (registered trademark) cable, a coaxial cable, and an optical fiber. Among these, UTP is inexpensive, but cannot transmit in the millimeter wave / microwave band because the band is about 20 MHz to several hundred MHz. On the other hand, a part of the coaxial cable has a transmission band up to a millimeter wave / microwave band, but a loss per unit distance is large, and propagation of about km is practically impossible. Therefore, transmission using an optical fiber that gives a loss of 0.2 dB / km to a frequency of 200 THz is a promising candidate.
However, the signal actually propagating in the optical fiber is not a signal having a millimeter-wave / microwave band as a carrier frequency, but has a carrier frequency of about 200 THz and a detuning frequency of the millimeter-wave / microwave band frequency. Optical signal. That is, light having a beat corresponding to a millimeter wave / microwave band frequency (see Non-Patent Documents 1 and 3) is transmitted through an optical fiber. The sideband obtained when the optical carrier is modulated in the millimeter wave / microwave band is also called a subcarrier.
When this light wave is input to a photodiode (PD), electrons having no response speed of THz or higher respond only to the envelope of the light wave. This makes it possible to extract a millimeter-wave / microwave signal from the optical signal. Extracting information from the beat in this manner is called heterodyne reception, and in particular, light reception is called heterodyne reception.
[0004]
There is coherent optical communication as optical communication for performing heterodyne light reception, which generally refers to an optical communication method in which information is carried on the amplitude and phase of light at a transmission end and information is obtained by heterodyne reception with reference light generated at a reception end. . With the recent narrowing of the light source line width, the practical use of this method is approaching. However, advanced optical phase control of the reference light generated at the light receiving end and the received light is necessary, and in this regard, the reference light is transmitted to the transmitting end. There is a difference from the optical millimeter-wave / microwave system generated by. Therefore, the heterodyne using the same light source is sometimes called a self-heterodyne.
[0005]
[Problems to be solved by the invention]
The present invention proposes a new optical millimeter-wave / microwave signal generation method, and can improve the optical millimeter-wave / microwave signal generation method in the downlink of an optical millimeter-wave / microwave access system. It can contribute to the simplification of the central office and the simplification of the central office. The improvement means mainly consists in changing the driving conditions of the electro-optic modulator.
[0006]
Next, the requirements imposed on the optical millimeter-wave / microwave signal generation method in the downlink of the optical millimeter-wave / microwave access system will be described in more detail.
Generally, a network is made up of "points" and "lines". "Points" in the optical millimeter-wave / microwave system are a central station, a base station, and a terminal, and "lines" are an optical fiber connecting the central station and the base station and a radio link connecting the base station and the terminal.
FIG. 32 is a conceptual diagram showing a spectrum of light having a detuning frequency corresponding to a millimeter wave / microwave band frequency. An optical signal having a detuning frequency of about 60 GHz is generated in an optical frequency band of about 200 THz (1.5 μm band), and this is fiber-transmitted from the central station to the base station. Here, the number of beat signals to be transmitted is important. When a plurality of millimeter-wave / microwave-band beat signals are propagated through an optical fiber, the plurality of millimeter-wave / microwave-band signals are superimposed after photoelectric conversion and interfere with each other. Due to the phase difference between the plurality of signals, the intensity fluctuation of the millimeter wave / microwave band signal is induced. Since this phase difference is generated by the wavelength dispersion of the single mode fiber, it has a periodic intensity characteristic with respect to the propagation length. In addition, since the phase difference between signals at a certain propagation distance differs depending on the frequency, the signal also has a periodic intensity characteristic with respect to the frequency. These effects of dispersion in an optical fiber are called dispersion penalty. In the optical millimeter-wave / microwave access system having the dispersion penalty, the strength of the millimeter-wave / microwave signal supplied to the base station depends on the distance from the central station and the frequency. Particularly, in the high-frequency band, the above-described period becomes short, which causes a serious problem.
[0007]
As an example of an optical modulation method in which a dispersion penalty occurs, there is intensity modulation by continuous wave (CW light) millimeter wave / microwave band electric signals. Since intensity modulation by an analog signal generally generates three spectral components, a conceptual diagram of an optical spectrum obtained by this modulation is as shown in FIG. Since the number of beats of the millimeter wave / microwave band frequency existing between these three spectra is 2, if the phase difference between these two beat signals changes to π due to optical fiber dispersion, these frequency components after heterodyne light reception. Is completely negated.
Therefore, conventionally, a modulation method that generates only two spectral components, that is, a modulation method called a DSB-SC method (see Non-Patent Document 3) has been proposed. As shown in FIG. 32A, the number of beat components in the millimeter-wave / microwave band in the two-mode optical spectrum is one. Therefore, the millimeter-wave / microwave signal obtained by the heterodyne light receiving is not affected by the optical fiber dispersion.
The present invention utilizes the two-mode optical millimeter-wave / microwave signal, and relates to a method of generating an optical millimeter-wave / microwave signal having various characteristics while having little influence of dispersion.
[0008]
Considering the spatial propagation characteristics of the millimeter wave, PSK modulation or QAM modulation is appropriate as a signal modulation method in wireless propagation. Many of the current wireless communication systems also use these PSK signals, QAM signals, or similar modulation formats. As an example of a similar modulation format, there is a modulation method called π / 4 shift QPSK used in a second generation Japanese mobile phone system. In Europe, a modulation method called GMSK is used. A modulation method for carrying information on the amplitude and phase of these electric fields is generally called a primary modulation method. On the other hand, there is a secondary modulation method as a method for further modulating a signal generated by the primary modulation method. Examples include spread spectrum CDMA, frequency hopping CDMA, OFDM, subcarrier multiplexing, and the like. These are collectively referred to herein as an arbitrary modulation format or an arbitrary modulation method.
The present invention relates to a two-mode optical millimeter-wave signal generation method suitable for PSK or QAM modulation or an arbitrary modulation scheme similar to a signal format in wireless propagation.
[0009]
In the present invention, a modulator using an electro-optic effect for light modulation is used.
Next, generation of optical millimeter-wave / microwave signals using the electro-optic effect modulator will be described.
The primary electro-optic effect is an effect of applying an electric field having a lower frequency than light in a crystal lacking anti-image symmetry to linearly change the difference between two refractive indexes felt by a light wave. . The magnitude and direction of the effect of the electric field on the refractive index change depends on the crystal. A commonly used electro-optic modulator is LiNbO ₃ (LN) modulator. Among the LN modulators, there are two types, x-cut and z-cut, depending on the direction of the refractive index change in the crystal used. Further, due to the difference in the structure of the optical waveguide, a Mach-Zehnder interferometer type optical modulator (MZM) in which two phase modulators are integrated, and an optical single sideband (MZM) in which four phase modulators are integrated SSB) modulator is present. Further, there are one-electrode, two-electrode and four-electrode modulators depending on the number of electrodes for RF input signals. These modulators modulate the phase of the light wave and are therefore called optical phase modulators.
[0010]
Next, an output light spectrum modulated by the optical phase modulator will be described.
Each of the modulators described above is configured by integrating an optical phase modulator formed on an electro-optic crystal. The optical phase modulator includes an optical waveguide and electrodes for applying an electric field to the optical waveguide. The optical phase modulator modulates the refractive index by the electric field and modulates the phase of light passing through the optical waveguide. Here, in order to quantitatively describe the DC bias (DC voltage bias) dependency of the optical phase modulator output optical spectrum, a quantitative description of the optical phase modulator output optical spectrum, which is a premise thereof, is given.
The optical electric field input to the optical phase modulator is E _in = E ₀ exp (jΩt), input RF voltage is A _{R F} cos (ωt + θ _RF ), The optical phase modulator output optical electric field E _out Is the first-order n-order Bessel function J for the modulation factor x (= α) _n It is expressed by the following equation using (x).
(Equation 1)

The modulation degree α in the equation is the optical phase modulator half-wavelength voltage Vπ and the amplitude A of the RF voltage. _RF And the following relationship.
(Equation 2)

Also, the Bessel function J _n (X) is shown in FIG.
The half-wavelength voltage Vπ is a voltage at which light transmitted through the optical phase modulator undergoes phase modulation π / 2. Vπ is a device parameter indicating the modulation efficiency, and a low voltage is desired. Currently 5V _pp However, recent studies have reduced this value. Here, an optical phase modulator that applies a constant phase with a constant DC (direct current) bias is called a fixed phase shifter. In some cases, the same optical phase modulator performs optical phase modulation by an RF signal and addition of a fixed phase, but these are not distinguished here. In the case where the same optical phase modulator is used, a bias T that combines a DC signal and an RF signal in an electric circuit is used.
[0011]
Next, the output light of the one-electrode MZM and the output light of the optical SSB modulator constituted by integrating the optical phase modulator will be described. The MZM is composed of two optical phase modulators and one fixed phase shifter, and a one-electrode MZM configuration for performing the DSB-SC method is shown in FIG. In the one-electrode MZM modulator, the phase difference between the optical phase modulators and the phase difference of the input RF signal are important parameters. The former can be controlled by the DC component of the electric field injected from the electrode, and the latter can be controlled by changing the electric circuit configuration. The normal one-electrode MZM inputs RF signals having phases opposite to each other to two internal optical phase modulators, that is, the latter parameter is fixed to π.
[0012]
The DC bias dependency of the output light of one electrode MZM will be described.
Here, the phase difference given by the DC bias is θ _DC The output electric field E of one electrode MZM when _out Represents Here, for simplicity, chirp and insertion loss are not included.
[Equation 3]

Where θ _DC Is π, the above expression is expanded as follows.
(Equation 4)

From this, it can be seen that the output light of one electrode MZM is composed of the odd-order sideband, and the carrier and the even-order sideband are suppressed. This modulation method is called carrier-suppressed double-sideband modulation (DSB-SC) (see Non-Patent Document 3) and is frequently used when generating millimeter-wave / microwave signals. Its features will be described later. When the phase term controlled by the DC bias is 0, the expansion is performed as follows.
(Equation 5)

Under this bias, even-order sidebands are output. When an RF intensity at which α is about 2.4 is input, an optical carrier component having a quadrupling function is suppressed, and a two-mode optical millimeter-wave / microwave signal is generated.
As described above, in an optical modulator such as an MZM including an optical phase modulator, a phase difference between modulated lights controlled by a DC bias is an important parameter in controlling an optical spectrum.
[0013]
Although the optical spectrum can be quantitatively grasped as described above, a spectrum schematic diagram capable of comprehensively expressing the phase and the frequency will be described for easier understanding. As a premise, the optical electric field output from the optical phase modulator represented by Expression 1 is developed while paying attention to the frequency and the phase.
(Equation 6)

Here, the optical electric field corresponding to the input RF signal phase can be represented in FIG. In the figure, the oblique axis indicates the frequency axis, and the axis rotation angle indicates the phase. Further, the phase of the optical carrier is defined as a reference phase in FIG. In the figure, the arrow length indicates the electric field strength, which corresponds to the Bessel function in each order.
[0014]
34, 35, and 36 correspond to the phase change of the RF signal, that is, the time change. (B) to (j) in each drawing are as follows. (B) is a phasor that represents an input RF signal. An actual electric field corresponding to an I (In-phase) axis component in the figure is applied to the modulator. (C) is a phasor expressing light that has been phase-modulated from this RF electric field. (D) to (f) decompose the electric field represented by the phasor. (D) is a stationary component (carrier component), (e) is a Q (Quadrature) axial direction fluctuation component, and (f) is an I-axis direction fluctuation component. This corresponds to
(Equation 7)

Corresponding to Here, third-order and higher-order components are excluded. The relationship between (e) and (g) and (h) is
(Equation 8)

And the relationship between (f) and (i) and (j) is
(Equation 9)

Corresponding to That is, the phase modulation shown in (c) is decomposed into frequency components having phases shown in the phasors (d), (g), (h), (i), and (j). (A) is a diagram in which the phases of these frequency components are arranged on the same axis. Therefore, hereinafter, the phase of each spectral component is expressed in the expression format (a).
[0015]
The principle of DSB-SC modulation will be described in detail using this expression format.
FIG. 37 shows an MZM configuration for performing the DSB-SC method. The one-electrode MZM modulator supplies electric fields of opposite phases to two internal phase modulators. The DC bias gives a phase change to only one propagating light. FIG. 38 shows a schematic optical spectrum diagram corresponding to (a) to (c) in the figure. In FIG. 38, the phase of the transmitted light carrier at the position (a) in FIG. 37 and the phase of the input RF electric field are used as references. FIG. 38C shows that the modulator output light has an odd-order component, which corresponds to Expression 5.
Note that FIG. 38 (c) shows that the phase of the beat between the upper and lower primary modulation components and the phase of the beat between the primary and tertiary modulation components are different from each other by π. This shows that the output RF intensity is minimized in back-to-back transmission that is not affected by. In practical terms, the input RF intensity is reduced to reduce the effect of the higher-order sideband, and J in FIG. ₃ It is necessary to drive in a range where the component is small. Thus, three millimeter-wave / microwave-band beat components in FIG. 38C are approximately one.
Thus, since the DSB-SC method is a two-mode optical spectrum transmission, the number of effective beat components in the millimeter-wave / microwave band is one. Therefore, a millimeter-wave / microwave signal obtained by heterodyne light reception is hardly affected by optical fiber dispersion.
[0016]
The features of optical millimeter-wave / microwave signal generation using the DSB-SC method are mainly summarized as follows.
(A) Frequency multiplication function:
As can be seen from FIG. 38 (c), since the beat of the primary upper and lower sideband components is used as the output signal, the optical phase modulator may be modulated at half the required frequency of the output signal. In addition, the RF signal generation circuit is simplified.
(B) High dispersion resistance:
As can be seen from FIG. 38 (c), by reducing the intensity of the input RF signal and reducing the third-order modulation component, only the primary upper and lower sideband components are transmitted through the optical fiber. The type is limited, and therefore no optical fiber dispersion penalty occurs.
Optical millimeter-wave / microwave signal generation using the DSB-SC method is called a two-mode optical millimeter-wave / microwave signal generation method because the primary upper and lower sideband components are transmitted through an optical fiber.
Since the two-mode optical millimeter-wave / microwave signal generation method has the above-described excellent characteristics, a PSK (Phase Shift Keying) modulation method using the two-mode optical millimeter-wave / microwave signal generation method, that is, Mode optical millimeter-wave / microwave PSK signal generation methods have been studied (see Non-Patent Documents 4 and 5).
[0017]
Next, problems of the conventional two-mode optical millimeter-wave / microwave signal generation method will be described.
The principle of conventional two-mode optical millimeter-wave / microwave PSK signal generation will be described.
The electric field generated by the DSB-SC operation is expressed by the following equation from Equation 5, excluding higher-order modulation components. Note that this equation is correct when the input RF signal strength is low.
(Equation 10)

The intended millimeter-wave / microwave signal component of the photocurrent obtained by heterodyne receiving this light with a photodiode (PD) is:
[Equation 11]

It becomes. This produces a double wave of the input signal, but also doubles the phase shift at the same time. For this reason, for example, when a PSK signal is used as an input signal, a PSK signal cannot be obtained after PD light reception. Therefore, a signal having a phase shift of 1 / frequency multiple of the target PSK signal is generated. This is realized by the circuit shown in FIG. The schematic diagram of the optical spectrum is shown in FIG.
[0018]
However, this conventional method has the following problems.
(A) When two independent BPSK signals are superimposed as input signals, that is, when a QPSK signal is generated, an interference component of the two PSK signals is generated, and a QPSK signal is generated by adding orthogonal BPSK signals. Is impossible.
(B) The spectrum is expanded due to the transient transition process between the symbol 1 and the symbol 0 of the BPSK signal, so that the frequency use efficiency is reduced and the code error rate is increased.
(C) Since the modulation format of the signal generated at the modulator input terminal is different from the modulation format of the signal obtained after heterodyne reception, it is impossible to use an arbitrary modulation format.
[0019]
[Non-patent document 1]
U. Gliese, S.M. Norskov and T.W. N. Nielsen, "Chromatic Dispersion in Fiber-Optical Microwave and Millimeter-Wave Links", IEEE Trans. Microwave Theory Tech. , Vol. 44, pp. 1716-1724, 1996.
[Non-patent document 2]
U. Gliese, T .; N. Nielsen, S .; Norskov and K.K. E. FIG. Stubkjar, "Multifunctional fiber-optic microwave links based on remote heterodyne detection", IEEE Trans. Microwave Theory Tech. , Vol. 46, pp. 458-468, 1998.
[Non-Patent Document 3]
J. J. O'Reilly, P .; M. Lane, R.A. Heidemann and R.S. Hofsetter, "Optical Generation of Very Narrow Linewidth Millimeter Wave Signals", Electron. Lett. , Vol. 28 pp. 2309-2311, 1992.
[Non-patent document 4]
Y. Ozeki, M .; Kishi and M.S. Tsuchiya, "Dual-Mode Optical MM-Wave PSK Signal Generation Technique for 60 GHz-Band Fiber-Radio Systems", Technical Digest of the 2nd Korea-Japan Joint Workshop on Microwave-Photonics, IV-3, pp. 125-128, Seoul, February 2001.
[Non-Patent Document 5]
Yasuyuki Koseki, Masato Kishi, Masahiro Tsuchiya "Generating 60-GHz 2-mode optical millimeter-wave and microwave PSK signals using a Mach-Zehnder optical modulator" 2001 IEICE General Conference, Ritsumeikan University.
[Non-Patent Document 6]
M. Izutsu, S .; Shikama, and T.S. Sueta, “Integrated Optical SSB Modulator / Frequency Shifter”, IEEE J. Quantum Electron. , Vol. QE-17, pp. 2225-2227, 1981.
[Non-Patent Document 7]
S. Shimototsu, S .; Oikawa, T .; Saitou, N .; Mitsugi, K .; Kubodera, T .; Kawanishi and M.S. Izutsu, "Single Side-Band Modulation Performance of a LiNbO ₃ Integrated Modulator Consisting of Four-Phase Modulator Wavewaves ", IEEE Photonics Tech. Lett., Vol. 13, pp. 364-366, 2001.
[Non-Patent Document 8]
A. Narasimha, X .; J. Meng, M .; C. Wu and E. Yablonovitch, "Tandem single sideband modulation scheme for double spectrum spectral efficiency of analog fiber links, Electronics Letters. 36, pp. 1135-1136, 2000.
[0020]
In view of the above problems, it is a first object of the present invention to provide a two-mode optical millimeter-wave / microwave QPSK signal generation method that does not generate signal interference components.
It is a second object of the present invention to provide a two-mode optical millimeter-wave / microwave BPSK signal generation method that does not cause spectrum expansion.
It is a third object of the present invention to provide an optical millimeter-wave / microwave signal generation method capable of using an arbitrary modulation format.
[0021]
[Means for Solving the Problems]
To achieve the above object, an optical millimeter-wave / microwave signal generation method according to the present invention divides a single-wavelength light into a first light and a second light, and divides the first light and the second light. Are input to the first optical modulator and the second optical modulator, respectively, and the first RF signal and the second RF signal are input to the first optical modulator and the second optical modulator, respectively. The first light and the second light are respectively subjected to DSB (double side band) modulation, and a fixed phase π / 2 is added between the first light and the second light subjected to DSB modulation and added. The interference component of the first RF signal and the second RF signal is suppressed in the added heterodyne received light output of the first light and the second light.
According to this method, a plurality of millimeter-wave / microwave-band signals can be transmitted without interference.
[0022]
Further, the single-wavelength light splitting, adding means, light modulating means and fixed phase π / 2 adding means apply an electric field to the branching optical waveguide and the optical waveguide formed integrally on the electro-optic crystal. An optical SSB modulator composed of a plurality of optical phase modulators that add a phase to light propagating through an optical waveguide and a π / 2 fixed phase shifter that applies a constant electric field to the optical waveguide may be used.
Alternatively, the single-wavelength light splitting, adding means, light modulating means, and fixed phase π / 2 adding means apply an electric field to the branching optical waveguide and the optical waveguide integrated on the electro-optic crystal. A two-electrode MZM optical modulator light composed of a plurality of optical phase modulators for adding a phase to light propagating through the optical waveguide, and a π / 2 fixed phase shifter for applying a constant electric field to the optical waveguide; It is only necessary to have an optical band rejection filter for removing the carrier after the addition of the second light and the second light.
According to the above method, the heterodyne received light output is a heterodyne output between the primary modulation components of the upper and lower sidebands of the first RF signal and the second RF signal, respectively, and the first RF signal and the second RF signal are output. Has a frequency that is twice as high as the frequency of each RF signal, and has a phase that is twice as high as the input phase of each of the first RF signal and the second RF signal.
Further, the heterodyne received light output may be a heterodyne output between the secondary modulation components of the upper and lower sidebands of the first RF signal and the second RF signal, respectively, and the first RF signal and the second RF signal may be used. The first RF signal and the second RF signal may have four times the frequency of each signal, and may have four times the input phase of each of the first RF signal and the second RF signal.
[0023]
Further, in the method of obtaining the above twice frequency and phase, the first RF signal and the second RF signal are separated from each other by an NA-BPSK (narrow angle binary phase) having a phase shift amount of π / 2. (shift keyig) signal, and by delaying the first RF signal and the second RF signal so as to have a phase difference of π / 4 from each other, a π / 2 signal between two BPSK signals generated in heterodyne light reception is obtained. A phase difference occurs, and the two BPSK signals are added to obtain a QPSK (quadrature phase shift keying) signal.
Further, in the method for obtaining the frequency and phase four times as described above, the first RF signal and the second RF signal are mutually independent NA-BPSK signals having a phase shift of π / 4, and By delaying the first RF signal and the second RF signal so as to have a phase difference of π / 8 from each other, a phase difference of π / 2 occurs between two BPSK signals generated in heterodyne reception, and the two BPSK signals are generated. The BPSK signal is added to obtain a QPSK signal.
[0024]
Further, in the method for obtaining the above-mentioned double frequency and phase, the first RF signal and the second RF signal are converted to NA-s with a phase shift of π / 2 and opposite phase changes. By making them BPSK signals and delaying the first RF signal and the second RF signal so as to have a phase difference of π / 2 from each other, two BPSK signals generated in heterodyne light reception are generated between the symbols of the signals. In the above transient transition state, unnecessary components cancel each other out to prevent spectrum broadening.
Further, in the method of obtaining the frequency and the phase four times as described above, the first RF signal and the second RF signal may have a phase shift amount of π / 4 and a phase change amount opposite to each other. The two BPSK signals generated in the heterodyne light reception are converted into the BPSK signals by delaying the first RF signal and the second RF signal so as to have a phase difference of π / 4 from each other, so that It is characterized in that unnecessary components cancel each other out in a transient transition state between them, thereby preventing spectrum spread.
[0025]
Further, by making the frequencies of the first RF signal and the second RF signal different from each other, the frequencies of the two signals generated in the heterodyne light receiving are different, and the interference component between the two signals is suppressed. It is characterized in that subcarrier multiplexing is possible.
[0026]
The addition of the first RF signal to the third RF signal that has not been modulated and the sixth RF signal obtained by giving a phase change of 90 ° to the fifth RF signal modulated by an arbitrary modulation method. And a second RF signal is generated by adding a fourth RF signal and a fifth RF signal obtained by giving a phase change of 90 ° to the third RF signal. By inputting the signals to the respective optical modulators, in heterodyne light reception, the signal has the frequency of the sum of the frequency of the third RF signal and the frequency of the fifth RF signal, and A signal modulated by the given modulation method is obtained.
[0027]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that substantially the same members will be described with the same reference numerals.
First, a first embodiment of the present invention will be described.
FIG. 1 is a diagram showing a configuration of an optical millimeter-wave / microwave signal generation method of the present invention when an OSSBM (optical SSB modulator) is used as an optical modulator. Light from a single-wavelength light source 1 such as a single-wavelength laser light source is input to a branching optical waveguide 3 formed on an electro-optic crystal 2. One of the lights branched by the branch optical waveguide 3 is input to the optical modulator 4 via one branch 3 a of the branch optical waveguide 3. The other of the branched lights is input to the optical modulator 5 via the other branch 3 b of the branch optical waveguide 3 and via the π / 2 fixed phase shifter 6. The light subjected to the optical phase modulation by the optical modulator 4 and the light subjected to the optical phase modulation by the optical modulator 5 are added by the branch optical waveguide 7 via one branch 7a and the other branch 7b of the branch optical waveguide 7. , And is input to a photodetector 9 such as a photodiode provided at an output end of the optical fiber 8.
The optical modulators 4 and 5 are MZMs, and a configuration using two MZMs is called an OSSBM (optical SSB modulator). The optical modulator 4 has an optical phase modulator 11a at a branch 10a of the branch optical waveguide 10, and has a π fixed phase shifter 12 and an optical phase modulator 11b at a branch 10b of the branch optical waveguide 10 in series. I have. The optical modulator 5 has the same configuration as the optical modulator 4.
[0028]
In order to operate with this configuration, the RF input signal 13 is input to the optical phase modulators 11a and 11b of the optical modulator 4 in opposite phases by using the RF phase shifter 14. Similarly, the RF input signal 15 is input to the optical phase modulators 11a and 11b of the optical modulator 5 in opposite phases using the π phase shifter 14 for RF.
[0029]
FIG. 2 is a diagram showing a configuration of an optical millimeter-wave / microwave signal generation method of the present invention when a DE-MZM (two-electrode Mach-Zehnder optical modulator) is used as an optical modulator. This configuration uses the optical phase modulators 11a and 11b for the modulators 4 and 5 in FIG. 1, and is called DE-MZM. One branch 3a of the optical branch waveguide 3 has an optical phase modulator 11a, and the other branch 3b has an optical phase modulator 11a and a π / 2 fixed phase shifter 6. After the lights modulated by the optical modulators 4 and 5 are added by the optical branching waveguide 6, an optical band stop filter 16 for removing a carrier (light of the light source 1) is provided. The optical band stop filter 16 is, for example, a fiber Bragg grating (FBG). Other configurations are the same as those in FIG.
To operate in this configuration, RF input signals 13 and 15 are input to optical modulators 4 and 5, respectively.
[0030]
According to the configuration of FIG. 1 or FIG. 2, the operation is as follows.
The optical carrier frequency of the light output from the light source 1 is Ω, and the frequencies of the RF signals 13 and 15 are ω ₁ And the electric field strength and the phase of the RF signal 13 and the RF signal 15 are E ₁ , E ₂ And φ ₁ , Φ ₂ Then, the upper and lower primary sidebands output from the optical modulators 4 and 5 are respectively
E ₁ expj ｛(Ω−ω ₁ ) T-φ ₁ ｝,
E ₂ expj ｛(Ω + ω ₂ ) T + φ ₂ ｝
It is expressed as A phase of π / 2 is added to the one primary sideband by the π / 2 fixed phase shifter 6 and added, and the added upper and lower primary sidebands are heterodyne-received by the photodiode 8. Photocurrent i _p Is given by
(Equation 12)

As is clear from this equation, the RF signal 13 and the RF signal 15 can be extracted without interference.
[0031]
Next, a first mode of the present invention, a two-mode optical millimeter-wave / microwave QPSK signal generation method in which no signal interference component occurs, will be described.
The QPSK signal used in the two-mode optical millimeter-wave / microwave QPSK signal generation method of the present invention will be described. There are two types of QPSK signals called quadrature phase shift keying and quadrature phase shift keying. The former is realized by the addition of orthogonal BPSK. FIG. 3A shows a configuration of a circuit for generating the quadrature phase shift keying signal, and FIG. 3B shows an I / Q diagram of the phasor. On the other hand, the latter is generated using a phase modulator. An example of the configuration is shown in FIG. 4A, and a phasor is shown in FIG. The difference between the two is in the transition process between symbols. Although details will be described later, a phase fluctuation component in a transition process between symbols induces spectrum spreading. Four-phase phase shift keying has a particularly large spectrum spread. For this reason, the present invention uses quadrature phase shift keying.
[0032]
FIG. 5 is a diagram for explaining the principle and configuration of the two-mode optical millimeter-wave / microwave QPSK signal generation method of the present invention.
As described in FIG. 1 or FIG. 2, the integrated optical modulator uses OSSBM (optical SSB modulator; see Non-Patent Documents 6 and 7) or DE-MZM (two-electrode MZM). However, when the DE-MZM is used, since a DSB-SC operation cannot be obtained, an optical band rejection filter (BEF: band elimination filter) such as a fiber Bragg grating is used to suppress a carrier wave.
The input RF signals are respectively ω using the frequency / phase identifier i. _i , Φ _i , And in this figure, ω ₁ = Ω ₂ , Φ ₁ = (2m-1) π / 4, φ ₂ = Mπ / 2, and m and n are digital signals having a value of 0 or 1. A 45 ° phase difference is provided between the two BPSK signals at the input end. The fixed phase difference provided between the two modulators is φ _bias And
[0033]
According to this configuration, the phase difference between the two BPSK signals is doubled due to the nonlinearity of the modulator, and an orthogonal BPSK signal is obtained at the output end. For the optical carrier frequency Ω, the upper and lower primary sidebands are
E _i expj ｛(Ω−ω _i ) T-φ _i ｝,
E _i expj ｛(Ω + ω _i ) T + φ _i ｝
It is expressed as
The correspondence between the input RF signal and the optical spectrum is as follows.
(Equation 13)

Photocurrent i obtained by heterodyne receiving this optical spectrum _p Is given by
[Equation 14]

In this equation, the phase φ including the data signal ₁ , Φ ₂ Are generated, and correspond to the interference components of the two PSK signals described in the conventional method. The possibility that two BPSK signals generated by the input NA-BPSK (Narrow Arrow Binary Phase Shift Keying) signal may cause interference at the output terminal is shown again using a spectrum schematic diagram.
[0034]
FIG. ₁ And φ ₂ FIG. 2 is a schematic diagram of an optical spectrum when two RF signals having information, RF1 and RF2, are input to each modulator.
RF1 and RF2 generate an upper sideband and a lower sideband, respectively, which will be referred to as USB1, LSB1 and USB2, LSB2. The RF signals generated by the heterodyne reception of LSB1 and USB1 and the heterodyne reception of LSB2 and USB2 are 2φ. ₁ And 2φ ₂ Has the phase information of However, here, due to the heterodyne reception of USB1 and LSB2 and of USB2 and LSB1, 2φ ₁ + 2φ ₂ Is output at the same time.
[0035]
The method of the present invention _bias Is set to π / 2, and this interference component is removed. That is, the two interference components are canceled out of phase with each other. That is, the phase difference φ between the two waveguides controlled by the DC bias _bias To π / 2, one of the interference components has a leading phase of π / 2 and the other has a lagging phase of π / 2. As a result, both components have opposite phases to each other. The phase relationship of the light spectrum inside the modulator at this time is as shown in FIG. In Equation 14, φ _bias = Π / 2, the photocurrent i obtained by heterodyne light reception _p Is given by Expression 12. Thus, the input signal to the optical modulator is added without interference at the output end, and two BPSK signals can be extracted without interference.
Note that ω ₁ = Ω ₂ But ω ₁ ≠ ω ₂ In this case, the photocurrent i obtained by heterodyne light reception may be used. _p Is given by
(Equation 15)

That is, even if the BPSK signal is sub-carrier multiplexed (SCM), no interference component occurs. In general, when an MZM modulator is used, generation of unnecessary components is unavoidable in SCM due to nonlinearity of the modulator. However, this method can theoretically suppress generation of unnecessary components.
[0036]
Next, an example of the first embodiment will be described.
Each of the methods using OSSBM and DE-MZM as the optical phase modulator at the RF frequency of the 60 GHz band was performed.
FIG. 8 is a diagram showing an experimental system using OSSBM. OSSBM is LiNbO ₃ An x-cut 2 electrode modulator. A 1 GHz band NA-BPSK signal generated based on a 156 Mbps pseudo random bit sequence (PRBS) signal output from a pulse pattern generator (PPG) is up-converted into a 30 GHz band, and a transmission bandwidth of 600 MHz BPPF is used. Image components were removed. The generation of the 2NA-BPSK signal was realized by splitting this signal into two and adding a one-bit length time delay to one of them. A delay corresponding to π / 4 of a 30 GHz component is added between the two signals. A tunable laser was used as a light source, and this output light was input to the OSSBM via a polarization controller (PC). The output light modulated according to the present method was PD-received through an EDFA (Erbium Doped Fiber Amp) and a BPF. Here, a traveling-wave type photodiode was used as a light receiving element, and a millimeter wave / microwave band signal after light reception was amplified. This was down-converted to a 2 GHz band, and an I / Q diagram was obtained by quadrature demodulation.
[0037]
The measurement results are shown below. FIG. 9 shows the BPF output light spectrum.
(A) shows two modes modulated by an NA-BPSK signal, while (b) shows two modes modulated by an unmodulated RF signal. In any case, there is an intensity difference between the upper and lower sidebands. The intensity ratio of the former is 5.9 dB, and that of the latter is 7.2 dB. This difference in intensity is determined by φ set by the DC bias. _bias = 90 ° and a 45 ° phase difference provided between two input RF signals. That is, the former sets the phase difference between the two signals to 90 ° with respect to both sidebands, while the latter sets the phase difference between the two signals to −45 ° with respect to the upper sideband and the phase difference between the two signals to the lower sideband. To + 45 °. Therefore, constructive interference occurs in the upper sideband (short wavelength side), and weaker interference occurs in the lower sideband (long wavelength side). There is 7.2 dB between the two optical modes obtained by the single tone modulation, which is the theoretical value,
(Equation 16)

And almost match.
FIG. 10 shows a spectrum of an RF signal obtained by heterodyne receiving this two-mode light spectrum with a PD. Although signal generation in the 60 GHz band was confirmed, periodic distortion was observed on the spectrum. This is considered because there is a steady propagation time difference between the two waveguides in the OSSBM. This time difference is due to the creation of two NA-BPSK signals from one NA-BPSK signal generated from a single PPG in order to simplify the experimental system. That is, since a time difference of about 1 bit is added to the electric circuit, it is considered that the Mach-Zehnder filter has a reciprocal period of about 150 MHz in the optical circuit. This does not occur if the NA-BPSK signal has different digital data.
FIG. 11B shows a phasor obtained by down-converting the 60 GHz band QPSK signal to a 2 GHz band. FIG. 3A shows an input NA-BPSK signal to the OSSBM. It is clear from these figures that the generation of the QPSK signal was confirmed, demonstrating the present invention.
[0038]
Next, an embodiment using DE-MZM as an optical phase modulator will be described.
The method of the present invention is to obtain beats in the upper and lower sidebands generated by optical phase modulation, and it is possible to obtain a QPSK signal without using OSSBM. The purpose of using the OSSBM is to provide a carrier suppression operation to the MZM inside the modulator. Therefore, modulation by DE-MZM and carrier removal by an optical band rejection filter are also possible. As a result, it is possible to use a commercial DE-MZM without using the OSSBM which can be said to be a special modulator, so that the convenience is high.
[0039]
The experimental system is the same as FIG. 8 except for the configuration of the phase modulator. The phase modulator configuration has a difference between (a) and (b) in FIG.
FIG. 13 shows the optical spectrum of the two-mode optical millimeter-wave / microwave signal generated in this embodiment. A major difference from the optical spectrum generated by OSSBM is that a secondary modulation component is generated. This is because the OSSBM can suppress even-order sidebands including a carrier wave, but this method cannot. However, this secondary modulation component and the carrier which is the component that makes the desired millimeter-wave / microwave frequency detuning are suppressed. Therefore, there is no effect on the dispersion penalty. FIG. 14 shows a phasor and BER obtained by receiving the two-mode optical millimeter-wave / microwave signal and performing down-conversion. As is clear from the figure, the error-free operation was realized at the received light intensity of -17 dBm, and the generation of the QPSK signal was confirmed. Thus, the present invention has been demonstrated.
[0040]
Next, a second embodiment of the present invention will be described.
This embodiment relates to a two-mode optical millimeter-wave / microwave BSK signal generation method for suppressing spectrum expansion accompanying a transient transition process between symbol 1 and symbol 0 of a BPSK signal.
First, a description will be given of factors that cause spectrum expansion in a BPSK signal generated by MZM input of an NA-BPSK signal.
FIG. 15 shows the phasor of the NA-BPSK signal and the BPSK signal generated by this doubling. Here, paying attention to the transient transition process between symbol 1 and symbol 0, the generated BPSK signal has a phase fluctuation component. This component is divided into an I-axis parallel component and a Q-axis parallel component as shown in FIG. If the latter varies with the bit rate, the former varies with twice the bit rate. This causes spectral broadening.
[0041]
FIG. 17A is a calculation result of a spectrum after passing a baseband NA-BPSK signal modulated at 100 Mbps through an LPF (α = 0.2) of 50 MHz. FIG. 3B shows a BPSK signal obtained by doubling the BPSK signal. A bandwidth of about 100 MHz has been expanded to a maximum of about 200 MHz, and a carrier wave has appeared. No carrier exists in an ideal BPSK signal.
[0042]
Here, FIG. 18 shows the spectrums of the aforementioned I-axis parallel component and Q-axis parallel component. From this figure, it is clear that the I-axis parallel component is a factor of spectrum expansion.
[0043]
Next, a method of suppressing spectrum expansion due to the I-axis parallel component will be described.
As shown in FIG. 19, by adding a Q-axis symmetric BPSK signal to one BPSK signal after receiving the PD, the I-axis fluctuation component is suppressed. The phase modulator used here is OSSBM or DE-MZM. In the latter, carrier removal by FBG or the like is required to increase dispersion tolerance.
As in the first embodiment, two NA-BPSK signals are input to OSSBM or DE-MZM. However, here, unlike Embodiment 1, a signal having the same data and a signal having a phase change opposite to each other is generated. The phase difference between the two RF signals is 90 °, and a 180 ° phase difference is obtained after heterodyne reception by the PD. As a result, a BPSK signal symmetric with respect to the Q axis is obtained.
Further, as described in the first embodiment, DC bias φ for suppressing interference components is used. _bias = Π / 2. However, unlike the first embodiment, the interference component here is not the interference component of two signals having different data, but the interference between signals having the same data, and this is what causes the output PSK signal to have a carrier. is there.
According to this method, the I-axis parallel component in the transient transition process between symbol 1 and symbol 0 is canceled by combining the I-axis symmetric BPSK signal and the Q-axis symmetric BPSK signal, and spectrum expansion is suppressed.
[0044]
Next, examples will be described.
To confirm the above principle, a 1.5 GHz band BPSK signal generation experiment was performed using OSSBM. FIG. 20 shows the experimental system. The 200 Mbps PRBS signal output from the PPG is converted into a 750 MHz band BPSK signal via a 100 MHz LPF. In order to remove side lobe components generated here, a BPF having a transmission bandwidth of 300 MHz in a 750 MHz band is used. After amplifying the BPSK signal and inputting it to the 90 ° hybrid circuit together with the original 750 MHz single tone signal, two target NA-BPSK signals are obtained.
After performing this operation on DFB-LD output light, PD light reception was performed via EDFA and BPF. Quadrature demodulation was performed on the obtained 1.5 GHz band BPSK signal to obtain an I / Q diagram. FIG. 21 shows the obtained RF spectrum, and FIG. 22 shows the I / Q diagram. FIG. 21 shows that the occupied spectrum band according to the method of the present invention shown in (b) is narrowed and the carrier component is suppressed as compared with the band according to the conventional method shown in (a).
[0045]
FIGS. 22A and 22B are BPSK signals according to the conventional method. Here, the same phasor was obtained by injecting and not injecting an input RF signal into the OSSBM. FIG. 7C shows a BPSK signal obtained by adding the BPSK signals corresponding to FIGS. The trajectory between the symbols slightly deviates from the straight line, which is considered to be caused by the difference in the intensity between the 2 BPSK signals. FIG. 22C demonstrates the principle of the present invention.
[0046]
Next, a third embodiment of the present invention will be described.
This embodiment relates to an optical millimeter wave signal generation method capable of obtaining a signal having the same modulation scheme as that at the modulator input end after heterodyne reception, regardless of the modulation format of the signal.
FIG. 23 is a diagram showing the configuration of the optical millimeter-wave / microwave signal generation method of the present invention. The oscillator output signal is split into two, and one of them is modulated in an arbitrary format. This and the single tone signal are input to the 90 ° hybrid circuit. The 90 ° hybrid circuit is a device having two input terminals (input terminal 21 and input terminal 22) and two output terminals (output terminal 23 and output terminal 24) as shown in FIG. The signal input from the input terminal 21 is split into two, and one of the split signals is output to the output terminal 24 after a 90 ° phase change is applied to the split signal. The other branched signal is output to the output terminal 23 without any phase change. Similarly, the signal input from the input terminal 22 is split into two, and one of them is output to the output terminal 23 after a phase change of 90 ° is applied, and the other is output to the output terminal 24 without the phase change. Is output to In the second embodiment, the NA-BPSK signal is obtained by adjusting the intensity and phase between the two input signals. However, in this method, a strict adjustment is not required because a 90 ° hybrid circuit is used. When strict amplitude / phase adjustment is required, this function is realized using a divider, a phase shifter, an attenuator, and the like.
[0047]
Note that there is a conventional example (see Non-Patent Document 8) for inputting an RF signal to the two-electrode MZM via a 90 ° hybrid. However, in this conventional example, two RF signals having different information for the same light are used. Therefore, the two signals input to the 90 ° hybrid circuit are both modulated, which is different from the present invention. In the present invention, beat signals of the upper and lower sidebands are obtained instead of obtaining the beat signals of the optical carrier and the modulation sideband by heterodyne light reception.
[0048]
It can be said that the BPSK modulation method of the second embodiment is a modulation method under one condition of this method. Therefore, a process of generating an NA-BPSK signal and a process of turning the optical sideband into a single tone in the second embodiment will be described here.
FIG. 25 and FIG. 26 show an RF signal phasor and a phasor having upper and lower sidebands. (A) to (i) correspond to FIGS. 23 (a) to (i).
In FIG. 26 (i), it can be seen that the upper sideband is single tone and the lower sideband is BPSK modulated light. With the configuration shown in FIG. 23, one of the upper and lower sidebands becomes a single tone, and the other becomes a modulated spectrum.
From this, it can be considered that this modulation method is a modulation method including two SSB modulations. That is, in the normal usage of the SSB modulator, an upper RF band or a lower RF band is generated by inputting an input RF signal to the SSB modulator via a 90 ° hybrid circuit. At this time, whether the generated sideband is the upper sideband or the lower sideband is determined depending on whether the wave is advanced by 90 ° or delayed by 90 °. In this method, a single tone signal and an arbitrary modulation signal are input to an optical SSB modulator via a 90 ° hybrid circuit. There is an opposite relationship between them. Therefore, it is independent of the modulation format.
However, taking into account the RF intensity dependency of the optical modulator DC bias, the RF intensity changes depending on the symbol at the time of QPSK signal generation or the like. For example, when the phasor corresponding to FIG. 25C becomes as shown in FIG. 27A, the RF intensity becomes zero, which may cause a bias shift of the LN modulator. Therefore, the symbols may be shifted from the zero point of the amplitude as shown in FIG. This can be realized by making the RF intensity in FIG. 23A larger than that in FIG.
[0049]
Next, an example of the third embodiment will be described.
An experiment was performed in a 60 GHz band to confirm the above principle. This experimental system is shown in FIG.
A 750 MHz band QPSK signal was generated from two signals of 156 Mbps PRBS, and after amplifying and filtering, it was input to a 90 ° hybrid circuit together with a single tone. The resulting IF band 2 signal was up-converted to a 29.5 GHz band, amplified, and input to the optical SSB modulator. FIG. 29 shows an optical spectrum measured after EDFA amplification and BPF of the DFB-LD output light modulated by this signal. Although some carrier waves remain, there is no secondary sideband for generating beats in the millimeter-wave band and the microwave band, so that the influence on the dispersion penalty can be ignored. Further, since the 28 GHz band image component is not removed at the time of input to the optical modulator, the upper and lower sidebands are considered to be composed of two or more sidebands. However, the wavelength resolution of the optical spectrum analyzer is 0.01 nm. (About 1.25 GHz), which is not recognized on the optical spectrum. The two-mode optical millimeter-wave / microwave signal was received by the UTC-PD. A 59 GHz band BPF was used after PD light reception for the purpose of removing a 56 GHz band image component and the like. I / Q diagrams and IF band spectra obtained after down-conversion to the 1.5 GHz band are shown in FIGS. Here, it was confirmed that the desired QPSK signal was obtained. It should be noted that optimization of the modulation operation, noise elimination and the like is insufficient for the experimental results.
[0050]
【The invention's effect】
As understood from the above description, according to the optical millimeter wave / microwave signal generation method of the present invention, a plurality of signals can be transmitted without interference, for example, a QPSK signal can be transmitted. Further, a BPSK signal can be transmitted without spectrum expansion. Further, any modulation format can be used.
It is extremely useful if the method of the present invention is used for generating optical millimeter-wave / microwave signals in the downlink of an optical millimeter-wave / microwave access system.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of an optical millimeter-wave / microwave signal generation method of the present invention when an OSSBM (optical SSB modulator) is used as an optical modulator.
FIG. 2 is a diagram illustrating a configuration of an optical millimeter-wave / microwave signal generation method according to the present invention when a DE-MZM (two-electrode Mach-Zehnder optical modulator) is used as an optical modulator.
FIG. 3 is a diagram illustrating a configuration of a QPSK signal generation circuit using quadrature phase shift keying and a phasor thereof.
FIG. 4 is a diagram illustrating a configuration of a circuit for generating a four-phase phase shift keying signal and a phasor thereof.
FIG. 5 is a diagram illustrating a configuration of a QPSK signal generation method according to the first embodiment of the present invention.
FIG. 6 is a diagram schematically showing a spectrum in a modulator.
FIG. 7 is a diagram schematically showing a spectrum when a DC bias is set to π / 2.
FIG. 8 is a diagram showing an experimental system when OSSBM is used for an optical modulator.
FIG. 9 is a diagram showing a spectrum of BPF output light obtained in the example.
FIG. 10 is a diagram showing an RF signal spectrum obtained in the example.
FIG. 11 is a diagram showing an input signal phasor and an output signal phasor to the OSSBM.
FIG. 12 is a diagram showing a difference between an experimental system using OSSBM and an experimental system using DE-MZM.
FIG. 13 is a diagram showing an optical spectrum of a two-mode optical millimeter-wave / microwave signal when DE-MZM is used.
FIG. 14 is a diagram illustrating a code error rate and a phasor of a QPSK signal obtained in an example.
FIG. 15 is a diagram illustrating a phasor of an NA-BPSK signal and a BPSK signal generated by doubling.
FIG. 16 is a diagram illustrating a phase variation component in a process of transitioning between symbols of a BPSK signal.
FIG. 17 is a diagram showing a spectrum of a baseband NA-BPSK signal and a spectrum of a BPSK signal obtained by doubling.
FIG. 18 is a diagram showing a Q-axis parallel component and an I-axis parallel component of a spectrum of a BPSK signal obtained by doubling.
FIG. 19 is a diagram illustrating a method of suppressing spectrum expansion of a BPSK signal according to the second embodiment of the present invention.
FIG. 20 is a diagram showing an experimental system used in Examples.
FIG. 21 is a diagram showing an RF spectrum obtained in an example.
FIG. 22 is a diagram showing an I / Q diagram obtained in the example.
FIG. 23 is a diagram illustrating a method according to a third embodiment of the present invention, which can perform modulation with an arbitrary modulation format.
FIG. 24 is a diagram illustrating the operation of the 90 ° hybrid circuit.
FIG. 25 is a diagram illustrating a phasor of each unit when a BPSK signal is generated by the method of the present invention.
FIG. 26 is a diagram illustrating a phasor of each unit when a BPSK signal is generated by the method of the present invention.
FIG. 27 is a diagram illustrating a QPSK signal phasor taking into consideration the RF intensity dependence of the DC bias of the optical modulator.
FIG. 28 is a diagram showing an experimental system used in Examples.
FIG. 29 is a diagram showing an RF spectrum obtained in an example.
FIG. 30 is a diagram showing an I / Q diagram obtained in the example.
FIG. 31 is a diagram showing an IF band spectrum obtained in the example.
FIG. 32 is a conceptual diagram showing an optical spectrum having a detuning frequency corresponding to a millimeter wave / microwave band frequency.
FIG. 33 is a diagram showing a Bessel function.
FIG. 34 is a diagram illustrating a schematic spectrum diagram capable of comprehensively expressing phases and frequencies.
FIG. 35 is a diagram illustrating a schematic spectrum diagram capable of comprehensively expressing phases and frequencies.
FIG. 36 is a diagram illustrating a schematic spectrum diagram capable of comprehensively expressing phases and frequencies.
FIG. 37 is a diagram illustrating a configuration of an MZM that performs an optical DSC-SC method.
FIG. 38 is a diagram schematically illustrating the spectrum of light of each unit that has been subjected to optical DSC-SC modulation.
FIG. 39 is a diagram showing an NA-PSK signal generation circuit.
FIG. 40 is a schematic diagram of the spectrum of a two-mode optical millimeter-wave / microwave BPSK signal.
[Explanation of symbols]
1 Single wavelength light source (single wavelength laser diode)
2 Electro-optic crystal
3 Branch optical waveguide
4 Optical modulator
5 Optical modulator
6 π / 2 fixed phase shifter
7 Branch optical waveguide
8 Optical fiber
9. Photodetector (photodiode)
10 Branch optical waveguide
11a, 11b Optical phase modulator
12 π fixed phase shifter
13 RF signal
14. RF fixed phase shifter for RF
15 RF signal
16 Optical Band Rejection Filter (Fiber Bragg Grating)

Claims (10)

Splitting the single-wavelength light into a first light and a second light, and inputting the split first light and the second light to a first light modulator and a second light modulator, respectively; A first RF signal and a second RF signal are input to the first optical modulator and the second optical modulator, respectively, and the first light and the second light are respectively subjected to DSB (double sideband) modulation. Then, a fixed phase π / 2 is added between the first light and the second light subjected to the DSB modulation and added,
Optical millimeter-wave / microwave, characterized in that at the heterodyne reception output of the added first light and second light, interference components of the first RF signal and the second RF signal are suppressed. Signal generation method. The single-wavelength light splitting, adding means, light modulating means, and fixed phase π / 2 adding means are integrated on an electro-optic crystal to form an optical waveguide by applying an electric field to the branched optical waveguide and the optical waveguide. 2. An optical SSB modulator comprising a plurality of optical phase modulators for adding a phase to light propagating in a waveguide, and a π / 2 fixed phaser for applying a constant electric field to the optical waveguide. 3. An optical millimeter-wave / microwave signal generation method according to claim 1. The single-wavelength light splitting, adding means, light modulating means, and fixed phase π / 2 adding means are integrated on an electro-optic crystal to form an optical waveguide by applying an electric field to the branched optical waveguide and the optical waveguide. A two-electrode MZM optical modulator composed of a plurality of optical phase modulators for adding a phase to light propagating in a waveguide, and a π / 2 fixed phaser for applying a constant electric field to the optical waveguide; The optical millimeter-wave / microwave signal generation method according to claim 1, further comprising an optical band rejection filter for removing a carrier after adding the light and the second light. The heterodyne received light output is a heterodyne output between primary modulation components of upper and lower sidebands of the first RF signal and the second RF signal, respectively, and the first RF signal and the second RF signal respectively. The frequency of the first RF signal and the second RF signal is twice as high as that of the first RF signal and the second RF signal. An optical millimeter-wave / microwave signal generation method as described in the above. The heterodyne received light output is a heterodyne output between the secondary modulation components of the upper and lower sidebands of the first RF signal and the second RF signal, respectively, and the first RF signal and the second RF signal respectively. The frequency of the first RF signal and the second RF signal is four times the input frequency of the first RF signal and the second RF signal. An optical millimeter-wave / microwave signal generation method as described in the above. The first RF signal and the second RF signal are mutually independent NA-BPSK (narrow anglebinary phase shift keyig) signals having a phase shift amount of π / 2 or π / 4, and the first RF signal And the second RF signal are delayed so as to have a phase difference of π / 4 or π / 8 from each other, so that a phase difference of π / 2 occurs between two BPSK signals generated in the heterodyne light reception. 6. The optical millimeter-wave / microwave signal generation method according to claim 4, wherein two BPSK signals are added to obtain a quadrature phase shift keying (QPSK) signal. The first RF signal and the second RF signal are NA-BPSK signals having a phase shift amount of π / 2 or π / 4, and having phase changes opposite to each other, and By delaying the signal and the second RF signal so as to have a phase difference of π / 2 or π / 4 from each other, the two BPSK signals generated in the heterodyne light reception are changed in a transient transition state between symbols of the signal. 6. The optical millimeter-wave / microwave signal generation method according to claim 4, wherein the unnecessary components cancel each other to prevent spectrum broadening. By making the frequencies of the first RF signal and the second RF signal different from each other, the frequencies of the two signals generated in the heterodyne light reception are different, and the interference component between the two signals is suppressed. The method of generating an optical millimeter-wave / microwave signal according to any one of claims 1 to 7, wherein subcarrier multiplexing is possible. A means for generating the first RF signal and the second RF signal, wherein the third RF signal not modulated and the fourth RF signal obtained by giving a phase change of 90 degrees to the third RF signal; Circuit for synthesizing a fifth RF signal modulated by an arbitrary modulation method for giving information on the phase and amplitude of the signal and a sixth RF signal obtained by giving a phase change of 90 degrees to the fifth RF signal And the electric circuit combines the third RF signal and the sixth RF signal to generate a first RF signal, and the fourth RF signal and the fifth RF signal The optical millimeter-wave / microwave signal generation method according to any one of claims 1 to 8, wherein the electrical circuit is a circuit that generates a second RF signal by combining the signal and the signal. An optical millimeter-wave / microwave signal generation apparatus using the optical millimeter-wave / microwave signal generation method according to claim 1.

Images