JP3811134B2 - Optical amplifier - Google Patents

Description

【００４８】
σｅ：誘導放出断面積
σａ：誘導吸収断面積
Ｎ２（ｘ）：位置ｘでの増幅の始凖位(^４Ｉ_１３／２）のイオン数
Ｎ１（ｘ）：位置ｘでの増幅の終凖位(^４Ｉ_１５／２）のイオン数
Ｎｔｏｔａｌ：全イオン数
【００４９】
ここで（ｅｑ１）、（ｅｑ２）式より、
Ｇａｉｎ（λ，ｘ）＝(σｅ（λ）Ｎ２（ｘ）−（σａ（λ）（Ｎｔｏｔａｌ−Ｎ２（ｘ）） （ｅｑ３）
を得る。
【００５０】
（ｅｑ３）式から解るように、変数はＮ２のみになる。
【００５１】
ここで、誘導放出断面積および誘導吸収断面積は、増幅媒体に添加される希土類イオンの物性で決定しているため、利得スペクトルを固定するためにはＮ２を固定すること、即ち増幅始準位と増幅終凖位のイオン数を固定する必要がある。
【００５２】
ＷＤＭ伝送などでは、伝送路の損失の経時変化や信号チャネル数の変化により光増幅器への入力信号光のパワーが変化する。このような変化は、石英系ＥＤＦＡでは、^４Ｉ_１３／２に存在するＥｒ^３＋の誘導放出速度が変化することに繋がり、結果として、Ｎ２が変化し、（ｅｑ３）から解るように利得スペクトルが変化する。
【００５３】
従って、石英系ＥＤＦを増幅媒体として用いる光増幅器のように、増幅に関与する準位が２つである場合には、任意の信号光波長をモニタし、この信号波長での利得を一定に保つように励起光パワーを制御すれば、光増幅器への入力パワーに関わらず（ｅｑ３）に基づいて利得スペクトルは一定に保たれる。
【００５４】
利得を一定に制御するための具体的手順は、増幅用光ファイバの入力端および出力端で信号光を分波して、それぞれの場所での信号光パワーを検出し、その検出値から信号光の利得を算出し、算出された利得値と設定値との誤差成分が零になるように励起光量を調節する。これが石英系ＥＤＦを増幅用光ファイバとして用いた光増幅器で利得プロファイルを一定に制御するため原理である。
【００５５】
ところで、このように増幅メカニズムが増幅始凖位と終凖位にいるイオンのみで記述できる系は非常にまれである。たとえば、エルビウム添加光ファイバで前記と同じ誘導放出を利用する場合においても、光ファイバを構成するホストガラスをフッ化物ガラスに変えただけでも、増幅の始凖位よりさらに高エネルギーの励起凖位への遷移と、遷移した先の凖位での寿命が無視できなくなる。このような場合には、ある波長（例えば入力信号光の１つの波長）での利得を一定に制御したとしても、利得プロファイルは一意的に決まらなくなる。これは、増幅動作を記述するにあたり、増幅始凖位、増幅終凖位およびそれ以外のイオン数を考慮する必要が生じたためである。即ち、例えばＴＤＦＡでは、総イオン数が下記の（ｅｑ２’）により表され、上記（ｅｑ３）のように１つのパラメータ（Ｎ２）を考慮するのみでは不十分となり、その他の準位のイオン数を考慮しなければならなくなったためである。
【００５６】
【数２】

【００５７】
３つの準位が増幅に大きく関与する希土類添加光ファイバを用いる光増幅器には、ツリウム添加光ファイバ（ＴＤＦ）を用いた光増幅器がある。この光増幅器では、図２（Ｂ）に示すように、４準位の光増幅器であるが、増幅に大きく関与する準位が３つ存在する。即ち、ＴＤＦＡでは、増幅に大きく関与するエネルギー凖位は、増幅の始凖位、終凖位および基底凖位が存在する。従って、これらの準位を全て考慮する必要があるため、ＥＤＦＡでの議論のように１波長の利得をモニターし、これを一定にするように制御するだけでは前記３つの凖位間のイオン数を一定に保つことはできない。
【００５８】
図３は、ＷＤＭ信号を増幅する場合、ＴＤＦＡにおいて光増幅器へ入射する波長λ_１の信号光のパワーと、ＴＤＦＡより増幅されて出射された波長λ_１の信号光のパワーより計算される利得を一定にするようにＴＤＦＡの励起光量を調節した場合の、光増幅器の特性を示す図である。
【００５９】
図３中に示すａ〜ｃは、ＴＤＦＡへの全入力信号パワーが低い場合をａ、中間の値の場合をｂ、高い場合をｃとして、光増幅器の特性を示したものである。この特性を評価した条件は、ＴＤＦＡのＴｍ^３＋添加濃度が６０００ｐｐｍであり、励起波長が１４００ｎｍであった。増幅用のＴｍ添加フッ化物光ファイバの励起は、この光ファイバに対して前方向および後方向より励起を行う双方向励起である。前方向励起光パワーと後方向励起パワーの比は常に一定とした。図３より明らかなように、単に一波長の利得を一定にしただけでは入力信号光パワーの変化に対して利得プロファイルが変化する。
【００６０】
このような増幅に大きく関与する準位が３つ存在する増幅媒体を含む光増幅器を制御する方法として次のような制御方法が提案されている。この制御方法は、ＴＤＦＡ光増幅器に関するもので、２通りの方法がある。この制御方法の１つは、双方向励起のＴＤＦＡ光増幅器で利用される。この制御方法では、２種類の信号光の利得をモニタする。このモニタ情報に基づいて、増幅用光ファイバの前方と後方の励起光のパワーの比を制御することにより利得帯域を調節し、且つ、総励起光パワーを制御することにより全体の利得を調節する方法である。ＴＤＦＡでは、前方と後方の励起光のパワー比を変化させると利得スペクトルが波長シフトするという現象が発生する。
【００６１】
また、ＴＤＦＡの第２の制御方法として、２波長（１．４μｍ、１．５６μｍ）の励起光源の光量を制御し、利得スペクトルを調節する方法がある（Won Jae Lee et al., "Gain excursion & tilt compensation algorithm for TDFA using 1.4 μm/1.5μm dual wavelength control"., OFC2002, ThZ3（非特許文献５）参照）。
【００６２】
上述のような実質的に増幅に関与する準位が２つである増幅媒体を含む光増幅器を除き、希土類イオンを添加した増幅媒体を用いた光増幅器の場合、励起光源の光量の調節により一波長の利得を一定値に制御するのみでは、入力信号のレベルの変動や他の条件の変化（例えば温度変化）に対して、利得プロファイルを一定に制御することはできない。このため、このような系では基本的に２種類の信号波長での利得をモニタすることが必要となり、複雑な制御をする必要があった。
【００６３】
本発明の光増幅器および制御方法では、増幅始凖位、増幅終凖位およびその他の凖位を考慮する必要がある希土類を添加した増幅媒体を用いることができる。以下では、その他の準位が第３の準位のみである場合について説明するが、本発明はこれに限定されず、さらに多くの準位が関与する場合も含む。
【００６４】
本発明は、入力信号のレベルの変動に応じて、その制御光の光量を調節することで増幅媒体の反転分布状態を一定にすることを特徴とする。つまり、本発明の基本的な概念は、入力信号のレベルの変動分を増幅帯域内の波長の制御光で補償することにある。本発明によれば、あらゆる希土類イオンの光増幅器に対し、増幅帯域内の一波長の利得を一定にするように制御光を調節することで利得プロファイル一定制御を実現する。本発明では、制御光源を調節するために、光のある波長（またはある波長範囲）における利得をモニタすることを特徴とする。このように制御光または少なくとも１つの信号光をモニタして、制御光源を調節するのは、例えば、制御光と信号光の総出力または総入力パワーをモニタして制御光のパワーを制御する方法などでは、利得スペクトルが歪み、利得プロファイルを一定に制御することはできないためである。このことを、制御光と信号光の総出力または総入力パワーをモニタして、制御光のパワーを制御する場合を例に取り、図４および図５を参照して説明する。
【００６５】
図４は、総入力パワーを一定にしただけでは利得プロファイルが一定に制御できないことを示す図である。図４は、信号波長以外の増幅帯域内の波長を有する制御光源を入力側に備えたＴＤＦＡにおいてＷＤＭ信号を増幅する際に、光増幅器への総信号光パワーの入力の変動に対して、制御光の入力パワーと総入力信号光パワーの合計（以下、総入力パワーとも称する）が常に一定になるように制御光源の光量を調節した場合の利得スペクトルを表す図である。
【００６６】
図４に示すａ〜ｃは、ＴＤＦＡへの総入力信号パワーが低い場合をａ、中間の場合をｂ、高い場合をｃとしたときの利得スペクトルである。この特性を評価した条件は、ＴＤＦＡのＴｍ^３＋添加濃度が４０００ｐｐｍであり、励起波長が１４１０ｎｍであった。増幅用のＴｍ添加フッ化物光ファイバの励起は、この光ファイバに対して前方向および後方向より励起を行う双方向励起であり、総励起光パワーは常に一定とした。図４より明らかなように、光増幅器への総入力パワーを一定にしただけでは入力信号光パワーの変化に対して利得プロファイルが変化する。
【００６７】
図５は、総出力パワーを一定にしただけでは利得プロファイルが一定に制御できないことを示す図である。図５は、信号は長以外の増幅帯域内の波長を有する制御光源を入力側に備えたＴＤＦＡにおいてＷＤＭ信号を増幅する際に、光増幅器への総信号光パワーの入力の変動に対して、制御光の出力パワーと総出力信号光パワーの合計（以下、総出力パワーとも称する）が常に一定になるように制御光源の光量を調節した場合の利得スペクトルを表す図である。
【００６８】
図５に示すａ〜ｃは、ＴＤＦＡへの総入力信号パワーが低い場合をａ、中間の場合をｂ、高い場合をｃとしたときの利得スペクトルである。この特性を評価した条件は、ＴＤＦＡのＴｍ^３＋添加濃度が５０００ｐｐｍであり、励起波長が１４２０ｎｍであった。増幅用のＴｍ添加フッ化物光ファイバの励起は、この光ファイバに対して前方向および後方向より励起を行う双方向励起であり、総励起光パワーは常に一定とした。図５より明らかなように、光増幅器への総出力パワーを一定にしただけでは入力信号光パワーの変化に対して利得プロファイルが変化する。
【００６９】
このように、総入力パワーまたは総出力パワーを一定にしただけでは、利得プロファイルを一定に制御することができない。
【００７０】
本発明では、光増幅器は、増幅媒体、該増幅媒体を励起するための励起手段、前記増幅媒体の増幅帯域内の波長を有する制御光を前記増幅媒体に入力し、その入力パワーを変化させることで前記増幅媒体の反転分布状態を制御できる１以上の制御光源、前記増幅媒体に入出力する光パワーをモニタする監視手段、および、前記監視手段より得られた値から該光の利得を算出し、予め定められた値若しくは外部より入力される値と一致するように前記制御光源を制御するための制御部とを含む。
【００７１】
また、本発明の光増幅器は、制御光源の制御光の光量を調節する手段を具備する。この手段には、制御光源の電流値を調整することや、制御光源の出力端に可変光減衰器を設けて減衰量を調節すること等が含まれる。
【００７２】
本発明には、増幅媒体、該増幅媒体を励起するための励起手段、前記増幅媒体の増幅帯域内の波長を有する制御光を前記増幅媒体に入力し、その入力パワーを変化させることで前記増幅媒体の反転分布状態を制御できる１以上の制御光源、前記増幅媒体に入出力する複数の波長の光のパワーをモニタする監視手段、前記監視手段より得られた値から複数の波長の利得を算出後、複数の利得の差と和を算出し、複数の利得の差が予め定められた値若しくは外部より入力される値と一致するように前記制御光源を制御し、複数の利得の和が予め定められた値若しくは外部より入力される値と一致するように前記励起手段を制御するための制御部とを含む光増幅器が含まれる。
【００７３】
本発明では、監視手段は、制御光をモニタするもの、１チャンネル以上の信号光をモニタするもの、または、増幅帯域内で予め定められた波長範囲の光をモニタするものでありうる。
【００７４】
さらに、本発明では、監視手段より得られた値に応じ、それらの値若しくは前記制御部で算出される利得、利得の差、または利得の和を補正する補正手段を含むことが好ましい。
【００７５】
また、本発明では、監視手段には、光バンドパスフィルタを含むことが好ましい。
【００７６】
上記本発明の利得プロファイル一定制御の原理に従えば、本発明の制御方法も提供される。
【００７７】
この制御方法の１つは、増幅媒体として希土類イオンをコアおよび／またはクラッドに添加した希土類添加光ファイバまたは希土類添加光導波路、該増幅媒体を励起するための励起手段、１以上の制御光源であって、該制御光源からの制御光の入力パワーを変化させることで前記増幅媒体の反転分布状態を制御できるもの、監視手段、および、制御部とを含む光増幅器の利得プロファイルを一定に制御する方法であって、前記増幅媒体の増幅帯域内の波長を有する制御光を前記増幅媒体に入力することと、前記増幅媒体に入出力する光パワーをモニタすることと、前記監視手段より得られた値から該光の利得を算出し、予め定められた値若しくは外部より入力される値と一致するように前記制御光源を制御することとを含む。
【００７８】
本発明では、増幅媒体に添加される希土類イオンは、１種類であっても複数種類であってもよい。
【００７９】
以下に図面を参照して本発明の光増幅器および利得プロファイル一定制御の方法をさらに具体的に説明するが、これらは例示であり、本発明はこれらに限定されない。
【００８０】
本発明は、光増幅器の利得プロファイル一定制御に、制御光を利用するように構成された光増幅器に関する。
【００８１】
光増幅器に制御光を導入し、この制御光を利用して光増幅器の利得プロファイルを一定に保つように制御する。本発明では、制御光源を調節するために、光のある波長（またはある波長範囲）における利得をモニタすることを特徴とする。
【００８２】
本発明の概要を、図６を参照して説明する。
【００８３】
本発明の光増幅器６００は、監視手段６０１，６０３、増幅部６０２と制御システム６０４を具備する。監視手段６０１は、制御光源６１０内に設けられた光検出器から構成される。また、監視手段６０３は、光分岐素子６３４と光電変換素子のような光検出器６３５から構成される。図６に示す監視手段６０１は、制御光源６１０内に光検出器を含む例を示したが、本発明はこれに限定されず、光検出器を別途設けてもよい。この場合、光分岐素子（分波器）を設け、信号光入力の一部を分波すればよい。光検出器からの信号は制御システム６０４に入力される。光検出器は、例えば光電変換素子などが含まれる。増幅部６０２は、制御光源６１０、増幅媒体６１２、励起光源６１４，６１６、制御光源からの光を合波するための合波器６１８、励起光源からの光を合波するための合波器６２０，６２２を具備する。増幅部は、図６に示されるようにアイソレータ６２４，６２６を設けることも好ましい。
【００８４】
制御光源６１０は、内部に光検出器を有するものでも、有しないものでもよい。本発明では、内部に光検出器を有する制御光源が好ましい。内部に光検出器を有する制御光源は、モニタ用の第１の信号６２８を出力できるものである。また、制御光源は複数存在していてもよい。
【００８５】
制御システム６０４は、制御光源を制御するための制御部６３０、および演算部６３２を有する。この演算部へは増幅部６０２から出力された光の一部を取り出す光分岐素子（分波器）６３４からの第２の信号６３６が入力される。演算部６３２では、制御光源６１０からの第１の信号と、光分岐素子６３４からの第２の信号から制御光の利得を算出し、予め設定された値と一致するような制御光源６１０の電流値を算出する。その値は制御部６３０へ送られる。制御部６３０は、演算部６３２から送られてきた電流値になるように制御光源６１０の光量を調整する。本明細書では、制御システム６０４は、制御部とも称される。この場合、制御部は、上記機能を有する制御部６３０と演算部６３２を含むものである。
【００８６】
なお、本発明において、各構成要素は、図６に示される配置のみではなく、種々の配置をとることができる。例えば、図６に示される光増幅器では、励起光は双方向励起であるが、本発明はこれに限定されず、前方励起および後方励起であってもよい。
【００８７】
増幅媒体には、希土類元素イオンをコアおよび／またはクラッドに添加した希土類添加光ファイバを好適に用いることができる。特に本発明では、ツリウム添加光ファイバ、エルビウム添加光ファイバ、ホルミウム添加光ファイバなどを使用することができる。具体的には、増幅媒体としては、希土類イオンを添加した増幅用光ファイバを用いることができ、これは、ホストガラスとして石英ガラス、ビスマス系ガラス、非輻射遷移の起こりにくいフッ化物系ガラスのＺＢＬＡＮガラス、Ｉｎ−Ｐｂガラスまたはテルライトガラス等である。希土類イオンは、いずれのイオンであってもよいが、ツリウムイオン（Ｔｍ^３＋）、ホルミウムイオン（Ｈｏ^３＋）、エルビウムイオン（Ｅｒ^３＋）が好ましい。
【００８８】
合波器および光分岐素子（分波器）は、溶融延伸のファイバ型（分岐型、波長分割多重型いずれも）、誘電体多層膜型またはファイバグレーティングと組み合わせたサーキュレータ等である。また、合波器は、励起光／信号光合波カップラのような通常の合波器を使用することもできる。
【００８９】
制御光を発生する制御光源、励起光を発生する励起光源は、例えばＮｄ−ＹＬＦレーザ、Ｎｄ−ＹＡＧレーザ、Ｔｉサファイアレーザなどの固体レーザや半導体レーザまたはファイバレーザ等がある。図６に示される光増幅器では、励起光は双方向励起であるが、本発明はこれに限定されず、前方励起および後方励起であってもよい。
【００９０】
以下に本発明の光増幅器を用いた制御方法（本発明の光増幅器の動作）について具体的に説明する。以下では、監視手段６０１の光検出器が制御光源に含まれ、制御光の利得をモニタする場合を例に取り説明する。
【００９１】
信号光は合波器６１８で制御光源６１０からの制御光と合波される。次いで、合波された信号光はアイソレータ６２４を通過し、合波器６２０で励起光源６１４からの励起光と合波される。励起光と合波された信号光は、増幅媒体６１２に入る。増幅媒体は、励起光源６１４と励起光源６１６で双方向から励起される。増幅媒体を出た信号光は、合波器６２２、アイソレータ６２６を通過し、分波器６３４で分波される。分波器６３４で分波されなかった信号光の一部は信号出力光として出力される。
【００９２】
制御光源から増幅用光ファイバ６１２へ入力される制御光は、制御光源６１０の内部の光検出器で検出され、その制御光のパワーは、第１の信号として演算部６３２へ送られる。増幅媒体から出力された制御光は、分波器６３４で分波され、その制御光のパワーは、第２の信号６３６として演算部６３２に送られる。入力および出力端での制御光のパワーは、電気信号に変換され演算部６３２に入力されることが好ましい。
【００９３】
演算部６３２では、受信した検出値から制御光の利得を算出し、予め設定された値と一致するような制御光源６１０の電流値を算出し、その値は制御部６３０へ送られる。制御部６３０は、演算部６３２から送られてきた電流値になるように制御光源６１０の光量を調整する。
【００９４】
制御光の波長は増幅帯域内にあるため、制御光の利得を一定にするように制御光を調節することで利得プロファイルを一定に制御することができる。
【００９５】
本発明では、演算部に入力される信号は、制御光に基づくもののみではなく、１チャンネル以上の信号光に基づくもの、または、増幅帯域内で予め定められた波長範囲の光に基づくものであってもよい。また、本発明の制御方法では、補正手段により、監視手段から得られた値に応じ、それらの値、または、前記制御部で算出される利得、利得の差、若しくは利得の和を補正することが好ましい。
【００９６】
また、本発明では、監視手段に光バンドパスフィルタを含め、モニタ信号をこれに通すことが好ましい。
【００９７】
以上述べた各発明の光増幅器では、特に増幅媒体内での信号光の伝搬方向については規定していなかったが、光増幅器が以下の（ａ）〜（ｃ）の構成を有し、信号光がこれらに示される伝搬方向をとることができる。
【００９８】
（ａ）増幅媒体の入力側または出力側の両方、或いは、これらのどちらか一方に光アイソレータが接続される。信号光は、増幅媒体内を一方向のみに伝搬する（本明細書中で、シングルパスという）。
【００９９】
（ｂ）増幅媒体の一方に光サーキュレータが接続され、増幅媒体の他端には少なくとも信号光を反射させるミラー機能を有するデバイスが接続される。信号光は光サーキュレータを通過して増幅媒体中を進行し、増幅される。次いで、信号光は、これを反射させるミラーにより折り返されて増幅媒体中を反対方向に進行し、増幅される。増幅された信号光は、サーキュレータを通過して出射する（本明細書で、ダブルパスという）。
【０１００】
（ｃ）増幅媒体の両方に光サーキュレータが接続される。２種類の進行方向の異なる信号光が同時または時間的に分離された状態で増幅媒体中を互いに逆方向に進行し、増幅される（以下、双方向光増幅器という）。
【０１０１】
（実施例）
以下、図面を参照して本発明の実施例について説明する。
【０１０２】
（実施例１）
図７は、本発明の光ファイバ増幅器の本実施例を示す構成図である。図７に示すように、この光ファイバ増幅器は、増幅媒体、例えばコアに希土類イオンを添加した増幅用光ファイバ３３０４、アイソレータ３３０２、３３０５、３３０８、制御光と信号光を合波する合波器３３０１、励起光源３３０９からアイソレータ３３０８を介して供給される励起光とアイソレータ３３０２からの信号光・制御光とを合波する合波器３３０３、増幅用光ファイバ３３０４からアイソレータ３３０５を介して取り出された出力光を分岐する分波器３３０６、分波器３３０６の一方出力から制御光を抽出するためのフィルタ３３１０、内部に光検出器を有する制御光源３３０７、励起光を発生する励起光源３３０９、フィルタ３３１０から出力された制御光を検出する光検出器３３１１、制御光源３３０７からの信号と光検出器３３１１からの信号とを供給される演算部３３１３、および演算部３３１３からの出力により制御光源３３０７を制御する制御部３３１２より構成される。
【０１０３】
希土類イオンを添加した増幅用光ファイバのホストガラスとして石英ガラス、ビスマス系ガラス、非輻射遷移の起こりにくいフッ化物系ガラスのＺＢＬＡＮガラス、Ｉｎ−Ｐｂガラスまたはテルライトガラス等がある。
【０１０４】
合波器３３０１、分波器３３０６として溶融延伸のファイバ型（分岐型、波長分割多重型いずれも）、誘電体多層膜型またはファイバグレーティングと組み合わせたサーキュレータ等が使用可能である。合波器３３０３として、溶融延伸のファイバ型、誘電体多層膜型またはファイバグレーティングと組み合わせたサーキュレータ等が使用可能である。制御光を発生する制御光源３３０７、励起光を発生する励起光源３３０９としてＮｄ−ＹＬＦレーザ、Ｎｄ−ＹＡＧレーザ、Ｔｉサファイアレーザなどの固体レーザや半導体レーザまたはファイバレーザ等が使用可能である。フィルタ３３１０としては、誘電体多層膜フィルタ、ファイバグレーティング等がある。
【０１０５】
本実施例では、信号光は１４８０〜１５１０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを６０００ｐｐｍ添加した５ｍの光ファイバ、制御光源の波長は１５２０ｎｍ、励起波長は１４００ｎｍ、アイソレータ３３０２、３３０５は中心波長１４８０ｎｍ、合波器３３０１および分波器３３０６は１５１２ｎｍ以下と１５１８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、合波器３３０３は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、およびフィルタ３３１０は、中心波長１５２０ｎｍ、透過帯域の半値幅０．８ｎｍのものを使用した。また励起光パワーは３００ｍＷに固定とした。
【０１０６】
図７に示すように、信号光は合波器３３０１にて制御光源３３０７からの制御光と合波され、アイソレータ３３０２を通過し、合波器３３０３にて更にアイソレータ３３０８からの励起光と合波され、増幅用光ファイバ３３０４を通過後、アイソレータ３３０５を通過し、分波器３３０６にて分波され、その一方は信号光として出力される。
【０１０７】
増幅用光ファイバ３３０４への入力制御光源パワーは、制御光源３３０７の内部の光検出器にて検出される。増幅用光ファイバ３３０４からの出力制御光パワーは、分波器３３０６で分波され、フィルタ３３１０を通過後、光検出器３３１１にて検出され、その値は電気信号として演算部３３１３に送られる。
【０１０８】
演算部３３１３では、受信した検出値から制御光の利得を算出し、予め設定された値と一致するような制御光源３３０７の電流値を算出し、その値は制御部３３１２へ送られる。
【０１０９】
制御部３３１２は、演算部３３１３から送られてきた電流値になるように制御光源３３０７の光量を調整する。
【０１１０】
制御光の波長は増幅帯域内にあるため、制御光の利得を一定にするように制御光を調節することで利得プロファイル一定制御を実現することが可能となる。
【０１１１】
図８は、実施例１において各入力信号レベルを変えた場合の利得プロファイルの偏差の最大値をプロットした図である。つまり、各入力信号レベルを−２５、−２０、−１５、−１０、−５ｄＢｍ／ｃｈ．に設定し入力した場合の利得スペクトルについて、−２５ｄＢｍ／ｃｈ．での利得スペクトルを基準として、入力が変化した際の利得スペクトルと基準スペクトルとの差の最大値をプロットした図である。利得の評価は、−３５ｄＢｍの小信号プローブ光をスキャンすることにより測定した。図８の点線は、モニタする信号光による利得制御をかけない従来例の結果である。図８の実線は、１５２０ｎｍの制御光のモニタ系を用いて、１５２０ｎｍでの利得が一定になるように制御した本発明の場合の値をプロットした結果である。この結果、本発明によれば利得プロファイル制御が可能となることが明らかになった。
【０１１２】
（実施例２）
図９は、本発明の光ファイバ増幅器の第２実施例を示す構成図である。図９において図７と同様の個所には同一の参照番号を付してその説明を省略する。図９に示すように、この光ファイバ増幅器は、図７の構成に加え、増幅器へ入出力する信号光を分波する分波器３５０１、分波器３５０１からの信号光を抽出するためのフィルタ３５０２、制御光源３５０４、フィルタ３５０２から入力する信号光パワーを検出する光検出器３５０３より構成される。本実施例においては、分波器３３０６、フィルタ３３１０および光検出器３３１１は、それぞれ出力する信号光を分波し、抽出し、およびパワーを検出する。
【０１１３】
分波器３５０１として溶融延伸のファイバ型（分岐型、波長分割多重型いずれも）、誘電体多層膜型、およびファイバグレーティングと組み合わせたサーキュレータ等が使用可能である。制御光を発生する制御光源３５０４としてＮｄ−ＹＬＦレーザ、Ｎｄ−ＹＡＧレーザ、Ｔｉサファイアレーザなどの固体レーザや半導体レーザまたはファイバレーザ等が使用可能である。フィルタ３５０２としては、誘電体多層膜フィルタ、ファイバグレーティング等がある。
【０１１４】
本実施例では、信号光は１４７０〜１５００ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを４０００ｐｐｍ添加した１０ｍの光ファイバ、制御光源３３０４の波長は１４６０ｎｍ、励起波長は１４００ｎｍ、アイソレータ３３０２、３３０５は中心波長１４８０ｎｍ、分波器３５０１および分波器３３０６は分岐比９５：５のファイルカプラ、合波器３３０１は１４６２ｎｍ以下と１４６８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、合波器３３０１は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、フィルタ３５０２、３３１０は、中心波長１５１０ｎｍ、透過帯域の半値幅０．８ｎｍのものを使用した。また励起光パワーは３００ｍＷに固定とした。なお、モニタする波長多重信号光のうち１チャネルは、１５１０ｎｍの入力光とする。
【０１１５】
図９に示すように、信号光は分波器３５０１にて分波され、分波器３５０１からの信号光は、合波器３３０１にて制御光源３５０４からの制御光と合波され、アイソレータ３３０２を通過し、波長分割多重カプラ３３０３にて更にアイソレータ３３０８からの励起光と合波され、増幅用光ファイバ３３０４を通過後、アイソレータ３３０５を通過し、分波器３３０６にて分波され、その一方は信号光として出力される。
【０１１６】
１５１０ｎｍの入力信号光パワーは、分波器３５０１にて分岐され、フィルタ３５０２を通過後、光検出器３５０３にて検出され、１５１０ｎｍの出力信号光パワーは、分波器３３０６で分波されフィルタ３３１０を通過後、光検出器３３１１にて検出され、それらの値は電気信号にて演算部３３１３に送られる。
【０１１７】
演算部３３１３では、受信した検出値から１５１０ｎｍの信号光の利得を算出し、予め設定された値と一致するように制御光源３５０４の電流値を算出し、その値は制御部３３１２へ送られる。制御部３３１２は演算部３３１３から送られてきた電流値になるように制御光源３５０４の光量を調整する。ここでは、１５１０ｎｍの１チャネルの信号光パワーの検出値に基づいて制御光源３５０４の光量を制御したが、２以上の複数のチャネルを用いて、制御する他の実施例も可能である。さらに、モニタする信号光として予め定めた波長範囲の全信号光量を用いて、制御を行う実施例も可能である。この場合、本実施例では例えば、１４７０〜１５００ｎｍの範囲の各チャネルについての信号光パワーの総量を算出して、これをもとに利得を計算することにより、上述のとおり制御光源３５０４の光量を調整することができる。
【０１１８】
図１０は、各入力信号レベルを変えた場合の、利得プロファイルの偏差の最大値をプロットした図である。つまり、各入力信号レベルを−２５、−２０、−１５、−１０、−５ｄＢｍ／ｃｈ．に設定し入力した場合の利得スペクトルについて、−２５ｄＢｍ／ｃｈ．での利得スペクトルを基準として、入力が変化した際の利得スペクトルと基準スペクトルとの差の最大値をプロットした図である。利得の評価、−３５ｄＢｍの小信号プローブ光をスキャンすることにより測定した。図１０の点線は、モニタする信号光による利得制御をかけない従来例の結果である。図１０の実線は、１５１０ｎｍの信号光のモニタ系を用いて、１５１０ｎｍでの利得が一定になるように制御した本発明の場合の値をプロットした結果である。この結果、本発明によれば利得プロファイル制御が可能となることが明らかになった。
【０１１９】
（実施例３）
図１１は、本発明の光ファイバ増幅器の第３実施例を示す構成図である。図１１において図９と同様の個所には同一の参照番号を付してその説明を省略する。図１１に示すように、この光ファイバ増幅器は、図９に加えて出力光のうち分波器３３０６から分岐された出力光を分岐する分波器３７０２、分波器３７０２からの制御光を抽出するためのフィルタ３７０３、フィルタから抽出された制御光パワーを検出する光検出器３７０４、アイソレータ３３０８からの励起光を合波器３３０３および３７０５に分岐させる分波器３７０１および増幅用光ファイバ３３０４からの信号光と分波器３７０１からの励起光を合波する合波器３７０５より構成される。ここで、制御光源３３０７は、入力制御光源パワーを検出するための光検出器を内部に有する。
【０１２０】
合波器３７０５として、溶融延伸のファイバ型、誘電体多層膜型またはファイバグレーティングと組み合わせたサーキュレータ等が使用可能である。フィルタ３７０３としては誘電体多層膜フィルタ、ファイバグレーティング等がある。
【０１２１】
本実施例では、信号光は１４６０〜１４９０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを２０００ｐｐｍ添加した２０ｍの光ファイバ、制御光源３３０７の波長は１４５０ｎｍ、励起波長は１３９０ｎｍ、アイソレータ３３０２、３３０５は中心波長１４８０ｎｍ、分波器３５０１，３３０６は、分岐比９５：５のファイルカプラ、合波器３３０１は１４５２ｎｍ以下と１４５８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、分波器３７０１、３７０２は分岐比３ｄＢのファイバカプラ、合波器３３０３，３７０５は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、フィルタ３５０２、３３１０は、中心波長１４９０ｎｍ、透過帯域の半値幅０．８ｎｍのもの、フィルタ３７０３は、中心波長１４５０ｎｍ、透過帯域の半値幅が０．８ｎｍのものを使用した。なお、モニタする信号光は１４５０ｎｍの制御光と１４９０ｎｍの信号光とした。
【０１２２】
ここで、２つのモニタする信号光の利得をＧ（１４５０ｎｍ）およびＧ（１４９０ｎｍ）とし、合計利得Ｇ_Ｓおよび利得差ΔＧを以下のように定義する。
【０１２３】
Ｇ_Ｓ＝Ｇ（１４５０ｎｍ）＋Ｇ（１４９０ｎｍ） …（１）
ΔＧ＝Ｇ（１４５０ｎｍ）−Ｇ（１４９０ｎｍ） …（２）
【０１２４】
Ｇ_ＳおよびΔＧが一意に決まればＧ（１４５０ｎｍ）およびＧ（１４９０ｎｍ）も決定するので、Ｇ_ＳおよびΔＧの値を制御することで、Ｇ（１４５０ｎｍ）およびＧ（１４９０ｎｍ）を制御可能である。
【０１２５】
図１２は、実施例３において、入力信号光パワーを−１５ｄＢｍ／ｃｈ．、励起光パワーを４００ｍＷに固定し、制御光のパワーを変化させた場合の、２つのモニタする信号光の合計利得Ｇ_Ｓと利得差ΔＧをプロットした図である。Ｇ_Ｓは実線で示され、△Ｇは破線で示される。Ｇ_ＳおよびΔＧの値は、共に制御光パワーの増加に伴い、単調的に減少している。
【０１２６】
図１３は、実施例３において、入力信号光パワーを−１５ｄＢｍ／ｃｈ．、制御光パワーを−１５ｄＢｍに固定し、励起光パワーを変化させた時の、２つのモニタする信号光の合計利得Ｇ_Ｓと利得差ΔＧをプロットした図である。図１２同様にＧ_Ｓは実線で示され、△Ｇは破線で示される。励起光の増加に伴い、Ｇ_Ｓは単調増加しているのに対し、ΔＧは殆ど変化していない。
【０１２７】
図１２および図１３の結果から、次のような制御アルゴリズムを適用することで利得スペクトル制御が可能となることがわかる。モニタする信号光の利得が変化した時、まず、励起光パワーを固定したまま、励起光パワーを変化させてもほとんど変化することのない△Ｇが所望の値になるように制御光のパワーを調整する。次に制御光のパワーを固定し、所望のＧ_Ｓとなるように励起光パワーを調整する。この結果、利得スペクトル制御が可能となる。
【０１２８】
図１１に示すように、信号光は、分波器３５０１にて分波され、合波器３３０１にて制御光源３３０７からの制御光と合波され、アイソレータ３３０２を通過し、合波器３３０３にて更に分波器３７０１からの励起光と合波され、増幅用光ファイバ３３０４を通過後、分波器３７０１からの励起光と信号光とを合波する合波器３７０５を通過し、アイソレータ３３０５を通過し、分波器３３０６にて分波され、その一方は信号光として出力される。
【０１２９】
分波器３５０１で分波された一方の信号光は、フィルタ３５０２にて１４９０ｎｍの信号光のみ抽出された後、光検出器３５０３にて検出され、増幅用光ファイバに入力される制御光パワーは制御光源３３０７内の光検出器で検出され、分波器３３０６で分波された出力光は、更に分波器３７０２で分岐され、一方はフィルタ３３１０にて１４９０ｎｍの信号光のみ抽出された後、光検出器３３１１で検出され、もう一方はフィルタ３７０３にて１４５０ｎｍの制御光のみ抽出された後、光検出器３７０４で検出され、それらの検出値は電気信号に変換後、演算部３３１３に送られる。
【０１３０】
演算部３３１３では、受信した検出値から信号光および制御光の利得を算出し、これをもとにＧ_ＳおよびΔＧを算出して、予め設定されたＧ_ＳおよびΔＧと一致するように制御光源および励起光源の電流値を算出し、それらの値は制御部３３１２へ送られる。
【０１３１】
制御部３３１２は、演算部３３１３から送られてきた電流値になるように制御光源３３０７および励起光源３３０９の光量を調整する。
【０１３２】
図１４は、本発明の利得制御の際、−１０ｄＢｍ／ｃｈ．入力時の利得スペクトルを基準として各入力条件（−２５，−２０，−１５，−１０ｄＢｍ／ｃｈ．）の利得スペクトルの偏差をプロットした図である。全入力条件に対し利得偏差０．３ｄＢを実現しており、本発明の有効性が実証された。
【０１３３】
（実施例４）
本実施例の構成は、図１１を参照して実施例３で説明したものと同様であるが、演算部において数値設定を行う例である。従って、演算部での処理以外、即ち各構成の部材、装置は実施例３と同様のものを使用する。以下の説明では、図１５を参照する。
【０１３４】
本実施例では、信号光は１４９０〜１５２０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを８０００ｐｐｍ添加した４ｍの光ファイバ、制御光源３３０７の波長は１４７０ｎｍ、励起波長は１４００ｎｍ、アイソレータ３３０２、３３０５、３３０８は中心波長１４７０ｎｍ、分波器３５０１、３３０６は分岐比９５：５のファイバカプラ、合波器３３０１は１４７２ｎｍ以下と１４７８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、分波器３７０１、３７０２は３ｄＢの分岐カプラ、合波器３３０３，３７０５は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、フィルタ３５０２、３３１０は、透過帯域が１４９０〜１５２０ｎｍのもの、フィルタ３７０３は、中心波長１４７０ｎｍ、透過帯域の半値幅が０.８ｎｍのもの、を使用した。なお、モニタする信号光は１４７０ｎｍの制御光と予め定めた波長範囲の光量として、１４９０〜１５２０ｎｍの波長範囲の全信号光とした。
【０１３５】
ここで、２種類のモニタする信号光の利得をＧ（１４７０ｎｍ）およびＧ（１４９０〜１５２０ｎｍ）とし、合計利得Ｇ_Ｓおよび利得差ΔＧを以下のように定義する。
【０１３６】
Ｇ_Ｓ＝Ｇ（１４７０ｎｍ）＋Ｇ（１４９０〜１５２０ｎｍ） …（１）
ΔＧ＝Ｇ（１４７０ｎｍ）−Ｇ（１４９０〜１５２０ｎｍ） …（２）
【０１３７】
Ｇ_ＳおよびΔＧが一意に決まればＧ（１４７０ｎｍ）およびＧ（１４９０〜１５２０ｎｍ）も決定するので、Ｇ_ＳおよびΔＧを制御することで、Ｇ（１４７０ｎｍ）およびＧ（１４９０〜１５２０ｎｍ）を制御可能である。
【０１３８】
本実施例のＧ_ＳとΔＧにおいても、励起光パワーや制御光パワーの変化に対し、実施例３のＧ_ＳとΔＧと同様の傾向を示すため、同様の制御アルゴリズムを適用することにより利得スペクトル制御が可能である。また、実施例３と同様、入力された信号光は、分波器３５０１に入力し、分波器３３０６にて分波され、その一方は信号光として出力される。
【０１３９】
分波器３５０１で分波された信号光は、フィルタ３５０２にて増幅帯域内で予め定めた波長範囲である１４９０〜１５２０ｎｍの全信号光が抽出された後、光検出器３５０３にて検出され、増幅用光ファイバに入力される制御光パワーは制御光源に内蔵された光検出器により検出され、分波器３３０６で分波された信号光および制御光は、分波器３７０２で分岐され、一方はフィルタ３３１０にて１４９０〜１５２０ｎｍの全信号光パワーが抽出された後、光検出器３３１１で検出され、もう一方はフィルタ３７０３にて１４７０ｎｍの制御光のみ抽出された後、光検出器３７０４で検出され、それらの検出値は電気信号に変換後、演算部３３１３に送られ、実施例３と同様これらの検出値に基づき、制御部３３１２は制御光源３３０７および励起光源３３０９の光量を調整する。
【０１４０】
本発明の利得制御時、−２５〜−１０ｄＢｍ／ｃｈ．入力条件に対し、利得偏差０．３ｄＢ以下を実現し、本制御の有効性を実証した。
【０１４１】
（実施例５）
本実施例の構成は、図１１を参照して実施例３で説明したものと同様であり、各部材、装置も実施例３と同様のものを使用するが、増幅用光ファイバとしてＨｏ添加光ファイバを使用する。この点が実施例３と異なる。以下の説明では、図１１を参照する。
【０１４２】
本実施例では、信号光は２８９０〜２９２０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにホルミウムイオン（Ｈｏ^３＋） を１０００ｐｐｍ添加した２０ｍの光ファイバ、制御光源３３０７の波長は２８８０ｎｍ、励起波長は１２００ｎｍ、アイソレータ３３０２、３３０５は中心波長２９００ｎｍ、分波器３５０１、３３０６は分岐比９５：５のファイバカプラ、合波器３３０１は２８８２ｎｍ以下と２８８８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、分波器３７０１、３７０２は３ｄＢの分岐カプラ、合波器３３０３、３７０５は１３００ｎｍ以下と１３１０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、フィルタ３５０２、３３１０は、透過帯域が２８９０〜２９２０ｎｍのもの、フィルタ３７０３は、中心波長２８８０ｎｍ、透過帯域の半値幅が０．８ｎｍのもの、を使用した。なお、モニタする信号光は２８８０ｎｍの制御光と予め定めた波長範囲の光量として、２８９０〜２９２０ｎｍの波長範囲の全信号光とした。
【０１４３】
ここで、２種類のモニタする信号光の利得をＧ（２８８０ｎｍ）およびＧ（２８９０〜２９２０ｎｍ）とし、合計利得Ｇ_Ｓおよび利得差ΔＧを以下のように定義し、実施例４と同様に合計利得Ｇ_Ｓおよび利得差ΔＧを求め、外部より入力されたＧ_ＳおよびΔＧと一致するよう制御光源および励起光源の電流値を算出し、それらの値は制御部３３１２へ送られる。制御部３３１２は受信した電流値になるように制御光源３３０７および励起光源３３０９を調整する。
【０１４４】
Ｇ_Ｓ＝Ｇ（２８８０ｎｍ）＋Ｇ（２８９０〜２９２０ｎｍ） …（１）
ΔＧ＝Ｇ（２８８０ｎｍ）−Ｇ（２８９０〜２９２０ｎｍ） …（２）
【０１４５】
本発明の利得制御時、−２５〜−１０ｄＢｍ／ｃｈ．入力条件に対し、利得偏差０．３ｄＢ以下を実現し、本制御の有効性が実証された。
【０１４６】
（実施例６）
本実施例の構成は、図１１を参照して実施例３で説明したものと同様であり、各部材、装置も実施例３と同様のものを使用するが、増幅用光ファイバとしてＥｒ添加光ファイバを使用する。この点が実施例３と異なる。なお、以下の説明では、図１１を参照する。
【０１４７】
本実施例では、信号光は１６９０〜１７２０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにエルビウムイオン（Ｅr^３＋）を２０００ｐｐｍ添加した１５ｍの光ファイバ、制御光源３３０７の波長は１６８０ｎｍ、励起波長は８００ｎｍ、アイソレータ３３０２、３３０５は中心波長１７００ｎｍ、分波器３５０１、３３０６は分岐比９５：５のファイバカプラ、合波器３３０１は１６８２ｎｍ以下と１６８８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、分波器３７０１、３７０２は３ｄＢの分岐カプラ、合波器３３０３、３７０５は１３００ｎｍ以下と１３１０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、フィルタ３５０２、３３１０は、透過帯域が１６９０〜１７２０ｎｍのもの、フィルタ３７０３は、中心波長１６８０ｎｍ、透過帯域の半値幅が０．８ｎｍのもの、を使用した。なお、モニタする信号光は１６８０ｎｍの制御光と予め定めた波長範囲の光量として、１６９０〜１７２０ｎｍの波長範囲の全信号光とした。
【０１４８】
ここで、２種類のモニタする信号光の利得をＧ（１６８０ｎｍ）およびＧ（１６９０〜１７２０ｎｍ）とし、合計利得Ｇ_Ｓおよび利得差ΔＧを以下のように定義し、実施例５と同様に合計利得Ｇ_Ｓおよび利得差ΔＧを求め、外部より入力されたＧ_ＳおよびΔＧと一致するよう制御光源および励起光源の電流値を算出し、それらの値は制御部３３１２へ送られる。制御部３３１２は受信した電流値になるように制御光源３３０７および励起光源３３０９を調整する。
【０１４９】
Ｇ_Ｓ＝Ｇ（１６８０ｎｍ）＋Ｇ（１６９０〜１７２０ｎｍ） …（１）
ΔＧ＝Ｇ（１６８０ｎｍ）−Ｇ（１６９０〜１７２０ｎｍ） …（２）
【０１５０】
本発明の利得制御時、−２５〜−１０ｄＢｍ／ｃｈ．入力条件に対し、利得偏差０．３ｄＢ以下を実現し、本制御の有効性が実証された。
【０１５１】
以上、実施例３〜実施例６では添加イオンとしてＴm^３＋、Ｈo^３＋、Ｅr^３＋について説明したが、本発明はこれらのイオンを使用する場合に限定されない。即ち、増幅の始凖位と終凖位の合計イオン数が一定となる条件が成立しないようなタイプの希土類添加光増幅器に本発明を適用することができる。
【０１５２】
（実施例７）
図１６は、本発明の光ファイバ増幅器の第７実施例を示す構成図である。図１６において図１１と同様の個所には同一の参照番号を付してその説明を省略する。図１６に示すように、この光ファイバ増幅器は、図１１の構成に加え、信号光を分波する分波器４１０１、分波器４１０１からの信号光を抽出するためのフィルタ４１０２、フィルタ４１０２からの入力する信号光パワーを検出する光検出器４１０３が含まれる。また、本実施例では、フィルタ３７０３は、分波器３７０２からの信号光を抽出し、光検出器３７０４は、この抽出された信号光パワーを検出する。なお、励起光パワーは４００ｍＷに固定とした。
【０１５３】
分波器４１０１として溶融延伸のファイバ型（分岐型、波長分割多重型いずれも）、誘電体多層膜型、ファイバグレーティングと組み合わせたサーキュレータ等が使用可能である。フィルタ４１０２として誘電体多層膜フィルタ、ファイバグレーティング等がある。
【０１５４】
本実施例では、信号光は１４８０〜１５１０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを６０００ｐｐｍ添加した５ｍの光ファイバ、制御光源３５０４の波長は１４７０ｎｍ、励起波長は１４１０ｎｍ、アイソレータ３３０２、３３０５、３３０８は中心波長１４８０ｎｍ、分波器３５０１、３３０６は分岐比９５：５のファイバカプラ、合波器３３０１は１４７２ｎｍ以下と１４７８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、分波器３７０１、３７０２、４１０１は３ｄＢの分岐カプラ、合波器３３０３、３７０５は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、フィルタ３３１０、３５０２は、中心波長１４８０ｎｍ、透過帯域の半値幅０．８ｎｍのもの、フィルタ４１０２、３７０３は、中心波長１５１０ｎｍ、透過帯域の半値幅が０．８ｎｍのもの、を使用した。なお、モニタする信号光は１４８０ｎｍと１５１０ｎｍの信号光とした。
【０１５５】
ここで、２種類のモニタする信号光の利得をＧ（１４８０ｎｍ）およびＧ（１５１０ｎｍ）とし、合計利得Ｇ_Ｓおよび利得差ΔＧを以下のように定義する。
【０１５６】
Ｇ_Ｓ＝Ｇ（１４８０ｎｍ）＋Ｇ（１５１０ｎｍ） …（１）
ΔＧ＝Ｇ（１４８０ｎｍ）−Ｇ（１５１０ｎｍ） …（２）
【０１５７】
Ｇ_ＳおよびΔＧが一意に決まればＧ（１４８０ｎｍ）およびＧ（１５１０ｎｍ）も決定するので、Ｇ_ＳおよびΔＧを制御することで、Ｇ（１４８０ｎｍ）およびＧ（１５１０ｎｍ）を制御可能である。
【０１５８】
本実施例のＧ_ＳとΔＧにおいても、励起光パワーや制御光パワーの変化に対し、実施例３のＧ_ＳとΔＧと同様の傾向を示すため、同様の制御アルゴリズムを適用することにより利得スペクトル制御が可能であり、実施例３と同様にＧ_ＳとΔＧを算出し、これをもとに制御光源３５０４および励起光源３３０９の光量を調整する。
【０１５９】
本実施例においては、波長多重信号光のうち２チャネル（１４８０ｎｍおよび１５１０ｎｍ）を用いて利得を求め、Ｇ_ＳおよびΔＧを算出して制御光源３５０４および励起光源３３０９の光量を制御したが、さらに多くのチャネルを用いて制御を行う実施例も可能である。この場合、例えば各チャネルの利得の差分としては△Ｇの代りに２種類の任意の２チャネルの利得の差分を用いる方法が考えられるが、２チャネルにおけるＧ_ＳとΔＧのように、各チャネルの利得の制御が可能な範囲で変数を自由に定義し、光量制御のために用いることができる。
【０１６０】
本発明の利得制御時、−２５〜−１０ｄＢｍ／ｃｈ．入力条件に対し、利得偏差０．３ｄＢ以下を実現し、本制御の有効性が実証された。
【０１６１】
（実施例８）
図１７は、本発明の光ファイバ増幅器の第８実施例を示す構成図である。図１７において図１１と同様の個所には同一の参照番号を付してその説明を省略する。図１７に示すように、この光ファイバ増幅器は、図１１の構成に加え、演算部４３１３を有する。演算部４３１３は、光検出器３５０３で検出された信号光パワーを補正する補正回路４３０１および補正された信号光パワーに基づいて演算を行ってその結果を制御部３３１２に送る演算回路４３０２を有する。実施例４同様、各部材、装置は実施例３と同様のものを使用する。
【０１６２】
本実施例では、信号光は１４８０〜１５１０ｎｍ、２ｎｍ間隔の１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを６０００ｐｐｍ添加した５ｍの光ファイバ、制御光源３３０７の波長は１４７０ｎｍ、励起波長は１３９０ｎｍ、アイソレータ３３０２、３３０５は中心波長１４８０ｎｍ、合波器３３０１は１４７２ｎｍ以下と１４７８ｎｍ以上の帯域を合分波する誘電多層膜型の波長分割多重カプラ、分波器３７０１、３７０２は３ｄＢの分岐カプラ、合波器３３０３、３７０５は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラ、分波器３５０１、３３０６は分岐比９５：５の分岐カプラ、フィルタ３５０２、３３１０は、透過帯域が１４８０〜１５１０ｎｍのもの、フィルタ３７０３は、中心波長１４７０ｎｍ、透過帯域の半値幅が０．８ｎｍのもの、を使用した。なお、モニタする信号光は１４７０ｎｍの制御光と予め定めた波長範囲の光量として、１４８０〜１５１０ｎｍの波長範囲の全信号光とした。
【０１６３】
ここで、２種類のモニタする信号光の利得をＧ（１４７０ｎｍ）およびＧ（１４８０〜１５１０ｎｍ）とする。しかし、光検出器３５０３に検出される全信号光パワーには信号光以外の光（増幅された自然放出光等）が含まれ、無視できないほど大きい場合がある。特に多段中継後の場合にはその傾向が顕著に表れる。そこで補正回路４３０１にて入力信号光パワーに応じたある関数から計算される補正値を発生し、その値を演算部で計算されたＧ（１４８０〜１５１０ｎｍ）に加算もしくは掛算することでＧ（１４８０〜１５１０ｎｍ）の補正を行う。補正後のＧ（１４８０〜１５１０ｎｍ）をＧ′（１４８０〜１５１０ｎｍ）とする。合計利得Ｇ_Ｓおよび利得差ΔＧを以下のように定義する。
【０１６４】
Ｇ_Ｓ＝Ｇ（１４７０ｎｍ）＋Ｇ′（１４８０〜１５１０ｎｍ） …（１）
ΔＧ＝Ｇ（１４７０ｎｍ）−Ｇ′（１４８０〜１５１０ｎｍ） …（２）
【０１６５】
Ｇ_ＳおよびΔＧが一意に決まればＧ（１４７０ｎｍ）およびＧ′（１４８０〜１５８０ｎｍ）も決定するので、Ｇ_ＳおよびΔＧを制御することで、Ｇ（１４７０ｎｍ）およびＧ′（１４８０〜１５１０ｎｍ）を制御可能である。
【０１６６】
本実施例のＧ_ＳとΔＧにおいても、励起光パワーや制御光パワーの変化に対し、実施例３のＧ_ＳとΔＧと同様の傾向を示すため、同様の制御アルゴリズムを適用することにより利得スペクトル制御が可能である。また、実施例３と同様、入力された信号光は、分波器３５０１で分波され、分波器３３０６にて分波され、その一方は信号光として出力される。
【０１６７】
分波器３５０１で分岐された信号光は、フィルタ３５０２にて１４８０〜１５１０ｎｍの全信号光が抽出された後、光検出器３５０３にて検出され、制御光は制御光源に内蔵された光検出器３３０７で検出され、分波器３３０６で分波された信号光および制御光は、分波器３７０２でさらに分岐され、一方はフィルタ３７０３にて１４７０ｎｍの制御光が抽出された後、光検出器３７０４で検出され、もう一方はフィルタ３３１０にて１４８０〜１５１０ｎｍの全信号光が抽出された後、光検出器３３１１で検出され、それらの検出値は電気信号に変換後、演算部４３１３の演算回路４３０２に送られる。また、光検出器３５０３にて検出された検出値は補正回路４３０１に送られる。
【０１６８】
光検出器３５０３で検出された検出値は、演算部３３１３内の補正回路４３０１にて以下の式に従って補正される。
【０１６９】
Ｐre＝Ｐc・(1.0029−0.2530・Ｐc−3.0981・Ｐc・Ｐc），
Ｐc≧0.005 …（３）
Ｐre：補正後制御光源出力（ｍＷ）
Ｐc：補正前制御光源出力（ｍＷ）
【０１７０】
本補正式は光検出器３５０３にて検出した制御光パワー値に応じて、その値に重み付けして補正するものであり、補正前後の制御光出力の関係を図１８に示す。横軸を補正前制御光出力、縦軸を補正前制御光出力と補正後制御光出力の差（ΔＰ）とした。補正前制御光出力が増加するにつれΔＰが急激に増加していることが分かる。なお本方程式はあくまでも一例であり、制御時の増幅器の動作条件を満たすのであれば多種多様の関数を取り得る。また、本実施例では、光検出器３５０３からの出力値の補正についてのみ説明したが、他の実施例としては、光検出器３７０４からの出力値を補正することまたは光検出器３５０３および３７０４からの出力値の双方を補正する構成も可能である。
【０１７１】
補正後の制御光源出力値は、演算回路４３０２へ送られる。
【０１７２】
演算回路４３０２では、受信した検出値や補正値から信号光および制御光の利得を算出した上で、Ｇ_ＳおよびΔＧを算出する。予め設定されたＧ_ＳおよびΔＧと一致するよう制御光源および励起光源の電流値を算出し、それらの値は制御部３３１２へ送られる。
【０１７３】
制御部３３１２は、演算回路４３０２から送られてきた電流値になるように制御光源３３０７および励起光源３３０９を調整する。
【０１７４】
本発明の利得制御時、−２５〜−１０ｄＢｍ／ｃｈ．入力条件に対し、利得偏差０．３ｄＢ以下を実現し、本制御の有効性が実証された。
【０１７５】
（実施例９）
図１９は、本発明の光ファイバ増幅器の第９実施例を示す構成図である。図１９において図１７と同様の個所には同一の参照番号を付してその説明を省略する。図１９に示すように、図１７の構成に加え、制御光と信号光を合分波するファイバグレーティング４５０１、４５０５と組み合わせたサーキュレータ４５０２、４５０４、アイソレータ３３０８からの励起光とサーキュレータ４５０２からの信号光・制御光とを合波する合波器４５０３が含まれる。
【０１７６】
ファイバグレーティング４５０１、４５０５と組み合わせたサーキュレータ４５０２、４５０４の代わりに溶融延伸のファイバ型（分岐型、波長分割多重型いずれも）、誘電体多層膜型等が使用可能である。合波器４５０３として、溶融延伸のファイバ型、誘電体多層膜型、ファイバグレーティングと組み合わせたサーキュレータ等が使用可能である。
【０１７７】
本実施例では、信号光は１４７８、１４８０、１４８２、１４８４、１４８６、１４８８、１４９０、１４９２、１４９６、１４９８、１５００、１５０２、１５０４、１５０６、１５０８、１５１０ｎｍの１６波、増幅用光ファイバ３３０４はＺＢＬＡＮフッ化物ガラスをホストとしコアにツリウムイオンを６０００ｐｐｍ添加した７ｍの光ファイバ、制御光源の波長は１４９４ｎｍ、励起波長は１４００ｎｍ、アイソレータ３３０８は中心波長１４００ｎｍ、ファイバグレーティング４５０１、４５０５と組み合わせたサーキュレータ４５０２、４５０４は合分波波長１４９４ｎｍ（帯域幅０.８ｎｍ）、合波器４５０３は１４３０ｎｍ以下と１４４０ｎｍ以上の帯域を合波する誘電多層膜型の波長分割多重カプラを使用した。また励起光パワーは３００ｍＷに固定とした。
【０１７８】
図１９に示すように、信号光は、分波器４５０１にて制御光と合波され、ファイバグレーティング４５０１と組み合わせたサーキュレータ４５０２を通過し、合波器４５０３にて更に励起光と合波され、増幅用光ファイバ３３０４を通過後、ファイバグレーティング４５０５と組み合わせたサーキュレータ４５０４にて分波され、その一方は信号光として出力される。
【０１７９】
増幅用光ファイバ３３０４に入力される制御光パワーは、制御光源３３０７の内部の光検出器にて検出され、増幅用光ファイバより出力される制御光パワーは、ファイバグレーティング４５０５と組み合わせたサーキュレータ４５０４で分波され、光検出器３３１１にて検出され、それらの値は電気信号にて演算部３３１３に送られる。
【０１８０】
制御光源３３０７内部の光検出器に検出された検出値は、演算部３３１３内の補正回路４３０１にて以下の式に従って実施例８と同様に補正され、補正前後の制御光出力の関係は図２０に示される。
【０１８１】
Ｐre＝Ｐc・(1.0106−0.73559・Ｐc），Ｐc≧0.005
Ｐre：補正後制御光源出力（ｍＷ）
Ｐc：補正前制御光源出力（ｍＷ）
【０１８２】
なお、本方程式はあくまでも一例であり、制御時の増幅器の動作条件を満たすのであれば多種多様の関数を取り得る。また、本実施例では、光検出器を内部の制御光源３３０７からの出力値の補正についてのみ説明したが、この他には、光検出器３３１１からの出力値を補正することまたは光検出器３３１１および制御光源３３０７からの出力値の双方を補正する構成も可能である。
【０１８３】
補正後の制御光源出力値は、演算回路４３０２へ送られ、補正値に基づき制御光源３３０７の光量が調整される。
【０１８４】
本発明の利得制御時、−２５〜−１０ｄＢｍ／ｃｈ．入力条件に対し、利得偏差０．３ｄＢ以下を実現し、本制御の有効性を実証した。
【０１８５】
【発明の効果】
本発明によれば、希土類イオンを添加した増幅媒体を用いた光増幅器において、少なくとも光増幅器内に制御光源を有し、制御光もしくは信号光の利得を制御すること、少なくとも光増幅器内に制御光源を有し、制御光利得と信号光利得を制御すること、または、少なくとも光増幅器内に制御光源を有し、２つの信号光の利得を制御することにより利得スペクトルを一定に制御することができる。
【０１８６】
また、本発明の光増幅器は、ＷＤＭ伝送システムに適用した場合、信号チャネル数の変化に伴う入力レベルの変動および温度変化による利得スペクトルの変化に対し、信号光の利得スペクトルの制御が可能である。
【０１８７】
また、本発明の光増幅器は、伝送路損失の経時変化時の補償に必要な出力を一定に制御することができ、さらに、ホールバーニング等にも対応することができる。
【０１８８】
本発明の制御方法は、広く一般的な希土類イオンを添加した増幅媒体を用いた光増幅器で利用可能である。
【図面の簡単な説明】
【図１】（Ａ）および（Ｂ）は、それぞれ、石英をホストガラスとする場合のＥｒイオン、およびＰｒイオンの、増幅に大きく関与するエネルギー準位を表す模式図である。
【図２】（Ａ）および（Ｂ）は、それぞれ、ＥｒイオンおよびＴｍイオンのエネルギー凖位を示す図である。
【図３】ＴＤＦＡにおいて波長λ_１での利得を一定にし、ＴＤＦＡの励起光量を調節した場合の特性を示す図である。
【図４】制御光と信号光の総入力パワーが常に一定になるように制御光源の光量を調節した場合の利得スペクトルを表す図である。
【図５】制御光と信号光の総出力パワーが常に一定になるように制御光源の光量を調節した場合の利得スペクトルを表す図である。
【図６】本発明の光増幅器を説明するための構成図である。
【図７】本発明の光ファイバ増幅器の第１実施例を説明するための構成図である。
【図８】第１実施例において各入力信号レベルを変えた場合の利得プロファイルの偏差の最大値をプロットした図である。
【図９】本発明の光ファイバ増幅器の第２実施例を説明するための構成図である。
【図１０】第２実施例において各入力信号レベルを変えた場合の利得プロファイルの偏差の最大値をプロットした図である。
【図１１】本発明の光ファイバ増幅器の第３実施例を説明するための構成図である。
【図１２】第３実施例において、入力信号光パワーを−１５ｄＢｍ／ｃｈ．、励起光パワーを４００ｍＷに固定し、制御光のパワーを変化させた場合の、２つのモニタする信号光の合計利得Ｇ_Ｓと利得差ΔＧをプロットした図である。
【図１３】第３実施例において、入力信号光パワーを−１５ｄＢｍ／ｃｈ．、制御光パワーを−１５ｄＢｍに固定し、励起光パワーを変化させた時の、２つのモニタする信号光の合計利得Ｇ_Ｓと利得差ΔＧをプロットした図である。
【図１４】本発明の利得制御時、−１０ｄＢｍ／ｃｈ．入力時の利得スペクトルを基準として各入力条件（−２５，−２０，−１５，−１０ｄＢｍ／ｃｈ．）の利得スペクトルの偏差をプロットした図である。
【図１５】第４〜第６実施例を説明するための構成図である。
【図１６】本発明の光ファイバ増幅器の第７実施例を説明するための構成図である。
【図１７】本発明の光ファイバ増幅器の第８実施例を説明するための構成図である。
【図１８】第８実施例において、補正前後の制御光出力の関係を示した図である。
【図１９】本発明の光ファイバ増幅器の第９実施例を説明するための構成図である。
【図２０】第９実施例において、補正前後の制御光出力の関係を示した図である。
【符号の説明】
６００ 光増幅器
６０１、６０３ 監視手段
６０２ 増幅部
６０４ 制御システム
６１０ 制御光源
６１２ 増幅媒体
６１４、６１６ 励起光源
６１８、６２０、６２２ 合波器
６２４、６２６ アイソレータ
６２８、６３６ 信号
６３０ 制御部
６３２ 演算部
６３４ 光分岐素子
６３４ 分波器
６３５ 光検出器
３３０１、３３０３、３７０５、４５０３ 合波器
３３０２、３３０５、３３０８ アイソレータ
３３０３ 波長分割多重カプラ（合波器）
３３０４ 増幅用光ファイバ
３３０６、３５０１、３７０１、３７０２、４１０１、４５０１ 分波器
３３０７、３３１１、３５０３、３７０４、４１０３ 光検出器
３３０７、３３０７、３５０４ 制御光源
３３０９ 励起光源
３３１０、３５０２、３７０３、４１０２ フィルタ
３３１２ 制御部
３３１３、４３１３ 演算部
４３０１ 補正回路
４３０２ 演算回路
４５０１、４５０５ ファイバグレーティング
４５０２、４５０４ サーキュレータ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an optical amplifier. More specifically, the present invention relates to a technique for controlling the gain profile of an optical amplifier including an amplification medium containing rare earth ions.
[0002]
[Prior art]
Wavelength division multiplexing (WDM) communication is one of the most effective means to cope with the rapid increase in optical communication capacity in recent years. The WDM communication is characterized in that the communication capacity is expanded by increasing the number of signal channels. For this purpose, it is indispensable to expand the signal wavelength range.
[0003]
In current WDM communications, rare earth ion doped optical fiber amplifiers are used. As an amplification band used in WDM communication, there is a C band (1530 nm to 1570 nm) which is an amplification band of an erbium-doped optical fiber amplifier (EDFA).
[0004]
On the other hand, there is an S band (1460 nm to 1530 nm) as a band having low loss and low dispersion equivalent to that of the C band, and is attracting attention as a next-generation signal wavelength range. The thulium-doped optical fiber amplifier (TDFA) has an amplification band in the S band, and has been intensively studied.
[0005]
In the S-band TDFA, the power conversion efficiency (40%) equivalent to that of the L-band EDFA has been achieved in the entire optical amplifier, and WDM transmission experiments have also been successful (for example, OFC2001 PD-1 (non-patent) Reference 1)). Furthermore, widening by band combining of an EDFA using the C-band amplification band and an EDFA using the L-band amplification band has been achieved, and wideband transmission of 10.9 Tbit / s has been reported. (For example, refer to OFC2001 PD-24 (Non-Patent Document 2)).
[0006]
By the way, in WDM communication using a plurality of amplifiers using rare-earth-doped optical fibers as amplification media as repeaters, the output spectrum of the amplifier after relay changes, and in some cases, the spectrum is greatly distorted.
[0007]
Such a phenomenon appears because the power of the input signal light to the optical amplifier changes due to the change in the loss of the transmission line over time and the change in the number of signal channels, thereby changing the gain spectrum of the optical amplifier. This is because the signal distortion occurs due to accumulation in the entire transmission path.
[0008]
The phenomenon that the gain spectrum of the amplifier described above is greatly distorted is a factor that limits the transmission distance, and therefore it is necessary to control the amplifier so that the gain spectrum of the amplifier becomes constant. As a control method thereof, it is effective to control each optical amplifier so as to keep the gain profile constant (hereinafter also referred to as constant gain profile control).
[0009]
For example, as an example of a control method for a quartz-based EDFA, there is a method of monitoring the gain of one channel of signal light and controlling the pumping light power so that the gain of the channel is always constant. A method for controlling the gain profile of the quartz-based EDFA to be constant is almost established by this method. Also, the constant control of the gain profile of the fluoride PDFA can be controlled by the same method. The amplification media (optical fibers) of these optical amplifiers operate by an amplification mechanism having substantially two levels involved in amplification. The mechanism is that it can be handled only by the initial level involved in amplification (hereinafter also referred to as amplification start position) and the final level involved in amplification (hereinafter also referred to as amplification end position) or the ground level. This is a rare case.
[0010]
These cases will be described with reference to FIG. FIG. 1A is a schematic diagram of excitation levels of a silica-based EDFA, and FIG. 1B is a schematic diagram of excitation levels of a praseodymium-doped optical fiber amplifier (PDFA).
[0011]
In the quartz-based EDFA, as shown in FIG. 1 (A), the amplification final level and the ground level coincide with each other. ⁴ I _13/2 ) And final amplification level ( ⁴ I _15/2 ). Accordingly, it is possible to control the gain profile to be kept constant by monitoring the gain of the signal light of one channel and adjusting the light amount of the excitation light source so that the gain of the channel is always constant.
[0012]
In the PDFA, as shown in FIG. 1B, there are three levels involved in amplification: an amplification start level, an amplification end level, and a ground level. However, in PDFA, the fluorescence lifetime of the amplification final level is much shorter than that of the amplification start level, so that the amplification final level can be ignored. Therefore, the gain profile can be controlled to be constant in the same manner as the quartz-based EDFA.
[0013]
Thus, when there are substantially two levels involved in amplification, it is relatively easy to keep the gain profile constant.
[0014]
However, in general, in an optical amplifier using an amplification medium to which rare earth ions are added, there are few cases where the level involved in amplification can be regarded as substantially two. In addition, it is impossible to apply a method of monitoring the gain of the wavelength of one signal and adjusting the light amount of the excitation light source from the change in the gain so as to keep the gain profile constant. In other words, the gain profile can be controlled to be constant with respect to fluctuations in the level of the input signal and other conditions (for example, temperature changes) simply by adjusting the light intensity of the excitation light source and controlling the gain of one signal wavelength to be constant. I can't do it.
[0015]
For this reason, in such a system, it is basically necessary to monitor the gain at the wavelengths of two different signals, and complicated control is required.
[0016]
As described above, in an optical amplifier using an amplification medium (an optical fiber or an optical waveguide) to which some rare earth ions are added, the gain at one signal wavelength is monitored, and the excitation light is controlled. Although the gain profile can be controlled to be constant, the gain profile cannot be controlled to be constant in an optical amplifier using an amplification medium to which other rare earth ions are added.
[0017]
Separately from the method of monitoring the gain of the wavelength of one signal as described above and adjusting the light amount of the excitation light source based on the change, the gain profile is kept constant using the control light. A method of keeping is also proposed.
[0018]
For example, E. Desurvire et al., “Dynamic gain compensation in saturated erbium-doped fiber amplifiers”, IEEE, Photonics technology letters, vol. , It is disclosed that the power of the control light is controlled so that the ASE level becomes constant.
[0019]
Japanese Patent Laid-Open No. 2000-261079 (Patent Document 1) inputs control light from behind a rare earth-doped optical fiber, monitors the power of control light output from the optical fiber, and keeps this power constant. An optical amplifier and a control method for controlling the light source of the control light are disclosed.
[0020]
JP-A-8-304856 (Patent Document 2) monitors the excitation light input to the amplification medium and the excitation light output from the amplification medium, and enters the optical amplifier from the ratio of these values. It is disclosed that the power of the control light to be controlled is controlled.
[0021]
In addition, the total input power of the control light and the signal light is monitored, the control light power is controlled so that the power is constant, and the total output power of the control light and the signal light is monitored. It is also disclosed to control the power of the control light so as to be constant.
[0022]
However, the method using control light as described above is a quartz-based EDFA in which there are only two levels (that is, levels to be considered in determining the state of the gain spectrum) that are greatly involved in amplification. However, TDFA and other rare-earth doped optical fiber amplifiers cannot be controlled.
[0023]
[Patent Document 1]
JP 2000-261079 A
[0024]
[Patent Document 2]
JP-A-8-304856
[0025]
[Non-Patent Document 1]
OFC2001 PD-1
[0026]
[Non-Patent Document 2]
OFC2001 PD-24
[0027]
[Non-Patent Document 3]
E. Desurvire et al., "Dynamic gain compensation in saturated erbium -doped fiber amplifiers"., IEEE, Photonics technology letters, vol. May 1991
[0028]
[Non-Patent Document 4]
E. Desurvire, "Erbium-Doped Fiber Amplifiers," A. Wiley-Interscience Publication, Chapter 1, 1994
[0029]
[Non-Patent Document 5]
Won Jae Lee et al., "Gain excursion & tilt compensation algorithm for TDFA using 1.4 μm / 1.5μm dual wavelength control"., OFC2002, ThZ3
[0030]
[Problems to be solved by the invention]
Therefore, there is a demand for an optical amplifier and a control method capable of controlling the gain profile to be constant, which can be applied to an optical amplifier using a general rare earth-doped amplification medium in a simple manner.
[0031]
The present invention has been made in view of such problems, and its object is to easily control a gain profile to be kept constant in an optical amplifier using an amplification medium to which a general rare earth ion is added. An object of the present invention is to provide an optical amplifier that can be used.
[0032]
It is another object of the present invention to provide a control method for controlling a gain profile to be constant using an optical amplifier using an amplification medium to which a general rare earth ion is added.
[0033]
[Means for Solving the Problems]
A first aspect of the present invention relates to an optical amplifier that can be controlled so as to keep a gain profile constant in an optical amplifier using an amplification medium to which a rare earth is added. In particular, the optical amplifier of the present invention relates to an optical amplifier using an amplification medium to which one or more general rare earth ions are added.
[0034]
According to the present invention, the amplifier of the present invention includes a rare earth-doped optical fiber or a rare earth-doped optical waveguide in which rare earth ions are added to a core and / or a clad. Amplifying medium comprising Exciting the amplification medium Excitation means whose excitation light power is constant One or more control light sources that can control the inversion distribution state of the amplification medium by inputting control light having a wavelength within the amplification band of the amplification medium to the amplification medium and changing the input power of the control light. Input and output Control light power, signal light power of one or more channels, light power in a predetermined wavelength range within the amplification band, or a combination thereof And monitoring means for monitoring the value obtained from the monitoring means Power of control light input / output to / from the amplification medium, power of signal light of one or more channels, power of light in a predetermined wavelength range within the amplification band, or a combination thereof Calculate the gain of the predetermined Gain Value or input from outside Gain Control the control light source to match the value To keep the gain profile constant And a control unit.
[0035]
The second aspect of the present invention is a method for controlling the gain profile of an optical amplifier to be constant.
[0036]
The control method of the present invention includes a rare earth-doped optical fiber or a rare earth-doped optical waveguide in which rare earth ions are added to the core and / or clad. Amplifying medium comprising Exciting the amplification medium Excitation means whose excitation light power is constant An optical amplifier including one or more control light sources that can control the inversion distribution state of the amplification medium by changing the input power of the control light from the control light source, a monitoring unit, and a control unit How to control the gain profile constant Because Control light having a wavelength within the amplification band of the amplification medium is input to the amplification medium and input / output to the amplification medium Control light power, signal light power of one or more channels, light power in a predetermined wavelength range within the amplification band, or a combination thereof Monitoring the value obtained from the monitoring means Control light power, signal light power of one or more channels, light power in a predetermined wavelength range within the amplification band, or a combination thereof Calculate the gain of the predetermined Gain Value or input from outside Gain Controlling the control light source to match the value Constant gain profile control Including.
[0037]
DETAILED DESCRIPTION OF THE INVENTION
The first aspect of the present invention relates to an optical amplifier that can be controlled so as to keep a gain profile constant in an optical amplifier using an amplification medium to which a rare earth is added.
[0038]
The second aspect of the present invention relates to a control method for maintaining a constant gain profile of an optical amplifier using an amplification medium to which a rare earth is added.
[0039]
The principle of the present invention will be described below.
[0040]
First, a description will be given of a method for controlling the optical amplifier and the gain profile to be constant in the case of using a rare earth-added amplification medium in which the levels involved in amplification are only two levels of the final amplification level and the initial amplification level. To do. Some amplifiers use a quartz-based erbium-doped optical fiber (EDF) as an amplification medium.
[0041]
In an optical amplifier using a quartz-based EDF as an amplification medium, the level involved in amplification is the final amplification level ( ⁴ I _15/2 ) And amplification start position ( ⁴ I _13/2 It is a relatively simple system.
[0042]
Such an amplifier is, for example, an optical amplifier (quartz EDFA) using a quartz EDF as an amplification medium, a quartz EDF, excitation means for exciting the quartz EDF, monitor means for monitoring input signal light, and the like. And at least a control unit for controlling the excitation means based on information from the monitoring means.
[0043]
FIG. 2A shows an Er of an optical amplifier using a quartz EDF as an amplification medium. ³⁺ It is a figure which shows the energy level of ion. In this optical amplifier, the final position of amplification ( ⁴ I _15/2 ) From the beginning of amplification ( ⁴ I _13/2 Amplification is performed using a method of excitation to An example of a method for controlling the gain profile of a quartz-based EDFA is to monitor the gain of signal light of one channel (hereinafter also referred to as “ch.”) And adjust the pump light power so that the gain of the channel is always constant. It is something to control.
[0044]
The principle of this control will be described below.
[0045]
The gain Gain (λ, x) per unit length at the position x in the longitudinal direction of the erbium-doped fiber amplifier at the wavelength λ is expressed by the following equation (for example, E. Desurvire, “Erbium-Doped Fiber Amplifiers, "See A. Wiley-Interscience Publication, Chapter 1, 1994 (Non-Patent Document 4)).
[0046]
Gain (λ, x) = (σe (λ) N2 (x) − (σa (λ) N1 (x)) (eq1)
[0047]
[Expression 1]

[0048]
σe: stimulated emission cross section
σa: induced absorption cross section
N2 (x): starting position of amplification at position x ( ⁴ I _13/2 ) Number of ions
N1 (x): final position of amplification at position x ( ⁴ I _15/2 ) Number of ions
Ntotal: total number of ions
[0049]
Here, from the equations (eq1) and (eq2),
Gain (λ, x) = (σe (λ) N2 (x) − (σa (λ) (Ntotal−N2 (x)) (eq3)
Get.
[0050]
As can be seen from the equation (eq3), the only variable is N2.
[0051]
Here, since the stimulated emission cross section and stimulated absorption cross section are determined by the physical properties of the rare earth ions added to the amplification medium, N2 is fixed to fix the gain spectrum, that is, the amplification start level. It is necessary to fix the number of ions at the end of amplification.
[0052]
In WDM transmission or the like, the power of input signal light to the optical amplifier changes due to a change in transmission path loss with time or a change in the number of signal channels. Such a change is caused by the quartz-based EDFA. ⁴ I _13/2 Er present in ³⁺ As a result, N2 changes, and the gain spectrum changes as shown in (eq3).
[0053]
Therefore, when there are two levels involved in amplification, such as an optical amplifier using a quartz-based EDF as an amplification medium, an arbitrary signal light wavelength is monitored and the gain at this signal wavelength is kept constant. If the pumping light power is controlled in this way, the gain spectrum is kept constant based on (eq3) regardless of the input power to the optical amplifier.
[0054]
The specific procedure for controlling the gain to be constant is to demultiplex the signal light at the input end and output end of the amplification optical fiber, detect the signal light power at each location, and detect the signal light from the detected value. And the amount of excitation light is adjusted so that the error component between the calculated gain value and the set value becomes zero. This is the principle for controlling the gain profile constant with an optical amplifier using a quartz-based EDF as an amplification optical fiber.
[0055]
By the way, a system in which the amplification mechanism can be described only with ions in the amplification starting position and the terminal position is very rare. For example, even in the case where the same stimulated emission as described above is used in an erbium-doped optical fiber, even if the host glass constituting the optical fiber is changed to a fluoride glass, the excitation position is higher than the starting position of amplification. And the lifetime at the transition position after the transition cannot be ignored. In such a case, even if the gain at a certain wavelength (for example, one wavelength of the input signal light) is controlled to be constant, the gain profile cannot be uniquely determined. This is because, in describing the amplification operation, it is necessary to consider the amplification start position, the amplification end position, and the number of other ions. That is, for example, in TDFA, the total number of ions is expressed by the following (eq2 ′), and it is not sufficient to consider only one parameter (N2) as in (eq3) above, and the number of ions of other levels is This is because it must be considered.
[0056]
[Expression 2]

[0057]
An optical amplifier using a rare earth-doped optical fiber in which three levels are greatly involved in amplification includes an optical amplifier using a thulium-doped optical fiber (TDF). As shown in FIG. 2B, this optical amplifier is a four-level optical amplifier, but there are three levels greatly involved in amplification. That is, in TDFA, the energy positions greatly involved in amplification include an amplification starting position, an end position, and a base position. Therefore, it is necessary to consider all of these levels. Therefore, the number of ions between the three saddle positions can be determined by simply monitoring the gain of one wavelength as in the EDFA discussion and controlling it to be constant. Cannot be kept constant.
[0058]
FIG. 3 shows a wavelength λ incident on an optical amplifier in TDFA when a WDM signal is amplified. ₁ And the wavelength λ amplified and emitted from the TDFA ₁ It is a figure which shows the characteristic of an optical amplifier at the time of adjusting the excitation light quantity of TDFA so that the gain calculated from the power of the signal light of this may be fixed.
[0059]
In FIG. 3, a to c show the characteristics of the optical amplifier, where a is a case where the total input signal power to the TDFA is low, b is an intermediate value, and c is a high value. The condition for evaluating this characteristic is Tm of TDFA. ³⁺ The addition concentration was 6000 ppm and the excitation wavelength was 1400 nm. Excitation of the Tm-doped fluoride optical fiber for amplification is bidirectional excitation in which the optical fiber is excited from the front and rear directions. The ratio between the forward pumping light power and the backward pumping power was always constant. As is apparent from FIG. 3, the gain profile changes with respect to the change in the input signal light power simply by making the gain of one wavelength constant.
[0060]
The following control method has been proposed as a method for controlling an optical amplifier including an amplification medium having three levels greatly involved in amplification. This control method relates to a TDFA optical amplifier, and there are two methods. One of the control methods is used in a bidirectionally pumped TDFA optical amplifier. In this control method, the gains of two types of signal light are monitored. Based on this monitor information, the gain band is adjusted by controlling the ratio of the power of the front and rear pumping light of the amplification optical fiber, and the overall gain is adjusted by controlling the total pumping light power. Is the method. In TDFA, when the power ratio between the front and rear pumping light is changed, a phenomenon occurs in which the gain spectrum shifts in wavelength.
[0061]
As a second control method of TDFA, there is a method of adjusting the gain spectrum by controlling the light quantity of the excitation light source of two wavelengths (1.4 μm, 1.56 μm) (Won Jae Lee et al., “Gain excursion”). & tilt compensation algorithm for TDFA using 1.4 μm / 1.5 μm dual wavelength control "., OFC2002, ThZ3 (Non-patent Document 5)).
[0062]
In the case of an optical amplifier using an amplification medium to which a rare earth ion is added, except for an optical amplifier including an amplification medium having two levels substantially involved in amplification as described above, the light source of the excitation light source can be adjusted by adjusting the light quantity. Only by controlling the wavelength gain to a constant value, the gain profile cannot be controlled to be constant with respect to fluctuations in the level of the input signal and changes in other conditions (for example, temperature changes). For this reason, in such a system, it is basically necessary to monitor gains at two types of signal wavelengths, and complicated control is required.
[0063]
In the optical amplifier and the control method of the present invention, it is possible to use an amplification medium to which a rare earth is added that needs to consider the amplification start position, the amplification end position, and other positions. In the following, the case where the other levels are only the third level will be described, but the present invention is not limited to this, and includes cases where more levels are involved.
[0064]
The present invention is characterized in that the inversion distribution state of the amplification medium is made constant by adjusting the amount of the control light according to the fluctuation of the level of the input signal. That is, the basic concept of the present invention is to compensate the fluctuation of the level of the input signal with the control light having the wavelength within the amplification band. According to the present invention, the gain profile constant control is realized by adjusting the control light so as to make the gain of one wavelength in the amplification band constant for all rare earth ion optical amplifiers. In the present invention, in order to adjust the control light source, the gain at a certain wavelength (or a certain wavelength range) of light is monitored. In this way, the control light or at least one signal light is monitored and the control light source is adjusted. For example, the total output or total input power of the control light and the signal light is monitored to control the control light power. This is because the gain spectrum is distorted and the gain profile cannot be controlled to be constant. This will be described with reference to FIGS. 4 and 5 by taking as an example a case where the total output or total input power of the control light and signal light is monitored to control the power of the control light.
[0065]
FIG. 4 is a diagram showing that the gain profile cannot be controlled to be constant only by keeping the total input power constant. FIG. 4 shows the control of fluctuations in the total signal light power input to the optical amplifier when a WDM signal is amplified in a TDFA equipped with a control light source having a wavelength in the amplification band other than the signal wavelength on the input side. It is a figure showing the gain spectrum at the time of adjusting the light quantity of a control light source so that the sum of light input power and total input signal light power (hereinafter also referred to as total input power) is always constant.
[0066]
4 are gain spectra when the total input signal power to the TDFA is a, the middle case is b, and the high case is c. The condition for evaluating this characteristic is Tm of TDFA. ³⁺ The addition concentration was 4000 ppm and the excitation wavelength was 1410 nm. Excitation of the Tm-doped fluoride optical fiber for amplification is bidirectional excitation in which the optical fiber is excited from the front and rear directions, and the total excitation light power is always constant. As is apparent from FIG. 4, the gain profile changes with respect to the change in the input signal light power only by keeping the total input power to the optical amplifier constant.
[0067]
FIG. 5 is a diagram showing that the gain profile cannot be controlled to be constant only by keeping the total output power constant. FIG. 5 shows that when a WDM signal is amplified in a TDFA provided with a control light source on the input side having a wavelength in an amplification band other than the length, the fluctuation of the total signal light power input to the optical amplifier is It is a figure showing the gain spectrum at the time of adjusting the light quantity of a control light source so that the sum of output power of control light and total output signal light power (hereinafter also referred to as total output power) is always constant.
[0068]
5A to 5C are gain spectra when the total input signal power to the TDFA is a, the intermediate case is b, and the high case is c. The condition for evaluating this characteristic is Tm of TDFA. ³⁺ The addition concentration was 5000 ppm and the excitation wavelength was 1420 nm. Excitation of the Tm-doped fluoride optical fiber for amplification is bidirectional excitation in which the optical fiber is excited from the front and rear directions, and the total excitation light power is always constant. As is clear from FIG. 5, the gain profile changes with respect to the change in the input signal light power only by keeping the total output power to the optical amplifier constant.
[0069]
Thus, the gain profile cannot be controlled to be constant only by making the total input power or the total output power constant.
[0070]
In the present invention, the optical amplifier is configured to input the amplification medium, excitation means for exciting the amplification medium, control light having a wavelength within the amplification band of the amplification medium to the amplification medium, and change its input power. The light gain is calculated from one or more control light sources capable of controlling the inversion distribution state of the amplification medium, monitoring means for monitoring the optical power input to and output from the amplification medium, and a value obtained from the monitoring means. A control unit for controlling the control light source so as to coincide with a predetermined value or a value input from the outside.
[0071]
The optical amplifier according to the present invention further includes means for adjusting the amount of control light from the control light source. This means includes adjusting the current value of the control light source, adjusting the amount of attenuation by providing a variable optical attenuator at the output end of the control light source, and the like.
[0072]
In the present invention, the amplification medium, excitation means for exciting the amplification medium, control light having a wavelength within the amplification band of the amplification medium is input to the amplification medium, and the amplification power is changed by changing the input power. One or more control light sources capable of controlling the inversion distribution state of the medium, monitoring means for monitoring the power of light of a plurality of wavelengths inputted to and outputted from the amplification medium, and calculating gains of a plurality of wavelengths from the values obtained from the monitoring means Thereafter, a difference and a sum of a plurality of gains are calculated, the control light source is controlled so that the difference between the plurality of gains matches a predetermined value or a value input from the outside, and the sum of the plurality of gains is calculated in advance. And an optical amplifier including a control unit for controlling the pumping means so as to coincide with a predetermined value or a value inputted from the outside.
[0073]
In the present invention, the monitoring means can be one that monitors control light, one that monitors signal light of one or more channels, or one that monitors light in a predetermined wavelength range within the amplification band.
[0074]
Further, according to the present invention, it is preferable to include correction means for correcting the values obtained by the monitoring means or the gain calculated by the control unit, the gain difference, or the gain sum.
[0075]
In the present invention, the monitoring means preferably includes an optical bandpass filter.
[0076]
According to the principle of constant gain profile control of the present invention, the control method of the present invention is also provided.
[0077]
One of the control methods is a rare earth-doped optical fiber or rare earth-doped optical waveguide in which rare earth ions are added to the core and / or clad as an amplification medium, excitation means for exciting the amplification medium, and one or more control light sources. A method of controlling the gain profile of an optical amplifier including a monitoring unit and a control unit that can control the inversion distribution state of the amplification medium by changing the input power of the control light from the control light source. The control light having a wavelength within the amplification band of the amplification medium is input to the amplification medium, the optical power input to and output from the amplification medium is monitored, and the value obtained from the monitoring unit And calculating the gain of the light and controlling the control light source so as to coincide with a predetermined value or a value input from the outside.
[0078]
In the present invention, the rare earth ions added to the amplification medium may be one kind or plural kinds.
[0079]
In the following, the optical amplifier and the method of constant gain profile control according to the present invention will be described more specifically with reference to the drawings. However, these are merely examples, and the present invention is not limited to these.
[0080]
The present invention relates to an optical amplifier configured to use control light for constant gain profile control of an optical amplifier.
[0081]
Control light is introduced into the optical amplifier, and control is performed using this control light so as to keep the gain profile of the optical amplifier constant. In the present invention, in order to adjust the control light source, the gain at a certain wavelength (or a certain wavelength range) of light is monitored.
[0082]
The outline of the present invention will be described with reference to FIG.
[0083]
The optical amplifier 600 of the present invention includes monitoring means 601 and 603, an amplification unit 602, and a control system 604. The monitoring unit 601 is constituted by a photodetector provided in the control light source 610. The monitoring unit 603 includes an optical branching element 634 and a photodetector 635 such as a photoelectric conversion element. Although the monitoring means 601 shown in FIG. 6 has shown the example which contains a photodetector in the control light source 610, this invention is not limited to this, You may provide a photodetector separately. In this case, an optical branching element (demultiplexer) may be provided to demultiplex a part of the signal light input. The signal from the photodetector is input to the control system 604. The photodetector includes, for example, a photoelectric conversion element. The amplifying unit 602 includes a control light source 610, an amplification medium 612, excitation light sources 614 and 616, a multiplexer 618 for multiplexing light from the control light source, and a multiplexer 620 for multiplexing light from the excitation light source. 622. The amplifying unit is preferably provided with isolators 624 and 626 as shown in FIG.
[0084]
The control light source 610 may or may not have a photodetector inside. In the present invention, a control light source having a photodetector inside is preferable. A control light source having a photodetector inside can output a first signal 628 for monitoring. A plurality of control light sources may exist.
[0085]
The control system 604 includes a control unit 630 and a calculation unit 632 for controlling the control light source. A second signal 636 from an optical branching element (demultiplexer) 634 that extracts part of the light output from the amplifying unit 602 is input to the arithmetic unit. In the calculation unit 632, the gain of the control light is calculated from the first signal from the control light source 610 and the second signal from the optical branching element 634, and the current of the control light source 610 that matches the preset value is calculated. Calculate the value. The value is sent to the control unit 630. The control unit 630 adjusts the light amount of the control light source 610 so that the current value sent from the calculation unit 632 is obtained. In the present specification, the control system 604 is also referred to as a control unit. In this case, the control unit includes a control unit 630 and a calculation unit 632 having the above functions.
[0086]
In the present invention, each component can take various arrangements in addition to the arrangement shown in FIG. For example, in the optical amplifier shown in FIG. 6, the pumping light is bidirectional pumping, but the present invention is not limited to this and may be forward pumping and backward pumping.
[0087]
As the amplification medium, a rare earth-doped optical fiber in which rare earth element ions are added to the core and / or the clad can be suitably used. In particular, in the present invention, a thulium-doped optical fiber, an erbium-doped optical fiber, a holmium-doped optical fiber, or the like can be used. Specifically, an amplification optical fiber to which rare earth ions are added can be used as the amplification medium. This is because the host glass is quartz glass, bismuth glass, ZBLAN of fluoride glass that is less likely to cause non-radiative transition. Glass, In-Pb glass, tellurite glass, or the like. The rare earth ion may be any ion, but thulium ion (Tm ³⁺ ), Holmium ion (Ho ³⁺ ), Erbium ion (Er ³⁺ ) Is preferred.
[0088]
The multiplexer and the optical branching element (demultiplexer) are a melt-drawn fiber type (both branch type and wavelength division multiplexing type), a dielectric multilayer film type, or a circulator combined with a fiber grating. In addition, an ordinary multiplexer such as a pumping light / signal light multiplexing coupler can be used as the multiplexer.
[0089]
Examples of the control light source that generates control light and the excitation light source that generates excitation light include solid-state lasers such as Nd-YLF laser, Nd-YAG laser, and Ti sapphire laser, semiconductor lasers, and fiber lasers. In the optical amplifier shown in FIG. 6, the pumping light is bidirectional pumping, but the present invention is not limited to this, and forward pumping and backward pumping may be used.
[0090]
The control method using the optical amplifier of the present invention (operation of the optical amplifier of the present invention) will be specifically described below. In the following, a case where the photodetector of the monitoring unit 601 is included in the control light source and the gain of the control light is monitored will be described as an example.
[0091]
The signal light is combined with the control light from the control light source 610 by the multiplexer 618. Next, the combined signal light passes through the isolator 624, and is combined with the excitation light from the excitation light source 614 by the multiplexer 620. The signal light combined with the excitation light enters the amplification medium 612. The amplification medium is excited in both directions by the excitation light source 614 and the excitation light source 616. The signal light exiting the amplification medium passes through the multiplexer 622 and the isolator 626 and is demultiplexed by the demultiplexer 634. A part of the signal light not demultiplexed by the demultiplexer 634 is output as signal output light.
[0092]
The control light input from the control light source to the amplification optical fiber 612 is detected by a photodetector inside the control light source 610, and the power of the control light is sent to the arithmetic unit 632 as a first signal. The control light output from the amplification medium is demultiplexed by the demultiplexer 634, and the power of the control light is sent to the calculation unit 632 as the second signal 636. The power of the control light at the input and output ends is preferably converted into an electrical signal and input to the calculation unit 632.
[0093]
The calculation unit 632 calculates the gain of the control light from the received detection value, calculates the current value of the control light source 610 that matches a preset value, and sends the value to the control unit 630. The control unit 630 adjusts the light amount of the control light source 610 so that the current value sent from the calculation unit 632 is obtained.
[0094]
Since the wavelength of the control light is within the amplification band, the gain profile can be controlled to be constant by adjusting the control light so that the gain of the control light is constant.
[0095]
In the present invention, the signal input to the arithmetic unit is not only based on the control light, but also based on signal light of one channel or more, or based on light in a predetermined wavelength range within the amplification band. There may be. In the control method of the present invention, the correction means corrects the values, the gain calculated by the control unit, the gain difference, or the sum of the gains according to the values obtained from the monitoring means. Is preferred.
[0096]
In the present invention, it is preferable to include an optical bandpass filter in the monitoring means and to pass the monitor signal therethrough.
[0097]
In the optical amplifiers of the inventions described above, the propagation direction of the signal light in the amplification medium is not particularly defined. However, the optical amplifier has the following configurations (a) to (c), and the signal light Can take the propagation directions shown in these.
[0098]
(A) An optical isolator is connected to both or one of the input side and output side of the amplification medium. The signal light propagates only in one direction in the amplification medium (referred to as a single path in this specification).
[0099]
(B) An optical circulator is connected to one of the amplification media, and a device having at least a mirror function for reflecting signal light is connected to the other end of the amplification media. The signal light passes through the optical circulator, travels through the amplification medium, and is amplified. Next, the signal light is folded back by a mirror that reflects the signal light, travels in the opposite direction in the amplification medium, and is amplified. The amplified signal light passes through the circulator and exits (referred to herein as a double pass).
[0100]
(C) An optical circulator is connected to both of the amplification media. Two types of signal lights having different traveling directions travel in opposite directions in the amplification medium in a state where they are separated simultaneously or temporally and are amplified (hereinafter referred to as a bidirectional optical amplifier).
[0101]
(Example)
Embodiments of the present invention will be described below with reference to the drawings.
[0102]
Example 1
FIG. 7 is a block diagram showing the present embodiment of the optical fiber amplifier of the present invention. As shown in FIG. 7, this optical fiber amplifier has an amplification medium, for example, an amplification optical fiber 3304 in which rare earth ions are added to a core, isolators 3302, 3305, and 3308, and a multiplexer 3301 that combines control light and signal light. , A multiplexer 3303 that combines the excitation light supplied from the excitation light source 3309 via the isolator 3308 and the signal light / control light from the isolator 3302, and an output that is extracted from the amplification optical fiber 3304 via the isolator 3305 From a demultiplexer 3306 for branching light, a filter 3310 for extracting control light from one output of the demultiplexer 3306, a control light source 3307 having a photodetector inside, an excitation light source 3309 for generating excitation light, and a filter 3310 Signals from the light detector 3311 and the control light source 3307 for detecting the output control light and the light detection. Calculating unit 3313 is supplied with a signal from the vessel 3311, and composed of a control unit 3312 for controlling the control light source 3307 by the output from the arithmetic unit 3313.
[0103]
As a host glass of an optical fiber for amplification to which rare earth ions are added, there are quartz glass, bismuth glass, fluoride glass ZBLAN glass, In-Pb glass, tellurite glass and the like that hardly cause non-radiative transition.
[0104]
As the multiplexer 3301 and the demultiplexer 3306, a melt-drawn fiber type (both branch type and wavelength division multiplexing type), a dielectric multilayer film type, or a circulator combined with a fiber grating can be used. As the multiplexer 3303, a melt-drawn fiber type, a dielectric multilayer type, or a circulator combined with a fiber grating can be used. As the control light source 3307 for generating control light and the excitation light source 3309 for generating excitation light, a solid-state laser such as an Nd-YLF laser, an Nd-YAG laser, or a Ti sapphire laser, a semiconductor laser, a fiber laser, or the like can be used. Examples of the filter 3310 include a dielectric multilayer filter and a fiber grating.
[0105]
In this embodiment, the signal light is 1480-1510 nm, 16 waves at 2 nm intervals, the amplification optical fiber 3304 is a 5 m optical fiber in which ZBLAN fluoride glass is used as a host and 6000 ppm of thulium ions are added to the core, and the wavelength of the control light source is 1520 nm. The excitation wavelength is 1400 nm, the isolators 3302 and 3305 are center wavelengths of 1480 nm, the multiplexer 3301 and the demultiplexer 3306 are dielectric multilayer type wavelength division multiplexing couplers and multiplexers that multiplex and demultiplex bands of 1512 nm or less and 1518 nm or more 3303 is a dielectric multilayer film type wavelength division multiplex coupler for combining bands of 1430 nm or less and 1440 nm or more, and a filter 3310 having a center wavelength of 1520 nm and a transmission band having a half-value width of 0.8 nm. The excitation light power was fixed at 300 mW.
[0106]
As shown in FIG. 7, the signal light is combined with the control light from the control light source 3307 in the multiplexer 3301, passes through the isolator 3302, and is further combined with the excitation light from the isolator 3308 in the multiplexer 3303. Then, after passing through the amplification optical fiber 3304, it passes through the isolator 3305 and is demultiplexed by the demultiplexer 3306, and one of them is output as signal light.
[0107]
The input control light source power to the amplification optical fiber 3304 is detected by a photodetector inside the control light source 3307. The output control light power from the amplification optical fiber 3304 is demultiplexed by the demultiplexer 3306, passes through the filter 3310, is detected by the photodetector 3311, and the value is sent as an electric signal to the arithmetic unit 3313.
[0108]
The calculation unit 3313 calculates the gain of the control light from the received detection value, calculates the current value of the control light source 3307 that matches a preset value, and sends the value to the control unit 3312.
[0109]
The control unit 3312 adjusts the light amount of the control light source 3307 so that the current value transmitted from the calculation unit 3313 is obtained.
[0110]
Since the wavelength of the control light is within the amplification band, the gain profile constant control can be realized by adjusting the control light so that the gain of the control light is constant.
[0111]
FIG. 8 is a graph plotting the maximum value of the gain profile deviation when each input signal level is changed in the first embodiment. That is, each input signal level is set to −25, −20, −15, −10, −5 dBm / ch. -25 dBm / ch. With respect to the gain spectrum when set and input. 6 is a diagram in which the maximum value of the difference between the gain spectrum and the reference spectrum when the input is changed is plotted on the basis of the gain spectrum in FIG. The gain was measured by scanning a small signal probe light of −35 dBm. The dotted line in FIG. 8 shows the result of the conventional example in which gain control by the signal light to be monitored is not applied. The solid line in FIG. 8 is a result of plotting values in the case of the present invention in which the gain at 1520 nm is controlled to be constant using a monitor system for 1520 nm control light. As a result, it has been clarified that the gain profile can be controlled according to the present invention.
[0112]
(Example 2)
FIG. 9 is a block diagram showing a second embodiment of the optical fiber amplifier of the present invention. 9, the same parts as those in FIG. 7 are denoted by the same reference numerals, and the description thereof is omitted. As shown in FIG. 9, in addition to the configuration of FIG. 7, this optical fiber amplifier includes a demultiplexer 3501 for demultiplexing signal light input to and output from the amplifier, and a filter for extracting signal light from the demultiplexer 3501. 3502, a control light source 3504, and a photodetector 3503 that detects signal light power input from the filter 3502. In the present embodiment, the demultiplexer 3306, the filter 3310, and the photodetector 3311 demultiplex and extract the signal light to be output, and detect the power.
[0113]
As the duplexer 3501, a melt-drawn fiber type (both branch type and wavelength division multiplexing type), a dielectric multilayer type, and a circulator combined with a fiber grating can be used. As a control light source 3504 for generating control light, a solid-state laser such as an Nd-YLF laser, an Nd-YAG laser, or a Ti sapphire laser, a semiconductor laser, a fiber laser, or the like can be used. Examples of the filter 3502 include a dielectric multilayer filter and a fiber grating.
[0114]
In this embodiment, the signal light is 1470 to 1500 nm, 16 waves at intervals of 2 nm, the amplification optical fiber 3304 is a 10 m optical fiber in which 4000 ppm of thulium ions is added to the core of ZBLAN fluoride glass, and the wavelength of the control light source 3304 is 1460 nm, excitation wavelength is 1400 nm, isolators 3302 and 3305 are center wavelengths of 1480 nm, demultiplexers 3501 and 3306 are file couplers with a branching ratio of 95: 5, and a multiplexer 3301 demultiplexes bands of 1462 nm or less and 1468 nm or more Wavelength division multi-layer wavelength division multiplex coupler, multiplexer 3301 is a dielectric multi-layer wavelength division multiplex coupler that multiplexes bands below 1430 nm and 1440 nm, filters 3502 and 3310 have a center wavelength of 1510 nm and a transmission band Using a half-value width of 0.8 nm. It was. The excitation light power was fixed at 300 mW. One channel of the wavelength multiplexed signal light to be monitored is assumed to be 1510 nm input light.
[0115]
As shown in FIG. 9, the signal light is demultiplexed by the demultiplexer 3501, and the signal light from the demultiplexer 3501 is combined with the control light from the control light source 3504 by the multiplexer 3301, and isolators 3302. , And is further combined with the excitation light from the isolator 3308 by the wavelength division multiplex coupler 3303, passes through the amplification optical fiber 3304, passes through the isolator 3305, and is demultiplexed by the demultiplexer 3306, Is output as signal light.
[0116]
The input signal light power of 1510 nm is branched by a demultiplexer 3501, passes through a filter 3502, is detected by a photodetector 3503, and the output signal light power of 1510 nm is demultiplexed by a demultiplexer 3306 and is filtered 3310. Are detected by the photodetector 3311 and their values are sent to the calculation unit 3313 as an electrical signal.
[0117]
The calculation unit 3313 calculates the gain of the 1510 nm signal light from the received detection value, calculates the current value of the control light source 3504 so as to match a preset value, and sends the value to the control unit 3312. The control unit 3312 adjusts the light amount of the control light source 3504 so that the current value transmitted from the calculation unit 3313 is obtained. Here, the light quantity of the control light source 3504 is controlled based on the detected value of the signal light power of one channel of 1510 nm, but other embodiments in which control is performed using two or more channels are also possible. Further, an embodiment in which the control is performed using the entire signal light amount in a predetermined wavelength range as the signal light to be monitored is also possible. In this case, in this embodiment, for example, the total amount of the signal light power for each channel in the range of 1470 to 1500 nm is calculated, and the gain is calculated based on the total amount. Can be adjusted.
[0118]
FIG. 10 is a graph plotting the maximum value of the gain profile deviation when each input signal level is changed. That is, each input signal level is set to −25, −20, −15, −10, −5 dBm / ch. -25 dBm / ch. With respect to the gain spectrum when set and input. 6 is a diagram in which the maximum value of the difference between the gain spectrum and the reference spectrum when the input is changed is plotted on the basis of the gain spectrum in FIG. Evaluation of gain was measured by scanning a small signal probe light of −35 dBm. The dotted line in FIG. 10 shows the result of the conventional example in which gain control by the signal light to be monitored is not applied. The solid line in FIG. 10 is a result of plotting values in the case of the present invention in which the gain at 1510 nm is controlled to be constant using a 1510 nm signal light monitor system. As a result, it has been clarified that the gain profile can be controlled according to the present invention.
[0119]
Example 3
FIG. 11 is a block diagram showing a third embodiment of the optical fiber amplifier of the present invention. In FIG. 11, the same parts as those in FIG. 9 are denoted by the same reference numerals, and the description thereof is omitted. As shown in FIG. 11, in addition to FIG. 9, the optical fiber amplifier extracts the control light from the demultiplexer 3702 and the demultiplexer 3702 that branch the output light branched from the demultiplexer 3306 out of the output light. Filter 3703 for detecting, control light power 3704 for detecting the control light power extracted from the filter, splitter 3701 for branching the excitation light from the isolator 3308 to the multiplexers 3303 and 3705, and the amplification optical fiber 3304 It comprises a multiplexer 3705 that combines the signal light and the excitation light from the demultiplexer 3701. Here, the control light source 3307 has a photodetector for detecting the input control light source power.
[0120]
As the multiplexer 3705, a melt-drawn fiber type, a dielectric multilayer type, or a circulator combined with a fiber grating can be used. Examples of the filter 3703 include a dielectric multilayer filter and a fiber grating.
[0121]
In this embodiment, the signal light is 1460 to 1490 nm, 16 waves at intervals of 2 nm, the amplification optical fiber 3304 is a 20 m optical fiber in which ZBLAN fluoride glass is used as a host and 2000 ppm of thulium ion is added to the core, and the wavelength of the control light source 3307 is 1450 nm, excitation wavelength is 1390 nm, isolators 3302 and 3305 are center wavelengths of 1480 nm, demultiplexers 3501 and 3306 are file couplers having a branching ratio of 95: 5, and a multiplexer 3301 multiplexes and demultiplexes bands of 1452 nm or less and 1458 nm or more. Dielectric multilayer type wavelength division multiplexing couplers, demultiplexers 3701 and 3702 are fiber couplers with a branching ratio of 3 dB, and multiplexers 3303 and 3705 are dielectric multilayer type wavelength division multiplexing that multiplexes bands of 1430 nm or less and 1440 nm or more. Couplers and filters 3502, 3310 are Heart wavelength 1490 nm, that of the half-width 0.8nm transmission band filter 3703, center wavelength 1450 nm, half-value width of the transmission band were from 0.8nm. Signal light to be monitored was 1450 nm control light and 1490 nm signal light.
[0122]
Here, the gains of the two signal lights to be monitored are G (1450 nm) and G (1490 nm), and the total gain G _S And the gain difference ΔG is defined as follows.
[0123]
G _S = G (1450 nm) + G (1490 nm) (1)
ΔG = G (1450 nm) −G (1490 nm) (2)
[0124]
G _S And ΔG are uniquely determined, G (1450 nm) and G (1490 nm) are also determined. _S And G (1450 nm) and G (1490 nm) can be controlled by controlling the values of ΔG and ΔG.
[0125]
12 shows that the input signal light power is −15 dBm / ch. , Pump light power 400 m The total gain G of the two signal lights to be monitored when the power of the control light is changed while being fixed at W _S And a gain difference ΔG. G _S Is indicated by a solid line, and ΔG is indicated by a broken line. G _S Both the values of ΔG and ΔG monotonously decrease with increasing control light power.
[0126]
FIG. 13 shows that in Example 3, the input signal light power is −15 dBm / ch. The total gain G of the two signal lights to be monitored when the control light power is fixed to -15 dBm and the pump light power is changed. _S And a gain difference ΔG. G as in FIG. _S Is indicated by a solid line, and ΔG is indicated by a broken line. As the excitation light increases, G _S Increases monotonically, while ΔG hardly changes.
[0127]
From the results of FIG. 12 and FIG. 13, it is understood that gain spectrum control can be performed by applying the following control algorithm. When the gain of the signal light to be monitored changes, first, the power of the control light is set so that ΔG that hardly changes even if the pump light power is changed while the pump light power is fixed becomes a desired value. adjust. Next, the control light power is fixed and the desired G _S The excitation light power is adjusted so that As a result, gain spectrum control becomes possible.
[0128]
As shown in FIG. 11, the signal light is demultiplexed by the demultiplexer 3501, is combined with the control light from the control light source 3307 by the multiplexer 3301, passes through the isolator 3302, and enters the multiplexer 3303. Further, after being combined with the excitation light from the demultiplexer 3701 and passing through the amplification optical fiber 3304, it is passed through the combiner 3705 that combines the excitation light from the demultiplexer 3701 and the signal light, and isolators 3305. , And is demultiplexed by the demultiplexer 3306, one of which is output as signal light.
[0129]
One of the signal lights demultiplexed by the demultiplexer 3501 is extracted by the filter 3502 only after the signal light of 1490 nm, is detected by the photodetector 3503, and the control light power input to the amplification optical fiber is The output light detected by the photodetector in the control light source 3307 and demultiplexed by the demultiplexer 3306 is further branched by the demultiplexer 3702, and after one of the signal light of 1490 nm is extracted by the filter 3310, The light is detected by the light detector 3311, and the other is extracted by the filter 3703 only after the 1450 nm control light, and then detected by the light detector 3704. The detected values are converted into electric signals and then sent to the calculation unit 3313. .
[0130]
The calculation unit 3313 calculates the gains of the signal light and the control light from the received detection value, and based on this, calculates G _S And ΔG are calculated, and the preset G _S And the current values of the control light source and the excitation light source are calculated so as to coincide with ΔG, and these values are sent to the control unit 3312.
[0131]
The control unit 3312 adjusts the light amounts of the control light source 3307 and the excitation light source 3309 so that the current value sent from the calculation unit 3313 is obtained.
[0132]
FIG. 14 shows that -10 dBm / ch. It is the figure which plotted the deviation of the gain spectrum of each input condition (-25, -20, -15, -10dBm / ch.) On the basis of the gain spectrum at the time of input. A gain deviation of 0.3 dB was realized for all input conditions, demonstrating the effectiveness of the present invention.
[0133]
Example 4
The configuration of the present embodiment is the same as that described in the third embodiment with reference to FIG. 11, but is an example in which numerical values are set in the arithmetic unit. Accordingly, the same components and devices as those of the third embodiment are used except for the processing in the calculation unit, that is, the members and devices of each component. In the following description, reference is made to FIG.
[0134]
In this embodiment, the signal light is 1490-1520 nm, 16 waves at 2 nm intervals, the amplification optical fiber 3304 is a 4 m optical fiber in which ZBLAN fluoride glass is used as a host, and 8000 ppm of thulium ions are added to the core, and the wavelength of the control light source 3307 is 1470 nm, excitation wavelength is 1400 nm, isolators 3302, 3305, and 3308 are center wavelengths of 1470 nm, demultiplexers 3501 and 3306 are fiber couplers with a branching ratio of 95: 5, and a multiplexer 3301 multiplexes and demultiplexes bands of 1472 nm and below and 1478 nm and above The dielectric multilayer type wavelength division multiplexing couplers, the demultiplexers 3701 and 3702 are 3 dB branch couplers, and the multiplexers 3303 and 3705 are dielectric multilayer type wavelength division multiplexing couplers that combine the bands of 1430 nm and below and 1440 nm and above. , Filters 3502, 3310, transmission What range of 1490～1520Nm, filter 3703, center wavelength 1470 nm, the half value width of the transmission band was used, one of 0.8 nm. The signal light to be monitored was 1470 nm control light and the light quantity in a predetermined wavelength range, and all signal light in the wavelength range of 1490 to 1520 nm.
[0135]
Here, the gains of the two types of signal light to be monitored are G (1470 nm) and G (1490-1520 nm), and the total gain G _S And the gain difference ΔG is defined as follows.
[0136]
G _S = G (1470 nm) + G (1490-1520 nm) (1)
ΔG = G (1470 nm) −G (1490-1520 nm) (2)
[0137]
G _S And ΔG are uniquely determined, G (1470 nm) and G (1490-1520 nm) are also determined. _S And ΔG can be controlled to control G (1470 nm) and G (1490 to 1520 nm).
[0138]
G of this example _S And ΔG, the change in G of Example 3 with respect to changes in pumping light power and control light power _S And ΔG show the same tendency, and gain spectrum control is possible by applying the same control algorithm. Similarly to the third embodiment, the input signal light is input to the demultiplexer 3501, and is demultiplexed by the demultiplexer 3306, and one of them is output as the signal light.
[0139]
The signal light demultiplexed by the demultiplexer 3501 is detected by the photodetector 3503 after all signal light of 1490 to 1520 nm, which is a predetermined wavelength range within the amplification band, is extracted by the filter 3502. The control light power input to the amplification optical fiber is detected by a photodetector built in the control light source, and the signal light and control light demultiplexed by the demultiplexer 3306 are branched by the demultiplexer 3702, while Is extracted by the filter 3310 after the entire signal light power of 1490 to 1520 nm is extracted and then detected by the photodetector 3311, and the other is extracted by the filter 3703 only after the control light of 1470 nm is detected by the photodetector 3704. These detected values are converted into electrical signals and then sent to the calculation unit 3313. Based on these detected values as in the third embodiment, the control unit 3312 controls the control light source 3307 and the control light source 3307. Adjusting the light intensity of the excitation light source 3309.
[0140]
During gain control of the present invention, −25 to −10 dBm / ch. A gain deviation of 0.3 dB or less was realized for the input conditions, and the effectiveness of this control was verified.
[0141]
(Example 5)
The configuration of this example is the same as that described in Example 3 with reference to FIG. 11, and the same members and devices as those in Example 3 are used. However, Ho-doped light is used as an amplification optical fiber. Use fiber. This point is different from the third embodiment. In the following description, reference is made to FIG.
[0142]
In this embodiment, the signal light is 2890-2920 nm, 16 waves at 2 nm intervals, and the amplification optical fiber 3304 uses ZBLAN fluoride glass as a host and holmium ions (Ho) in the core. ³⁺ ) Is added at 1000 ppm, the control light source 3307 has a wavelength of 2880 nm, the excitation wavelength is 1200 nm, the isolators 3302 and 3305 have center wavelengths of 2900 nm, the duplexers 3501 and 3306 have a branching ratio of 95: 5, and are combined. The device 3301 is a dielectric multilayer type wavelength division multiplexing coupler that multiplexes and demultiplexes bands of 2882 nm or less and 2888 nm or more, the demultiplexers 3701 and 3702 are 3 dB branch couplers, and the multiplexers 3303 and 3705 are 1300 nm or less and 1310 nm or more. Dielectric multilayer film type wavelength division multiplexing couplers for combining the bands, filters 3502 and 3310, having a transmission band of 2890-2920 nm, filter 3703 having a center wavelength of 2880 nm, and a half bandwidth of the transmission band of 0.8 nm, It was used. The signal light to be monitored was all signal light in the wavelength range of 2890 to 2920 nm as the control light of 2880 nm and the light quantity in the predetermined wavelength range.
[0143]
Here, G (2880 nm) and G (2890-2920 nm) are used as the gains of the two types of signal light to be monitored, and the total gain G _S And the gain difference ΔG is defined as follows, and the total gain G is the same as in the fourth embodiment. _S And gain difference ΔG, and G inputted from the outside _S And ΔG, the current values of the control light source and the excitation light source are calculated, and these values are sent to the control unit 3312. The control unit 3312 adjusts the control light source 3307 and the excitation light source 3309 so as to have the received current value.
[0144]
G _S = G (2880 nm) + G (2890-2920 nm) (1)
ΔG = G (2880 nm) −G (2890-2920 nm) (2)
[0145]
During gain control of the present invention, −25 to −10 dBm / ch. A gain deviation of 0.3 dB or less was realized for the input conditions, and the effectiveness of this control was demonstrated.
[0146]
(Example 6)
The configuration of this example is the same as that described in Example 3 with reference to FIG. 11, and the same members and devices as those in Example 3 are used. However, Er-doped light is used as an amplification optical fiber. Use fiber. This point is different from the third embodiment. In the following description, reference is made to FIG.
[0147]
In this embodiment, the signal light is 1690 to 1720 nm, 16 waves at 2 nm intervals, and the amplification optical fiber 3304 uses ZBLAN fluoride glass as a host and erbium ions (Er in the core). ³⁺ ) Is added at 2000 ppm, the wavelength of the control light source 3307 is 1680 nm, the excitation wavelength is 800 nm, the isolators 3302 and 3305 are center wavelengths of 1700 nm, the duplexers 3501 and 3306 are fiber couplers with a branching ratio of 95: 5 The device 3301 is a dielectric multilayer type wavelength division multiplexing coupler that multiplexes and demultiplexes the bands of 1682 nm or less and 1688 nm or more, the demultiplexers 3701 and 3702 are 3 dB branch couplers, and the multiplexers 3303 and 3705 are 1300 nm or less and 1310 nm or more. Dielectric multilayer type wavelength division multiplexing couplers for combining the bands, filters 3502 and 3310 have a transmission band of 1690 to 1720 nm, filter 3703 has a center wavelength of 1680 nm, and a half-width of the transmission band of 0.8 nm, It was used. The signal light to be monitored was 1680 nm control light and the total light quantity in the wavelength range of 1690 to 1720 nm as the light quantity in the predetermined wavelength range.
[0148]
Here, the gains of the two types of signal light to be monitored are G (1680 nm) and G (1690 to 1720 nm), and the total gain G _S And the gain difference ΔG is defined as follows, and the total gain G is the same as in the fifth embodiment. _S And gain difference ΔG, and G inputted from the outside _S And ΔG, the current values of the control light source and the excitation light source are calculated, and these values are sent to the control unit 3312. The control unit 3312 adjusts the control light source 3307 and the excitation light source 3309 so as to have the received current value.
[0149]
G _S = G (1680 nm) + G (1690 to 1720 nm) (1)
ΔG = G (1680 nm) −G (1690 to 1720 nm) (2)
[0150]
During gain control of the present invention, −25 to −10 dBm / ch. A gain deviation of 0.3 dB or less was realized for the input conditions, and the effectiveness of this control was demonstrated.
[0151]
As described above, in Examples 3 to 6, Tm is used as the added ion. ³⁺ , Ho ³⁺ , Er ³⁺ However, the present invention is not limited to the use of these ions. That is, the present invention can be applied to a rare earth-doped optical amplifier that does not satisfy the condition that the total number of ions at the start and end positions of amplification is constant.
[0152]
(Example 7)
FIG. 16 is a block diagram showing a seventh embodiment of the optical fiber amplifier of the present invention. In FIG. 16, the same parts as those in FIG. 11 are denoted by the same reference numerals, and the description thereof is omitted. As shown in FIG. 16, in addition to the configuration of FIG. 11, this optical fiber amplifier includes a demultiplexer 4101 for demultiplexing signal light, a filter 4102 for extracting signal light from the demultiplexer 4101, and a filter 4102. Includes a photodetector 4103 for detecting the signal light power input to. In this embodiment, the filter 3703 extracts the signal light from the branching filter 3702, and the photodetector 3704 detects the extracted signal light power. The excitation light power was fixed at 400 mW.
[0153]
As the demultiplexer 4101, a melt-drawn fiber type (both branch type and wavelength division multiplexing type), a dielectric multilayer film type, a circulator combined with a fiber grating, and the like can be used. Examples of the filter 4102 include a dielectric multilayer filter and a fiber grating.
[0154]
In this embodiment, the signal light is 1480-1510 nm, 16 waves at 2 nm intervals, the amplification optical fiber 3304 is a 5 m optical fiber in which ZBLAN fluoride glass is used as a host and 6000 ppm of thulium ions are added to the core, and the wavelength of the control light source 3504 is 1470 nm, excitation wavelength is 1410 nm, isolators 3302, 3305, and 3308 are center wavelengths of 1480 nm, demultiplexers 3501 and 3306 are fiber couplers with a branching ratio of 95: 5, and a multiplexer 3301 multiplexes and demultiplexes bands of 1472 nm and below and 1478 nm and above Dielectric multilayer type wavelength division multiplexing couplers, demultiplexers 3701, 3702, and 4101 are 3 dB branch couplers, and multiplexers 3303 and 3705 are dielectric multilayer type wavelength divisions that combine bands of 1430 nm or less and 1440 nm or more. Multiple coupler, filter 3310, 350 Is the center wavelength 1480 nm, that of the half-width 0.8nm transmission bands, filters 4102,3703, the center wavelength of 1510 nm, the half value width of the transmission band was used, one of 0.8nm. The signal light to be monitored was 1480 nm and 1510 nm signal light.
[0155]
Here, let G (1480 nm) and G (1510 nm) be the gains of the two types of signal light to be monitored, and the total gain G _S And the gain difference ΔG is defined as follows.
[0156]
G _S = G (1480 nm) + G (1510 nm) (1)
ΔG = G (1480 nm) −G (1510 nm) (2)
[0157]
G _S And ΔG are uniquely determined, G (1480 nm) and G (1510 nm) are also determined. _S And G (1480 nm) and G (1510 nm) can be controlled by controlling ΔG.
[0158]
G of this example _S And ΔG, the change in G of Example 3 with respect to changes in pumping light power and control light power _S And ΔG, the gain spectrum can be controlled by applying the same control algorithm. _S And ΔG are calculated, and the light amounts of the control light source 3504 and the excitation light source 3309 are adjusted based on this.
[0159]
In this embodiment, the gain is obtained using two channels (1480 nm and 1510 nm) of the wavelength multiplexed signal light, and G _S And ΔG are calculated to control the light amounts of the control light source 3504 and the excitation light source 3309. However, an embodiment in which control is performed using more channels is also possible. In this case, for example, a method of using a gain difference between two arbitrary two channels instead of ΔG as a gain difference between channels can be considered. _S And ΔG, variables can be freely defined within a range in which the gain of each channel can be controlled, and used for light quantity control.
[0160]
During gain control of the present invention, −25 to −10 dBm / ch. A gain deviation of 0.3 dB or less was realized for the input conditions, and the effectiveness of this control was demonstrated.
[0161]
(Example 8)
FIG. 17 is a block diagram showing an eighth embodiment of the optical fiber amplifier of the present invention. In FIG. 17, the same parts as those in FIG. 11 are denoted by the same reference numerals, and the description thereof is omitted. As shown in FIG. 17, this optical fiber amplifier has a calculation unit 4313 in addition to the configuration of FIG. The calculation unit 4313 includes a correction circuit 4301 for correcting the signal light power detected by the photodetector 3503 and a calculation circuit 4302 that performs a calculation based on the corrected signal light power and sends the result to the control unit 3312. As in the fourth embodiment, the same members and devices as those in the third embodiment are used.
[0162]
In this embodiment, the signal light is 1480 to 1510 nm, 16 waves at intervals of 2 nm, the amplification optical fiber 3304 is a 5 m optical fiber in which ZBLAN fluoride glass is used as a host and 6000 ppm of thulium ions are added to the core, and the wavelength of the control light source 3307 is 1470 nm, excitation wavelength is 1390 nm, isolators 3302 and 3305 are center wavelengths of 1480 nm, multiplexer 3301 is a dielectric multilayer type wavelength division multiplexing coupler that multiplexes and demultiplexes bands below 1472 nm and 1478 nm, and demultiplexers 3701 and 3702 are 3 dB branch couplers, multiplexers 3303 and 3705 are dielectric multi-layer wavelength division multiplexing couplers that combine bands below 1430 nm and 1440 nm, and duplexers 3501 and 3306 are branch couplers and filters with a branch ratio of 95: 5 3502 and 3310 have a transmission band of 1480. Those 1510 nm, the filter 3703, center wavelength 1470 nm, the half value width of the transmission band was used, one of 0.8 nm. The signal light to be monitored was 1470 nm control light and the total light quantity in the wavelength range of 1480-1510 nm as the light quantity in the predetermined wavelength range.
[0163]
Here, it is assumed that the gains of the two types of signal light to be monitored are G (1470 nm) and G (1480 to 1510 nm). However, the total signal light power detected by the photodetector 3503 includes light other than the signal light (amplified spontaneous emission light or the like) and may be so large that it cannot be ignored. This tendency is particularly noticeable after multistage relay. Accordingly, the correction circuit 4301 generates a correction value calculated from a certain function corresponding to the input signal light power, and adds or multiplies the value to G (1480 to 1510 nm) calculated by the calculation unit, thereby obtaining G (1480). ˜1510 nm). G (1480-1510 nm) after correction is defined as G ′ (1480-1510 nm). Total gain G _S And the gain difference ΔG is defined as follows.
[0164]
G _S = G (1470 nm) + G '(1480-1510 nm) (1)
ΔG = G (1470 nm) −G ′ (1480-1510 nm) (2)
[0165]
G _S And ΔG are uniquely determined, G (1470 nm) and G ′ (1480 to 1580 nm) are also determined. _S And ΔG can be controlled to control G (1470 nm) and G ′ (1480 to 1510 nm).
[0166]
G of this example _S And ΔG, the change in G of Example 3 with respect to changes in pumping light power and control light power _S And ΔG show the same tendency, and gain spectrum control is possible by applying the same control algorithm. Similarly to the third embodiment, the input signal light is demultiplexed by the demultiplexer 3501, demultiplexed by the demultiplexer 3306, and one of them is output as the signal light.
[0167]
The signal light branched by the demultiplexer 3501 is detected by a photodetector 3503 after all signal light of 1480 to 1510 nm is extracted by a filter 3502, and the control light is a photodetector built in the control light source. The signal light and the control light detected by 3307 and demultiplexed by the demultiplexer 3306 are further branched by the demultiplexer 3702, and one of them extracts the control light of 1470 nm by the filter 3703, and then the photodetector 3704. The other signal is extracted by the filter 3310 and is detected by the photodetector 3311. The detected value is converted into an electric signal, and then the arithmetic circuit 4302 of the arithmetic unit 4313 is detected. Sent to. Further, the detection value detected by the photodetector 3503 is sent to the correction circuit 4301.
[0168]
The detection value detected by the light detector 3503 is corrected by the correction circuit 4301 in the calculation unit 3313 according to the following expression.
[0169]
Pre = Pc (1.0029−0.2530 ・ Pc−3.0981 ・ Pc ・ Pc),
Pc ≧ 0.005 (3)
Pre: Control light source output after correction (mW)
Pc: Control light source output before correction (mW)
[0170]
This correction formula is used to correct the control light power value detected by the light detector 3503 by weighting the value, and the relationship between the control light output before and after the correction is shown in FIG. The horizontal axis is the control light output before correction, and the vertical axis is the difference (ΔP) between the control light output before correction and the control light output after correction. It can be seen that ΔP increases rapidly as the control light output before correction increases. This equation is merely an example, and various functions can be taken as long as the operating condition of the amplifier at the time of control is satisfied. In this embodiment, only the correction of the output value from the light detector 3503 has been described. However, in another embodiment, the output value from the light detector 3704 is corrected or the light detectors 3503 and 3704 are corrected. A configuration in which both of the output values are corrected is also possible.
[0171]
The corrected control light source output value is sent to the arithmetic circuit 4302.
[0172]
The arithmetic circuit 4302 calculates the gains of the signal light and the control light from the received detection value and correction value, and then calculates G _S And ΔG are calculated. Preset G _S And ΔG, the current values of the control light source and the excitation light source are calculated, and these values are sent to the control unit 3312.
[0173]
The control unit 3312 adjusts the control light source 3307 and the excitation light source 3309 so that the current value sent from the arithmetic circuit 4302 is obtained.
[0174]
During gain control of the present invention, −25 to −10 dBm / ch. A gain deviation of 0.3 dB or less was realized for the input conditions, and the effectiveness of this control was demonstrated.
[0175]
Example 9
FIG. 19 is a block diagram showing a ninth embodiment of the optical fiber amplifier of the present invention. 19, the same parts as those in FIG. 17 are denoted by the same reference numerals, and the description thereof is omitted. As shown in FIG. 19, in addition to the configuration of FIG. 17, excitation light from circulators 4502, 4504 and isolator 3308 combined with fiber gratings 4501, 4505 for multiplexing and demultiplexing control light and signal light, and signal light from circulator 4502 A multiplexer 4503 for multiplexing the control light is included.
[0176]
Instead of the circulators 4502 and 4504 combined with the fiber gratings 4501 and 4505, a melt-drawn fiber type (both branch type and wavelength division multiplexing type), a dielectric multilayer film type, and the like can be used. As the multiplexer 4503, a melt-drawn fiber type, a dielectric multilayer type, a circulator combined with a fiber grating, or the like can be used.
[0177]
In this embodiment, the signal light is 1478, 1480, 1482, 1484, 1486, 1488, 1490, 1492, 1496, 1498, 1500, 1502, 1504, 1506, 1508, 1510 nm, and the amplification optical fiber 3304 is ZBLAN. 7m optical fiber with 6000ppm thulium ion added to the core with fluoride glass as host, control light source wavelength 1494nm, excitation wavelength 1400nm, isolator 3308 center wavelength 1400nm, circulators 4502 and 4504 combined with fiber gratings 4501 and 4505 Is a wavelength division multiplexing coupler of a dielectric multilayer type that multiplexes bands of 1430 nm or less and 1440 nm or more. The multiplexer 4503 uses a wavelength division multiplexing coupler of 1494 nm (bandwidth 0.8 nm). The excitation light power was fixed at 300 mW.
[0178]
As shown in FIG. 19, the signal light is combined with the control light by the demultiplexer 4501, passes through the circulator 4502 combined with the fiber grating 4501, is further combined with the excitation light by the multiplexer 4503, After passing through the amplifying optical fiber 3304, it is demultiplexed by a circulator 4504 combined with a fiber grating 4505, and one of them is output as signal light.
[0179]
The control light power input to the amplification optical fiber 3304 is detected by a photodetector inside the control light source 3307, and the control light power output from the amplification optical fiber is output from a circulator 4504 combined with a fiber grating 4505. The signals are demultiplexed and detected by the photodetector 3311, and their values are sent to the calculation unit 3313 as an electric signal.
[0180]
The detection value detected by the photodetector inside the control light source 3307 is corrected in the same way as in the eighth embodiment by the correction circuit 4301 in the calculation unit 3313 according to the following formula, and the relationship between the control light output before and after the correction is shown in FIG. Shown in
[0181]
Pre = Pc · (1.0106−0.73559 · Pc), Pc ≧ 0.005
Pre: Control light source output after correction (mW)
Pc: Control light source output before correction (mW)
[0182]
This equation is merely an example, and various functions can be taken as long as the operating condition of the amplifier at the time of control is satisfied. In the present embodiment, only the correction of the output value from the internal control light source 3307 of the photodetector has been described, but in addition to this, the output value from the photodetector 3311 is corrected or the photodetector 3311 is corrected. In addition, a configuration in which both the output value from the control light source 3307 is corrected is also possible.
[0183]
The corrected control light source output value is sent to the arithmetic circuit 4302, and the light quantity of the control light source 3307 is adjusted based on the correction value.
[0184]
During gain control of the present invention, −25 to −10 dBm / ch. A gain deviation of 0.3 dB or less was realized for the input conditions, and the effectiveness of this control was verified.
[0185]
【The invention's effect】
According to the present invention, in an optical amplifier using an amplification medium to which rare earth ions are added, at least a control light source is provided in the optical amplifier, the gain of control light or signal light is controlled, and at least the control light source is provided in the optical amplifier. Can control the control light gain and the signal light gain, or at least have a control light source in the optical amplifier, and can control the gain spectrum of the two signal lights to be constant. .
[0186]
In addition, when applied to a WDM transmission system, the optical amplifier of the present invention can control the gain spectrum of signal light with respect to changes in the input spectrum accompanying changes in the number of signal channels and changes in the gain spectrum due to temperature changes. .
[0187]
Further, the optical amplifier of the present invention can control the output required for compensation when the transmission line loss changes with time, and can cope with hole burning and the like.
[0188]
The control method of the present invention can be used in an optical amplifier using an amplification medium to which a general rare earth ion is added.
[Brief description of the drawings]
FIGS. 1A and 1B are schematic views showing energy levels that are largely involved in amplification of Er ions and Pr ions when quartz is used as a host glass, respectively.
FIGS. 2A and 2B are diagrams showing energy positions of Er ions and Tm ions, respectively.
FIG. 3 shows a wavelength λ in TDFA. ₁ FIG. 6 is a diagram showing characteristics when the gain at is constant and the excitation light amount of TDFA is adjusted.
FIG. 4 is a diagram illustrating a gain spectrum when the light amount of the control light source is adjusted so that the total input power of the control light and the signal light is always constant.
FIG. 5 is a diagram illustrating a gain spectrum when the light amount of the control light source is adjusted so that the total output power of the control light and the signal light is always constant.
FIG. 6 is a configuration diagram for explaining an optical amplifier according to the present invention.
FIG. 7 is a configuration diagram for explaining a first embodiment of an optical fiber amplifier according to the present invention;
FIG. 8 is a diagram plotting the maximum value of the deviation of the gain profile when each input signal level is changed in the first embodiment.
FIG. 9 is a configuration diagram for explaining a second embodiment of the optical fiber amplifier according to the present invention;
FIG. 10 is a graph plotting a maximum value of a deviation of a gain profile when each input signal level is changed in the second embodiment.
FIG. 11 is a configuration diagram for explaining a third embodiment of the optical fiber amplifier according to the present invention;
FIG. 12 shows that in the third embodiment, the input signal light power is −15 dBm / ch. When the pumping light power is fixed at 400 mW and the control light power is changed, the total gain G of the two signal lights to be monitored _S And a gain difference ΔG.
FIG. 13 shows that in the third embodiment, the input signal light power is −15 dBm / ch. The total gain G of the two signal lights to be monitored when the control light power is fixed to -15 dBm and the pump light power is changed. _S And a gain difference ΔG.
FIG. 14 shows -10 dBm / ch. It is the figure which plotted the deviation of the gain spectrum of each input condition (-25, -20, -15, -10dBm / ch.) On the basis of the gain spectrum at the time of input.
FIG. 15 is a configuration diagram for explaining fourth to sixth embodiments;
FIG. 16 is a block diagram for explaining a seventh embodiment of the optical fiber amplifier of the present invention.
FIG. 17 is a configuration diagram for explaining an eighth embodiment of an optical fiber amplifier according to the present invention;
FIG. 18 is a diagram showing the relationship of control light output before and after correction in the eighth embodiment.
FIG. 19 is a configuration diagram for explaining a ninth embodiment of the optical fiber amplifier according to the present invention;
FIG. 20 is a diagram showing the relationship of control light output before and after correction in the ninth example.
[Explanation of symbols]
600 optical amplifier
601 and 603 monitoring means
602 Amplifier
604 control system
610 Control light source
612 Amplification medium
614, 616 Excitation light source
618, 620, 622 multiplexer
624, 626 Isolator
628, 636 signals
630 control unit
632 arithmetic unit
634 Optical branching element
634 duplexer
635 photodetector
3301, 3303, 3705, 4503 multiplexer
3302, 3305, 3308 Isolator
3303 Wavelength division multiplexing coupler (multiplexer)
3304 Optical fiber for amplification
3306, 3501, 3701, 3702, 4101, 4501
3307, 3311, 3503, 3704, 4103 photodetector
3307, 3307, 3504 Control light source
3309 excitation light source
3310, 3502, 3703, 4102 Filter
3312 control unit
3313, 4313 arithmetic unit
4301 Correction circuit
4302 arithmetic circuit
4501, 4505 Fiber grating
4502, 4504 Circulator

Claims (18)

An amplifying medium comprising a rare earth-doped optical fiber or a rare earth-doped optical waveguide in which rare earth ions are added to the core and / or clad;
Excitation means for exciting the amplification medium and the power of the excitation light being constant ;
One or more control light sources that can input control light having a wavelength within the amplification band of the amplification medium to the amplification medium and control the inversion distribution state of the amplification medium by changing its input power;
Monitoring means for monitoring the power of control light input to and output from the amplification medium, the power of signal light of one or more channels, the power of light in a predetermined wavelength range within the amplification band, or a combination thereof ; and
Control light power input / output to / from the amplification medium from the value obtained from the monitoring means , power of signal light of one or more channels, power of light in a predetermined wavelength range within the amplification band, or a combination thereof calculating the gain, characterized in that it comprises a control unit for controlling the gain profile constant by controlling the control light source to match the value of the gain input from the value or the outside of the gain to a predetermined Optical amplifier. 2. The optical amplifier according to claim 1, wherein the monitoring means monitors the power of the control light. 2. The optical amplifier according to claim 1, wherein the monitoring means monitors the power of signal light of one channel or more. 2. The optical amplifier according to claim 1, wherein the monitoring unit monitors the power of light in a predetermined wavelength range within an amplification band. An amplifying medium comprising a rare earth-doped optical fiber or a rare earth-doped optical waveguide in which rare earth ions are added to the core and / or clad;
Excitation means for exciting the amplification medium;
One or more control light sources that can input control light having a wavelength within the amplification band of the amplification medium to the amplification medium and control the inversion distribution state of the amplification medium by changing its input power;
Monitoring means for monitoring the power of light of a plurality of wavelengths input to and output from the amplification medium; and
A control unit that performs constant gain profile control, calculates gains of a plurality of wavelengths from values obtained from the monitoring unit, calculates a difference and sum of a plurality of gains, and the plurality of gain differences are determined in advance. was controlling the control light source to match the value of the difference of the gain input from the value or the outside of the difference in gain, the sum of a plurality of gain, input from the value or the outside of the sum of the gain to a predetermined An optical amplifier comprising: a control unit that controls the pumping means so as to coincide with a value of a sum of gains to perform constant gain profile control . 6. The optical amplifier according to claim 5, wherein the monitoring means monitors control light power and signal light power . 6. The optical amplifier according to claim 5, wherein the monitoring means monitors the power of a plurality of signal lights. Said monitoring means, the control light and optical amplifier according to claim 5, characterized in that in the amplification band is to monitor the power of the light of a predetermined wavelength range.   The control unit includes a correction unit that corrects a value calculated by the control unit, a gain calculated by the control unit, a gain difference, or a gain sum according to the values obtained from the monitoring unit. Item 9. The optical amplifier according to Item 1 to 8.   10. The optical amplifier according to claim 1, wherein at least one rare earth ion selected from thulium, holmium and erbium is used as the rare earth ion. An amplifying medium comprising a rare earth-doped optical fiber or a rare earth-doped optical waveguide in which rare earth ions are added to the core and / or clad;
Excitation means for exciting the amplification medium and the power of the excitation light being constant ;
One or more control light sources that can control the inversion distribution state of the amplification medium by changing the input power of the control light from the control light source;
Monitoring means, and
A method for controlling a gain profile of an optical amplifier including a control unit to be constant ,
Inputting control light having a wavelength within an amplification band of the amplification medium to the amplification medium;
Monitoring the power of control light input to and output from the amplification medium, the power of signal light of one or more channels, the power of light in a predetermined wavelength range within the amplification band, or a combination thereof ;
From the values obtained from the monitoring means, the power of control light, the power of signal light of one or more channels, the power of light in a predetermined wavelength range within the amplification band, or the gain of a combination of these is calculated and determined in advance. how to control the gain profile constant, characterized in that and controlling the gain profile constant the control light source controlled to so as to coincide with the gain value input from the value or the outside of the gain that has been . The control method according to claim 11 , wherein the monitoring unit monitors the power of the control light. The control method according to claim 11 , wherein the monitoring unit monitors the power of signal light of one channel or more. The control method according to claim 11 , wherein the monitoring unit monitors the power of light in a predetermined wavelength range within the amplification band. An amplifying medium comprising a rare earth-doped optical fiber or a rare earth-doped optical waveguide in which rare earth ions are added to the core and / or clad;
Excitation means for exciting the amplification medium;
One or more control light sources that can control the inversion distribution state of the amplification medium by changing the input power of the control light from the control light source;
A method for controlling a gain profile of an optical amplifier including a monitoring unit and a control unit to be constant ,
Inputting control light having a wavelength within an amplification band of the amplification medium to the amplification medium;
Monitoring the power of light of a plurality of wavelengths input to and output from the amplification medium;
After calculating gains of a plurality of wavelengths from the values obtained from the monitoring means, a difference and a sum of a plurality of gains are calculated, and the plurality of gain differences are input from a predetermined gain difference value or externally. The control light source is controlled so as to coincide with a value of gain difference, and the excitation is performed so that a sum of a plurality of gains coincides with a predetermined gain sum value or an externally input gain sum value. how to control the gain profile constant, characterized in that it comprises a controlling control to the gain profile means constant. The control method according to claim 15 , wherein the monitoring unit monitors the power of control light and signal light. The control method according to claim 15 , wherein the monitoring unit monitors the power of a plurality of signal lights. The method of claim 15 wherein said monitoring means, characterized by monitoring the power of the light of a predetermined wavelength range in the control light and the amplification band.

Images