JP3917919B2

JP3917919B2 - Quadrature demodulation error compensation method and circuit

Info

Publication number: JP3917919B2
Application number: JP2002272597A
Authority: JP
Inventors: 匡夫中川; 克彦荒木
Original assignee: Nippon Telegraph and Telephone Corp
Current assignee: Nippon Telegraph and Telephone Corp
Priority date: 2002-09-19
Filing date: 2002-09-19
Publication date: 2007-05-23
Anticipated expiration: 2022-09-19
Also published as: JP2004112384A

Description

【０００１】
【発明の属する技術分野】
本発明は送受信装置に用いられる直交復調器の位相誤差および振幅誤差を補償する方法および回路に関するものである。
【０００２】
【従来の技術】
直交復調器は、無線通信の受信機において、被変調信号から互いに９０度の位相差を持つ復調信号を取り出す復調器である。従来の直交復調器を図４に示す。図４において、入力端子１より入力される被変調信号は直交復調器８に入力される。この入力信号は２分配され、一方の信号はミキサ９において局部発振器１１の出力信号と乗算され、ローパスフィルタ１３によって高周波成分を除去されてベースバンド同相信号として出力端子２に出力する。また２分配された他方の信号はミキサ１０に入力され、局部発振器１１の出力信号を９０度移相器１２により９０度位相を変えられた信号と乗算され、ローパスフィルタ１４によって高周波成分を除去されてベースバンド直交信号として出力端子３に出力する。このとき、ミキサ９に入力される局部発振器１１の出力信号とミキサ１０に入力される９０度移相器１２の出力信号との間に９０度からの位相誤差や振幅誤差があると、復調特性が劣化する。
【０００３】
９０度移相器１２と同様の働きをする回路として、局部発振器１１の出力信号を９０度の位相差を持つ２信号に分配する９０度分配回路がよく知られているが、この場合も９０度分配回路の２出力信号の９０度からの位相誤差や振幅誤差があると、復調特性が劣化する。
【０００４】
従来では直交変調器における位相誤差、振幅誤差を補償することについては特許文献１に記載があったが、直交復調器についてはなかった。
【０００５】
【特許文献１】
特許第３１４４６４９号明細書
【０００６】
【発明が解決しようとする課題】
上述したように従来の直交復調器では、直交復調時に生じる位相誤差、振幅誤差があると復調特性が劣化する問題点があった。
【０００７】
本発明は、このような問題点を解消するためになされたもので、これら位相誤差、振幅誤差を精度良く補償できる直交復調誤差補償方法および回路を提供することを目的としている。
【０００８】
【課題を解決するための手段】
請求項１に係る発明は、被変調信号を入力してベースバンド信号を出力する直交復調器と、ベースバンド信号を入力して被変調信号を出力する直交変調器と、を備えた送受信装置における直交復調誤差補償方法であって、前記直交復調器の位相誤差および振幅誤差の検出時には、既知の信号を前記直交変調器に与えて前記直交変調器から得られる出力信号を前記直交復調器に入力し、前記直交復調器のベースバンド同相出力信号の平均値を第１ＤＣオフセットとして求めると共に、ベースバンド直交出力信号の平均値を第２ＤＣオフセットとして求めて、前記第１ＤＣオフセットを前記ベースバンド同相出力信号から差し引くと共に、前記第２ＤＣオフセットを前記ベースバンド直交出力信号から差し引き、複数の既知の信号を前記直交復調器に入力したときの、前記第１ＤＣオフセットが差し引かれたそれぞれのベースバンド同相出力信号および前記第２ＤＣオフセットが差し引かれたそれぞれのベースバンド直交出力信号をそれぞれ２乗した上で平均して得たそれぞれの２乗平均値から前記直交復調器の位相誤差および振幅誤差を補償する係数を求め、通常信号受信時には、前記直交復調器のベースバンド同相出力信号から前記第１ＤＣオフセットを差し引くと共に、ベースバンド直交出力信号から前記第２ＤＣオフセットを差し引き、前記第１ＤＣオフセットが差し引かれたベースバンド同相出力信号又は前記第２ＤＣオフセットが差し引かれたベースバンド直交出力信号に対して、前記係数により前記位相誤差および前記振幅誤差の補償処理を行うことを特徴とする直交復調誤差補償方法とした。
【０００９】
請求項２に係る発明は、請求項１に記載の直交復調誤差補償方法において、前記直交変調器に入力すべき前記既知の信号として、ベースバンド同相信号が１と０が交互に繰り返される周期信号でかつベースバンド直交信号が常時０である第１既知信号と、ベースバンド同相信号が常時０でかつベースバンド直交信号が１と０が交互に繰り返される周期信号である第２既知信号と、ベースバンド同相信号およびベースバンド直交信号が同じ１と０が交互に繰り返される周期信号である第３既知信号とを用意し、前記直交復調器に１と０が交互に繰り返される周期信号をベースバンド信号として与えて得られるベースバンド同相出力信号およびベースバンド直交出力信号をそれぞれ前記周期信号の１周期分またはそれ以上の時間で平均してベースバンド同相平均値δ_Iとベースバンド直交平均値δ_Qを求め、前記第１既知信号を用いたときの前記直交復調器のベースバンド同相出力信号から前記ベースバンド同相平均値δ_Iを差し引き、この差し引いた値を前記第１既知信号のうちの周期信号の１周期分またはそれ以上の時間で２乗平均した第１ベースバンド同相２乗平均値ＰＩ_dc1(t)を求め、前記第２既知信号を用いたときの前記直交復調器のベースバンド同相出力信号から前記ベースバンド同相平均値δ_Iを差し引き、この差し引いた値を前記第２既知信号のうちの周期信号の１周期分またはそれ以上の時間で２乗平均した第２ベースバンド同相２乗平均値ＰＩ_dc2(t)を求め、前記第１ベースバンド同相２乗平均値ＰＩ_dc1(t)と前記第２ベースバンド同相２乗平均値ＰＩ_dc2(t)を加算して定数ｋを求め、前記第３既知信号を用いたときの前記直交復調器のベースバンド同相出力信号から前記ベースバンド同相平均値δ_Iを差し引き、この差し引いた値を前記第３既知信号の１周期分またはそれ以上の時間で２乗平均した第３ベースバンド同相２乗平均値ＰＩ_dc3(t)を求め、該第３ベースバンド同相２乗平均値ＰＩ_dc3(t)と前記定数ｋから被変調信号の搬送波と局部発振器との位相差αを求め、前記第１既知信号を用いたときの前記直交復調器のベースバンド直交出力信号から前記ベースバンド直交平均値δ_Qを差し引き、この差し引いた値を前記第１既知信号のうちの周期信号の１周期分またはそれ以上の時間で２乗平均した第１ベースバンド直交２乗平均値ＰＱ_dc1(t)を求め、前記第２既知信号を用いたときの前記直交復調器のベースバンド直交出力信号から前記ベースバンド直交平均値δ_Qを差し引き、この差し引いた値を前記第２既知信号のうちの周期信号の１周期分またはそれ以上の時間で２乗平均した第２ベースバンド直交２乗平均値ＰＱ_dc2(t)を求め、前記第１ベースバンド直交２乗平均値ＰＱ_dc1(t)と前記第２ベースバンド直交２乗平均値ＰＱ_dc2(t)を加算した値と前記定数ｋから前記直交復調器の振幅誤差Ｇを求め、前記第３既知信号を用いたときの前記直交復調器のベースバンド直交出力信号から前記ベースバンド直交平均値δ_Qを差し引き、この差し引いた値を前記第３既知の信号の１周期分またはそれ以上の時間で２乗平均した第３ベースバンド直交２乗平均値ＰＱ_dc3(t)を求め、該第３ベースバンド直交２乗平均値ＰＱ_dc3(t)、前記定数ｋおよび前記位相差αから前記直交復調器の位相誤差φを求め、この位相誤差φおよび前記振幅誤差Ｇを補償する係数を求めることを特徴とする直交復調誤差補償方法とした。
【００１０】
請求項３に係る発明は、請求項１又は２に記載の直交復調誤差補償方法において、前記直交復調器の位相誤差および振幅誤差の検出時には、前記直交変調器に前記既知の信号を与えて得られる出力信号を、前記直交復調器に入力される受信信号と同じ周波数に周波数変換して、前記直交復調器に入力するようにしたことを特徴とする直交復調誤差補償方法とした。
【００１１】
請求項４に係る発明は、被変調信号を入力してベースバンド信号を出力する直交復調器と、ベースバンド信号を入力して被変調信号を出力する直交変調器と、を備えた送受信装置における直交復調誤差補正回路であって、前記直交復調器の既知のベースバンド同相出力信号の平均値を第１ＤＣオフセットとして検出する第１平均値検出回路と、前記直交復調器の既知のベースバンド直交出力信号の平均値を第２ＤＣオフセットとして検出する第２平均値検出回路と、前記直交復調器のベースバンド同相出力信号から前記第１ＤＣオフセットを差し引く第１加算器と、前記直交復調器のベースバンド直交出力信号から前記第２ＤＣオフセットを差し引く第２加算器と、前記直交復調器に複数の既知の信号を入力したとき、前記第１加算器から出力するそれぞれのベースバンド同相出力信号のそれぞれの２乗平均値と前記第２の加算器から出力するそれぞれのベースバンド直交出力信号のそれぞれの２乗平均値から前記直交復調器の位相誤差および振幅誤差を補償する係数を求める２乗平均・演算回路と、該２乗平均・演算回路で得られた前記係数により、通常信号受信時に前記第１加算器から出力するベースバンド同相出力信号又は第２加算器から出力するベースバンド直交信号に対して、前記振幅誤差および前記位相誤差を除去する位相・振幅補償回路と、を具備することを特徴とする直交復調誤差補償回路とした。
【００１２】
【発明の実施の形態】
［第１の実施形態］
図１は第１の実施形態である直交復調誤差補償回路を示すブロック図である。図１において、１は受信信号入力端子、２，３は復調信号出力端子、４，５は変調信号入力端子、６は送信信号出力端子、７，３０はスイッチ、８は直交復調器、１３，１４はローパスフィルタ、１５，１６は平均値検出回路、１７，１８は加算器、１９は２乗平均・演算回路、２０は位相・振幅補償回路、２４は直交変調器である。
【００１３】
直交復調器８はミキサ９，１０、局部発振器１１、９０度移相器１２により構成されている。位相・振幅補償回路２０は乗算器２１，２２、加算器２３により構成されている。直交変調器２４はミキサ２５，２６、局部発振器２７、９０度移相器２８、加算器２９により構成されている。
【００１４】
送信時には直交変調器２４の出力端子と送信信号出力端子６とをスイッチ３０により接続する。受信時には受信信号入力端子１と直交復調器８の入力端子とをスイッチ７により接続する。直交復調器８の位相誤差、振幅誤差の検出時には、スイッチ３０およびスイッチ７により直交変調器２４の出力端子と直交復調器８の入力端子とを接続する。
【００１５】
直交変調器の位相誤差、振幅誤差の補償方法については、特許第３１４４６４９号明細書に開示されている。これによって直交変調器２４の位相誤差、振幅誤差をまず補償し、補償された被変調信号を使って直交復調器８の位相誤差、振幅誤差を検出し、補償する。
【００１６】
直交復調器８の位相誤差、振幅誤差は周波数依存性があるため、誤差検出用に発生させる被変調信号の周波数は受信信号と同じにする。この角周波数をωとし、変調信号入力端子４に入力する信号をＩ_T(t)、変調信号入力端子５に入力する信号をＱ_T(t)とすると、直交変調器２４から出力される被変調信号Ｓ(t)は、

となる。
【００１７】
この信号Ｓ(t)をスイッチ３０およびスイッチ７を介して直交復調器８に入力させる。直交復調器８中の局部発振器１１よりミキサ９に入力される信号はcos（ωt＋α）、局部発振器１１より９０度移相器１２を介してミキサ１０に入力される信号はＧsin（ωt＋α＋φ）と書ける。ただし、Ｇは９０度移相器１２を通過する信号と通過しない信号との振幅誤差、φは９０度移相器１２を通過する信号と通過しない信号との位相誤差、αは被変調信号Ｓ(t)の搬送波と局部発振器１１との位相差である。
【００１８】
したがって、直交復調器８より出力された後、ローパスフィルタ１３を介して得られるベースバンド同相出力信号Ｉ_R(t)、およびローパスフィルタ１４を介して得られるベースバンド直交出力信号Ｑ_R(t)は、

となる。ただしｋは定数、δ_I、δ_QはＤＣオフセットである。一般に能動素子の出力にはＤＣオフセットが含まれているが、直交復調器８の出力信号はベースバンド信号であり、ＤＣ成分を信号として含んでいる。したがって、特に低速信号を復調する場合には、Ｃ（容量）カットによりＤＣオフセットを取り除くことができない。
【００１９】
ここで、直交変調器２４に入力する信号Ｉ_T(t)を１と０が等しい頻度で現れる信号とすると、（２）式のＩ_R(t)を平均値検出回路１５に入力して平均化した値はＤＣオフセットδ_Iとなる。同様に直交変調器２４に入力する信号Ｑ_T(t)を１と０が等しい頻度で現れる信号とすると、（３）式のＱ_R(t)を平均値検出回路１６に入力して平均化した値はＤＣオフセットδ_Qとなる。このとき、直交変調器２４に入力する信号Ｉ_T(t)およびＱ_T(t)を１と０が交互に繰り返される周期信号とすれば、平均値検出回路１５および１６の平均時間は周期信号の１周期分で良い。したがって未知の受信信号からＤＣオフセットを検出する場合に比べて、平均時間を短くすることが可能である。
【００２０】
検出したＤＣオフセットδ_I，δ_Qを加算器１７および１８により差し引いた信号Ｉ_dc(t)およびＱ_dc(t)は、

となる。（５）式は、

に変形できる。
【００２１】
振幅誤差および位相誤差がない場合には、Ｇ＝１、φ＝０であり、（４）式のＩ_dc(t)はそのまま理想的な同相信号として取り出すことができる。一方、（５）式のＱ_dc(t)から振幅誤差および位相誤差を除去した信号をＱ_cor(t)とすると、Ｇ＝１、φ＝０より、

となる。（４）式および（７）式を用いると、（６）式は、

となる。
【００２２】
したがって、φがφ≠ｎπ（rad）（ただし、ｎは０を含めた整数）の範囲で、Ｑ_cor(t)は、

により与えることができる。
【００２３】
すなわち振幅誤差Ｇおよび位相誤差φを求め、ＤＣオフセットを取り除いた信号に対して（９）式の演算を行うことにより、振幅誤差Ｇおよび位相誤差φを除去した理想的な直交信号を得ることができる。
【００２４】
次に、振幅誤差Ｇ、位相誤差φおよび定数ｋを求めるために、あらかじめ定めた信号をＩ_T(t)、Ｑ_T(t)として直交変調器２４に入力する。
【００２５】
まず、Ｉ_T(t)を１と０が交互に繰り返される周期信号とし、Ｑ_T(t)を常時０とする。これを変調信号１（第１既知信号）とする。Ｉ_T(t)＝１のとき、（４）式より加算器１７の出力Ｉ_dc(t)は、

となる。また、Ｉ_T(t)＝０のとき、Ｉ_dc(t)＝０である。したがって、Ｉ_T(t)として入力する周期信号の１周期分またはそれ以上の時間でＩ_dc(t)を２乗平均した値ＰＩ_dc1(t)は、

となる。
【００２６】
次に、Ｉ_T(t)を常時０とし、Ｑ_T(t)を１と０が交互に繰り返される周期信号とする。これを変調信号２（第２既知信号）とする。Ｑ_T(t)＝１のとき、（４）式よりＩ_dc(t)は、

となる。また、Ｑ_T(t)＝０のとき、Ｉ_dc(t)＝０である。したがって、Ｑ_T(t)として入力する周期信号の１周期分以上の時間でＩ_dc(t)を２乗平均した値ＰＩ_dc2(t)は、

となる。
【００２７】
（１１）式および（１３）式より、

が得られる。したがって、

により、定数ｋが求められる。
【００２８】
次に、Ｉ_T(t)およびＱ_T(t)を同じ１と０が交互に繰り返される周期信号とする。これを変調信号３（第３既知信号）とする。Ｉ_T(t)＝０でかつＱ_T(t)＝１のとき、（４）式よりＩ_dc(t)は、

となる。また、Ｉ_T(t)＝０でかつＱ_T(t)＝０のとき、Ｉ_dc(t)＝０である。したがって、Ｉ_T(t)、Ｑ_T(t)として入力する周期信号の１周期分以上の時間でＩ_dc(t)を２乗平均した値ＰＩ_dc3(t)は、

となる。
【００２９】
したがって、位相誤差αは、

であり、（１５）式のｋを代入することによって、位相差αを求めることができる。
【００３０】
次に同じ変調信号で得られる加算器１８の出力Ｑ_dc(t)に着目する。まず、変調信号１の場合に、Ｉ_T(t)＝１のとき、（５）式よりＱ_dc(t)は、

となる。また、Ｉ_T(t)＝０のとき、Ｑ_dc(t)＝０である。したがって、Ｉ_T(t)として入力する周期信号の１周期分以上の時間でＱ_dc(t)を２乗平均した値ＰＱ_dc1(t)は、

となる。
【００３１】
次に、変調信号２の場合に、Ｑ_T(t)＝１のとき、（５）式よりＱ_dc(t)は、

となる。また、Ｑ_T(t)＝０のとき、Ｉ_dc(t)＝０である。したがって、Ｑ_T(t)として入力する周期信号の１周期分以上の時間でＱ_dc(t)を２乗平均した値ＰＱ_dc2(t)は、

となる。
【００３２】
（２０）式および（２２）式より、

が得られる。したがって、振幅誤差Ｇは、

であり、（１５）式のｋを代入することにより、振幅誤差Ｇを求めることができる。
【００３３】
次に変調信号３の場合には、Ｉ_T(t)＝１でかつＱ_T(t)＝１のとき、（５）式よりＱ_dc(t)は、

となる。また、Ｉ_T(t)＝０でかつＱ_T(t)＝０のとき、Ｉ_dc(t)＝０である。したがって、Ｉ_T(t)、Ｑ_T(t)として入力する周期信号の１周期分以上の時間でＱ_dc(t)を２乗平均した値ＰＱ_dc3(t)は、

となる。したがって、位相誤差φは、

であり、（１５）式のｋ、（１８）式のα、（２４）式のＧを代入することにより、位相誤差φを求めることができる。
【００３４】
以上の（１０）式から（２７）式までの演算を２乗平均・演算回路１９で行い、さらに（９）式中のtanφおよびＧcosφを求め、位相・振幅補償回路２０中の乗算器２１および２２にそれぞれ係数として入力する。
【００３５】
この状態で受信信号入力端子１と直交復調器８の入力端子とをスイッチ７により接続することによって、受信信号に対してＤＣオフセット、振幅誤差および位相誤差を除去した理想的な直交信号を出力端子３に得ることができる。またＤＣオフセットを除去した理想的な同相信号は出力端子２に出力される。
【００３６】
本実施形態では受信していない時間を利用して定期的に直交復調器８の位相誤差、振幅誤差を検出することができるため、直交復調時に生じる位相誤差、振幅誤差を精度良く補償することが可能となる。
【００３７】
［第２の実施形態］
図２は第２の実施形態である直交復調誤差補償回路を示すブロック図である。図１と同一部分には同一符号を付与している。本実施形態では第１のローパスフィルタ１３の出力を第１のＡ／Ｄ変換器３１に入力し、第２のローパスフィルタ１４の出力を第２のＡ／Ｄ変換器３２に入力して、それぞれデジタル信号に変換した後に、平均値検出、２乗平均・演算、位相・振幅補償を行っている。これらをデジタル信号処理で行うことにより、精度良く動作させることができる。Ａ／Ｄ変換器は直交復調器８の前に置き、直交復調器をデジタル回路とする構成でも良い。
【００３８】
また、ミキサ９，１０には局部発振器１１の出力信号を同相で入力し、スイッチ７よりミキサ９，１０に入力される被変調信号に９０度移相器３３で９０度位相差を与えている。このように９０度位相の与え方は局部発振の出力信号であっても、被変調信号であっても直交復調器８の出力は同等であり、９０度位相器の振幅誤差、位相誤差を同じように補償できる。
【００３９】
［第３の実施形態］
図３は第３の実施形態である直交復調誤差補償回路を示すブロック図である。図１と同一部分には同一符号を付与している。本実施形態は直交変調器２４の出力信号の周波数と直交復調器８に入力される受信信号の周波数が異なる場合に有効な構成である。本実施形態ではスイッチ３０およびスイッチ７の間に周波数変換用の発振器３４およびミキサ３５を設けて、直交変調器２４の出力信号の周波数を、受信信号入力端子１より入力される信号の周波数に変換している。これにより第１の実施形態と同様に直交変調器２４の出力信号を直交復調器８の位相誤差、振幅誤差の検出に用いることができる。
【００４０】
［その他の実施形態］
なお、以上の第１〜第３の実施形態ではベースバンド同相出力信号を基準として、ベースバンド直交出力信号に対して位相誤差、振幅誤差を補償する処理を行ったが、ベースバンド直交出力信号を基準として、ベースバンド同相出力信号に対して同様に行うこともできる。
【００４１】
【発明の効果】
以上説明したように、本発明によれば、直交復調時に生じる位相誤差、振幅誤差を精度良く補償することが可能となる。
【図面の簡単な説明】
【図１】本発明の第１の実施形態である直交復調誤差補償回路を示すブロック図である。
【図２】本発明の第２の実施形態である直交復調誤差補償回路を示すブロック図である。
【図３】本発明の第３の実施形態である直交復調誤差補償回路を示すブロック図である。
【図４】従来の直交復調器を示すブロック図である。
【符号の説明】
１：受信信号入力端子
２，３：復調信号出力端子
４，５：変調信号入力端子
６：送信信号出力端子
７，３０：スイッチ
８：直交復調器
９，１０：ミキサ
１１：局部発振器
１２：９０度移相器
１３，１４：ローパスフィルタ
１５，１６：平均値検出回路
１７，１８：加算器
１９：２乗平均・演算回路
２０：位相・振幅補償回路
２１，２２：乗算器
２３：加算器
２４：直交変調器
２５，２６：ミキサ
２７：局部発振器
２８：９０度移相器
２９：加算器
３１，３２：Ａ／Ｄ変換器
３３：９０度移相器
３４：発振器
３５：ミキサ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method and a circuit for compensating for a phase error and an amplitude error of a quadrature demodulator used in a transmission / reception apparatus.
[0002]
[Prior art]
The quadrature demodulator is a demodulator that extracts a demodulated signal having a phase difference of 90 degrees from a modulated signal in a wireless communication receiver. A conventional quadrature demodulator is shown in FIG. In FIG. 4, the modulated signal input from the input terminal 1 is input to the quadrature demodulator 8. This input signal is divided into two, and one signal is multiplied by the output signal of the local oscillator 11 in the mixer 9, the high frequency component is removed by the low-pass filter 13 and output to the output terminal 2 as a baseband in-phase signal. The other divided signal is input to the mixer 10, the output signal of the local oscillator 11 is multiplied by the signal whose phase is changed by 90 degrees by the 90-degree phase shifter 12, and the high-frequency component is removed by the low-pass filter 14. And output to the output terminal 3 as a baseband orthogonal signal. At this time, if there is a phase error or amplitude error from 90 degrees between the output signal of the local oscillator 11 input to the mixer 9 and the output signal of the 90-degree phase shifter 12 input to the mixer 10, the demodulation characteristics Deteriorates.
[0003]
As a circuit that functions in the same manner as the 90-degree phase shifter 12, a 90-degree distribution circuit that distributes the output signal of the local oscillator 11 into two signals having a phase difference of 90 degrees is well known. If there is a phase error or amplitude error from 90 degrees of the two output signals of the degree distribution circuit, the demodulation characteristics deteriorate.
[0004]
Conventionally, Patent Document 1 describes compensation for phase error and amplitude error in a quadrature modulator, but not for a quadrature demodulator.
[0005]
[Patent Document 1]
Japanese Patent No. 3144649 Specification
[Problems to be solved by the invention]
As described above, the conventional quadrature demodulator has a problem that the demodulation characteristic deteriorates if there is a phase error or amplitude error that occurs during quadrature demodulation.
[0007]
The present invention has been made to solve such problems, and an object thereof is to provide a quadrature demodulation error compensation method and circuit capable of accurately compensating for these phase error and amplitude error.
[0008]
[Means for Solving the Problems]
The invention according to claim 1 is a transmission / reception apparatus comprising: a quadrature demodulator that inputs a modulated signal and outputs a baseband signal; and a quadrature modulator that inputs a baseband signal and outputs a modulated signal. A quadrature demodulation error compensation method, wherein when a phase error and an amplitude error of the quadrature demodulator are detected, a known signal is supplied to the quadrature modulator and an output signal obtained from the quadrature modulator is input to the quadrature demodulator An average value of the baseband in-phase output signal of the quadrature demodulator is obtained as a first DC offset, an average value of the baseband quadrature output signal is obtained as a second DC offset, and the first DC offset is obtained as the baseband in-phase output signal. with subtracted from subtracting the second 2DC offset from the baseband quadrature output signal, the quadrature demodulating the plurality of known signals When the input to each of the first 1DC respective offset is subtracted baseband in-phase output signal and the second 2DC offset each baseband quadrature output signal subtracted obtained by averaging in terms of squared respectively A coefficient for compensating for the phase error and amplitude error of the quadrature demodulator is obtained from the mean square value of the quadrature demodulator, and when receiving a normal signal, the first DC offset is subtracted from the baseband in-phase output signal of the quadrature demodulator and the baseband quadrature is obtained. Subtract the second DC offset from the output signal, and the baseband in-phase output signal from which the first DC offset is subtracted or the baseband quadrature output signal from which the second DC offset is subtracted, the phase error and the amplitude by the coefficient. Orthogonal reconstruction characterized by performing error compensation processing It was an error compensation method.
[0009]
According to a second aspect of the present invention, in the quadrature demodulation error compensation method according to the first aspect, as the known signal to be input to the quadrature modulator, a period in which a baseband in-phase signal is alternately repeated 1 and 0 A first known signal that is a signal and the baseband quadrature signal is always 0, and a second known signal that is a periodic signal in which the baseband in-phase signal is always 0 and the baseband quadrature signal is alternately repeated 1 and 0 A third known signal that is a periodic signal in which the baseband in-phase signal and the baseband quadrature signal have the same 1 and 0 alternately repeated, and the quadrature demodulator receives the periodic signal in which 1 and 0 are alternately repeated. The baseband in-phase output signal and baseband quadrature output signal obtained as baseband signals are averaged over the period of one period or more of the periodic signal, respectively. Seek command phase average [delta] _I and baseband quadrature average [delta] _Q, subtracting the baseband in-phase average value [delta] _I from the baseband in-phase output signal of the quadrature demodulator when using the first known signal, the A first baseband in-phase mean square value PI _dc1 (t) obtained by squaring the subtracted value over a period of one period or more of the periodic signal of the first known signal is obtained, and the second known signal is obtained. The baseband in-phase average value δ _I is subtracted from the baseband in-phase output signal of the quadrature demodulator using the quadrature demodulator, and this subtracted value is equal to or more than one period of the periodic signal of the second known signal. A second baseband in-phase mean square value PI _dc2 (t) obtained by squaring the time is obtained, and the first baseband in-phase mean square value PI _dc1 (t) and the second baseband in-phase mean square value PI _dc2 (t) constant by adding the seeking k, the third known signal and subtracting the quadrature demodulator baseband in-phase output signal from the baseband in-phase average value [delta] _I when using, one period of the subtracted value and the third known signal _{Alternatively, a} third baseband in-phase mean square value PI _dc3 (t) obtained by squaring the mean over a longer time is obtained, and the modulated signal is obtained from the third baseband in-phase mean square value PI _dc3 (t) and the constant k. of obtaining a phase difference α between the carrier and the local oscillator, subtracting the baseband quadrature average [delta] _Q from the baseband quadrature output signal of the quadrature demodulator when using the first known signal, the subtracted value When a first baseband quadrature mean square value PQ _dc1 (t) obtained by squaring over one period or more of the periodic signal of the first known signal is obtained, and the second known signal is used Base of the quadrature demodulator Command orthogonal subtracting the baseband quadrature average [delta] _Q from the output signal, the second baseband quadrature averaged squared in one period or more times of the periodic signal of the subtracted value and the second known signal A mean square value PQ _dc2 (t) is obtained, and a value obtained by adding the first baseband quadrature mean square value PQ _dc1 (t) and the second baseband quadrature mean square value PQ _dc2 (t) and the constant obtains an amplitude error G of the quadrature demodulator from k, the baseband quadrature output signal of the quadrature demodulator subtracts the baseband quadrature average [delta] _Q, the subtracted value obtained when using the third known signal A third baseband quadrature mean square value PQ _dc3 (t) obtained by squaring the mean of the third known signal for one period or more is obtained, and the third baseband quadrature mean square value PQ _dc3 ( t), the constant k and the phase difference α Obtaining a phase error φ of al the quadrature demodulator and the quadrature demodulator error compensating method characterized by obtaining the coefficients to compensate for this phase error φ and the amplitude error G.
[0010]
According to a third aspect of the present invention, in the quadrature demodulation error compensation method according to the first or second aspect, when the phase error and the amplitude error of the quadrature demodulator are detected, the known signal is supplied to the quadrature modulator. The quadrature demodulation error compensation method is characterized in that the output signal is frequency-converted to the same frequency as the reception signal input to the quadrature demodulator and input to the quadrature demodulator.
[0011]
The invention according to claim 4 is a transmission / reception apparatus comprising: a quadrature demodulator that inputs a modulated signal and outputs a baseband signal; and a quadrature modulator that inputs a baseband signal and outputs a modulated signal. A quadrature demodulation error correction circuit for detecting a mean value of a known baseband in-phase output signal of the quadrature demodulator as a first DC offset; and a known baseband quadrature output of the quadrature demodulator A second average value detection circuit that detects an average value of the signal as a second DC offset; a first adder that subtracts the first DC offset from a baseband in-phase output signal of the quadrature demodulator; and a baseband quadrature of the quadrature demodulator. a second adder subtracting said first 2DC offset from the output signal, when inputting a plurality of known signals to the quadrature demodulator, output from the first adder Each from each of the square mean value of the respective mean square value and the respective baseband quadrature output signal output from the second adder of the baseband in-phase output signal of the quadrature demodulator phase error and amplitude error to A mean-arithmetic / arithmetic circuit for obtaining a coefficient to compensate for, and a baseband in-phase output signal or second summing output from the first adder when a normal signal is received by the coefficient obtained by the mean-square / arithmetic circuit And a phase / amplitude compensation circuit for removing the amplitude error and the phase error with respect to the baseband quadrature signal output from the detector.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
[First Embodiment]
FIG. 1 is a block diagram showing an orthogonal demodulation error compensation circuit according to the first embodiment. In FIG. 1, 1 is a reception signal input terminal, 2 and 3 are demodulation signal output terminals, 4 and 5 are modulation signal input terminals, 6 is a transmission signal output terminal, 7 and 30 are switches, 8 is a quadrature demodulator, 14 is a low-pass filter, 15 and 16 are average value detection circuits, 17 and 18 are adders, 19 is a mean square / arithmetic circuit, 20 is a phase / amplitude compensation circuit, and 24 is a quadrature modulator.
[0013]
The quadrature demodulator 8 includes

mixers

9 and 10, a local oscillator 11, and a 90-degree phase shifter 12. The phase / amplitude compensation circuit 20 includes multipliers 21 and 22 and an adder 23. The quadrature modulator 24 includes

mixers

25 and 26, a local oscillator 27, a 90-degree phase shifter 28, and an adder 29.
[0014]
At the time of transmission, the output terminal of the quadrature modulator 24 and the transmission signal output terminal 6 are connected by the switch 30. During reception, the received signal input terminal 1 and the input terminal of the quadrature demodulator 8 are connected by a switch 7. When the phase error and amplitude error of the quadrature demodulator 8 are detected, the output terminal of the quadrature modulator 24 and the input terminal of the quadrature demodulator 8 are connected by the switch 30 and the switch 7.
[0015]
A method for compensating the phase error and amplitude error of the quadrature modulator is disclosed in Japanese Patent No. 3144649. Thus, the phase error and amplitude error of the quadrature modulator 24 are first compensated, and the phase error and amplitude error of the quadrature demodulator 8 are detected and compensated using the compensated modulated signal.
[0016]
Since the phase error and amplitude error of the quadrature demodulator 8 are frequency dependent, the frequency of the modulated signal generated for error detection is the same as that of the received signal. If this angular frequency is ω, a signal input to the modulation signal input terminal 4 is I _T (t), and a signal input to the modulation signal input terminal 5 is Q _T (t), the signal output from the quadrature modulator 24 is output. The modulation signal S (t) is

It becomes.
[0017]
This signal S (t) is input to the quadrature demodulator 8 via the switch 30 and the switch 7. A signal input to the mixer 9 from the local oscillator 11 in the quadrature demodulator 8 can be written as cos (ωt + α), and a signal input from the local oscillator 11 to the mixer 10 via the phase shifter 12 can be written as Gsin (ωt + α + φ). . Where G is the amplitude error between the signal passing through the 90-degree phase shifter 12 and the signal not passing through, φ is the phase error between the signal passing through the 90-degree phase shifter 12 and the signal not passing through, and α is the modulated signal S. This is the phase difference between the carrier wave of (t) and the local oscillator 11.
[0018]
Therefore, after being output from the quadrature demodulator 8, the baseband in-phase output signal I _R (t) obtained through the low-pass filter 13 and the baseband quadrature output signal Q _R (t) obtained through the low-pass filter 14 are obtained. Is

It becomes. Where k is a constant, δ _I, δ _Q is a DC offset. In general, the output of the active element includes a DC offset, but the output signal of the quadrature demodulator 8 is a baseband signal and includes a DC component as a signal. Therefore, especially when demodulating a low-speed signal, the DC offset cannot be removed by C (capacity) cut.
[0019]
Here, if the signal I _T (t) input to the quadrature modulator 24 is a signal in which 1 and 0 appear at the same frequency, I _R (t) in the equation (2) is input to the average value detection circuit 15 and averaged. phased value is a DC offset [delta] _I. Similarly, when the signal Q _T (t) input to the quadrature modulator 24 is a signal in which 1 and 0 appear at the same frequency, Q _R (t) in the equation (3) is input to the average value detection circuit 16 and averaged. value is the DC offset δ _Q. At this time, if the signals I _T (t) and Q _T (t) input to the quadrature modulator 24 are periodic signals in which 1 and 0 are alternately repeated, the average time of the average

value detection circuits

15 and 16 is the periodic signal. One cycle is sufficient. Therefore, it is possible to shorten the average time compared to the case of detecting the DC offset from the unknown received signal.
[0020]
The signals I _dc (t) and Q _dc (t) obtained by subtracting the detected DC offsets δ _I and δ _Q by the

adders

17 and 18 are

It becomes. Equation (5) is

Can be transformed into
[0021]
When there is no amplitude error and no phase error, G = 1 and φ = 0, and I _dc (t) in the equation (4) can be taken out as an ideal in-phase signal as it is. On the other hand, if Q _cor (t) is a signal obtained by removing the amplitude error and the phase error from Q _dc (t) in equation (5), G = 1, φ = 0,

It becomes. Using Equations (4) and (7), Equation (6) is

It becomes.
[0022]
Therefore, in a range where φ is φ ≠ nπ (rad) (where n is an integer including 0), Q _cor (t) is

Can be given by
[0023]
That is, an ideal orthogonal signal from which the amplitude error G and the phase error φ are removed can be obtained by calculating the amplitude error G and the phase error φ and performing the calculation of the equation (9) on the signal from which the DC offset is removed. it can.
[0024]
Next, in order to obtain the amplitude error G, the phase error φ, and the constant k, predetermined signals are input to the quadrature modulator 24 as I _T (t) and Q _T (t).
[0025]
First, I _T (t) is a periodic signal in which 1 and 0 are alternately repeated, and Q _T (t) is always 0. This is defined as a modulation signal 1 (first known signal). When I _T (t) = 1, the output I _dc (t) from the adder 17 is

It becomes. When I _T (t) = 0, I _dc (t) = 0. Therefore, the value obtained by averaging the square of the I _dc (t) in one period or longer periodic signal to be input as _{_{I T (t) PI dc1 (}} t) is

It becomes.
[0026]
Next, I _T (t) is always 0, and Q _T (t) is a periodic signal in which 1 and 0 are alternately repeated. This is referred to as a modulation signal 2 (second known signal). When Q _T (t) = 1, I _dc (t) is

It becomes. When Q _T (t) = 0, I _dc (t) = 0. Therefore, the value PI _dc2 (t) obtained by squaring I _dc (t) over a period of one period or more of the periodic signal input as Q _T (t) is

It becomes.
[0027]
From the equations (11) and (13),

Is obtained. Therefore,

Thus, a constant k is obtained.
[0028]
Next, let I _T (t) and Q _T (t) be periodic signals in which the same 1 and 0 are alternately repeated. This is defined as a modulation signal 3 (third known signal). When I _T (t) = 0 and Q _T (t) = 1, from equation (4), I _dc (t) is

It becomes. When I _T (t) = 0 and Q _T (t) = 0, I _dc (t) = 0. Therefore, a value PI _dc3 (t) obtained by averaging the I _dc (t) with a square for a period equal to or longer than one period of the periodic signal input as I _T (t) and Q _T (t) is

It becomes.
[0029]
Therefore, the phase error α is

The phase difference α can be obtained by substituting k in the equation (15).
[0030]
Next, attention is focused on the output Q _dc (t) of the adder 18 obtained with the same modulation signal. First, when the modulation signal 1, when the _{I T (t) = 1,} (5) Q dc (t) from expression

It becomes. When I _T (t) = 0, Q _dc (t) = 0. Therefore, a value PQ _dc1 (t) obtained by squaring Q _dc (t) over a period of one period or more of the periodic signal input as I _T (t) is

It becomes.
[0031]
Next, in the case of modulation signal 2, when Q _T (t) = 1, Q _dc (t) is

It becomes. When Q _T (t) = 0, I _dc (t) = 0. Therefore, a value PQ _dc2 (t) obtained by averaging the squares of Q _dc (t) over a period of one period or more of the periodic signal input as Q _T (t) is

It becomes.
[0032]
From the equations (20) and (22),

Is obtained. Therefore, the amplitude error G is

The amplitude error G can be obtained by substituting k in the equation (15).
[0033]
Next, in the case of the modulated signal 3, when I _T (t) = 1 and Q _T (t) = 1, Q _dc (t) is

It becomes. When I _T (t) = 0 and Q _T (t) = 0, I _dc (t) = 0. Therefore, the value PQ _dc3 (t) obtained by averaging the squares of Q _dc (t) over a period of one period or more of the periodic signal input as I _T (t) and Q _T (t) is

It becomes. Therefore, the phase error φ is

By substituting k in equation (15), α in equation (18), and G in equation (24), phase error φ can be obtained.
[0034]
The operations from the above equations (10) to (27) are performed by the root mean square / arithmetic circuit 19, tanφ and Gcosφ in the equation (9) are obtained, and the multiplier 21 in the phase /

amplitude compensation circuit

20 and 22 are input as coefficients.
[0035]
In this state, the received signal input terminal 1 and the input terminal of the quadrature demodulator 8 are connected by the switch 7 to output an ideal quadrature signal from which the DC offset, amplitude error and phase error have been removed from the received signal. 3 can be obtained. An ideal in-phase signal from which the DC offset is removed is output to the output terminal 2.
[0036]
In the present embodiment, since the phase error and amplitude error of the quadrature demodulator 8 can be detected periodically using the time not received, the phase error and amplitude error generated during quadrature demodulation can be compensated with high accuracy. It becomes possible.
[0037]
[Second Embodiment]
FIG. 2 is a block diagram showing an orthogonal demodulation error compensation circuit according to the second embodiment. The same parts as those in FIG. 1 are given the same reference numerals. In this embodiment, the output of the first low-pass filter 13 is input to the first A / D converter 31, the output of the second low-pass filter 14 is input to the second A / D converter 32, and After conversion to a digital signal, average value detection, squaring average / calculation, and phase / amplitude compensation are performed. By performing these by digital signal processing, the operation can be performed with high accuracy. The A / D converter may be placed before the quadrature demodulator 8 and the quadrature demodulator may be a digital circuit.
[0038]
Further, the output signals of the local oscillator 11 are input to the

mixers

9 and 10 in the same phase, and the 90-degree phase shifter 33 gives a 90-degree phase difference to the modulated signals input from the switch 7 to the

mixers

9 and 10. . As described above, whether the 90 degree phase is given is the same as the output of the quadrature demodulator 8 regardless of whether it is a local oscillation output signal or a modulated signal, and the amplitude error and phase error of the 90 degree phase shifter are the same. Can be compensated.
[0039]
[Third Embodiment]
FIG. 3 is a block diagram showing an orthogonal demodulation error compensation circuit according to the third embodiment. The same parts as those in FIG. 1 are given the same reference numerals. This embodiment is effective when the frequency of the output signal of the quadrature modulator 24 and the frequency of the received signal input to the quadrature demodulator 8 are different. In this embodiment, an oscillator 34 and a mixer 35 for frequency conversion are provided between the switch 30 and the switch 7 to convert the frequency of the output signal of the quadrature modulator 24 into the frequency of the signal input from the reception signal input terminal 1. is doing. As a result, similarly to the first embodiment, the output signal of the quadrature modulator 24 can be used to detect the phase error and amplitude error of the quadrature demodulator 8.
[0040]
[Other Embodiments]
In the above first to third embodiments, the baseband quadrature output signal is compensated for the phase error and the amplitude error on the basis of the baseband in-phase output signal. As a reference, the same can be applied to the baseband in-phase output signal.
[0041]
【The invention's effect】
As described above, according to the present invention, it is possible to accurately compensate for phase errors and amplitude errors that occur during quadrature demodulation.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an orthogonal demodulation error compensation circuit according to a first embodiment of the present invention.
FIG. 2 is a block diagram showing an orthogonal demodulation error compensation circuit according to a second embodiment of the present invention.
FIG. 3 is a block diagram showing an orthogonal demodulation error compensation circuit according to a third embodiment of the present invention.
FIG. 4 is a block diagram showing a conventional quadrature demodulator.
[Explanation of symbols]
1: reception signal input terminals 2, 3: demodulation signal output terminal 4, 5: modulation signal input terminal 6: transmission signal output terminal 7, 30: switch 8: quadrature demodulator 9, 10: mixer 11: local oscillator 12: 90 Phase shifters 13 and 14: Low-pass filter 15 and 16: Average value detection circuit 17 and 18: Adder 19: Mean square / arithmetic circuit 20: Phase / amplitude compensation circuit 21 and 22: Multiplier 23: Adder 24 : Quadrature modulators 25 and 26: mixer 27: local oscillator 28: 90 degree phase shifter 29: adder 31 and 32: A / D converter 33: 90 degree phase shifter 34: oscillator 35: mixer

Claims

A quadrature demodulation error compensation method in a transmission / reception apparatus comprising: a quadrature demodulator that inputs a modulated signal and outputs a baseband signal; and a quadrature modulator that inputs a baseband signal and outputs a modulated signal. ,
When detecting a phase error and an amplitude error of the quadrature demodulator, a known signal is supplied to the quadrature modulator and an output signal obtained from the quadrature modulator is input to the quadrature demodulator,
An average value of the baseband in-phase output signal of the quadrature demodulator is obtained as a first DC offset, and an average value of the baseband quadrature output signal is obtained as a second DC offset, and the first DC offset is subtracted from the baseband in-phase output signal. And subtracting the second DC offset from the baseband quadrature output signal,
A plurality of known signal when the input to the quadrature demodulator, the first 1DC respective offset is subtracted baseband in-phase output signal and the second 2DC offset each baseband quadrature output signals subtracted respectively 2 A coefficient for compensating for the phase error and the amplitude error of the quadrature demodulator is obtained from each square average value obtained by averaging after multiplication,
During normal signal reception, the first DC offset is subtracted from the baseband in-phase output signal of the quadrature demodulator, and the second DC offset is subtracted from the baseband quadrature output signal.
The phase error and the amplitude error are compensated by the coefficients for the baseband in-phase output signal from which the first DC offset has been subtracted or the baseband quadrature output signal from which the second DC offset has been subtracted. An orthogonal demodulation error compensation method.

The orthogonal demodulation error compensation method according to claim 1,
As the known signal to be input to the quadrature modulator, a baseband in-phase signal is a periodic signal in which 1 and 0 are alternately repeated, and a first known signal in which the baseband quadrature signal is always 0, and the same baseband signal. The second known signal, which is a periodic signal in which the phase signal is always 0 and the baseband quadrature signal is alternately repeated 1 and 0, and the same 1 and 0 are repeated alternately in the baseband in-phase signal and the baseband quadrature signal. Prepare a third known signal that is a periodic signal,
A baseband in-phase output signal and a baseband quadrature output signal obtained by applying a periodic signal in which 1 and 0 are alternately repeated to the quadrature demodulator as a baseband signal, respectively, for a period of one period or more of the periodic signal. Average baseband in-phase average value δ _I and baseband quadrature average value δ _Q ,
The orthogonal subtracting the baseband in-phase average value [delta] _I from the baseband in-phase output signal of the demodulator, one period of the periodic signal of the subtracted value of the first known signal in the case of using the first known signal _Find the first baseband in-phase mean square value PI _dc1 (t), which is the mean square over minutes or longer,
The orthogonal subtracting the baseband in-phase average value [delta] _I from the baseband in-phase output signal of the demodulator, one period of the periodic signal of the subtracted value and the second known signal when using the second known signal _Find the second baseband in-phase mean square value PI _dc2 (t), which is the mean square over minutes or longer,
The first baseband in-phase mean square value PI _dc1 (t) and the second baseband in-phase mean square value PI _dc2 (t) are added to obtain a constant k,
The baseband in-phase average value δ _I is subtracted from the baseband in-phase output signal of the quadrature demodulator when the third known signal is used, and this subtracted value is equal to or longer than one period of the third known signal. _Find the 3rd baseband in-phase mean square value PI _dc3 (t) that is squared over time,
A phase difference α between the carrier wave of the modulated signal and the local oscillator is obtained from the third baseband in-phase mean square value PI _dc3 (t) and the constant k,
The orthogonal subtracting the baseband quadrature average [delta] _Q from the baseband quadrature output signal of the demodulator, one period of the periodic signal of the subtracted value of the first known signal in the case of using the first known signal _Find the first baseband quadrature mean square value PQ _dc1 (t), which is the mean square over minutes or longer,
The orthogonal subtracting the baseband quadrature average [delta] _Q baseband quadrature output signal of demodulator, one period of a periodic signal of the subtracted value and the second known signal when using the second known signal _Find the second baseband quadrature mean square value PQ _dc2 (t), which is the mean square over minutes or longer,
An amplitude error G of the quadrature demodulator is obtained from a value obtained by adding the first baseband quadrature mean square value PQ _dc1 (t) and the second baseband quadrature mean square value PQ _dc2 (t) and the constant k. ,
The orthogonal subtracting the baseband quadrature average [delta] _Q from the baseband quadrature output signal of the demodulator, one cycle or more of the subtracted value and the third known signal when using the third known signal The third baseband quadrature mean square value PQ _dc3 (t) obtained by taking the mean square over time is calculated,
A phase error φ of the quadrature demodulator is obtained from the third baseband quadrature mean square value PQ _dc3 (t), the constant k and the phase difference α;
A quadrature demodulation error compensation method, wherein a coefficient for compensating for the phase error φ and the amplitude error G is obtained.

The orthogonal demodulation error compensation method according to claim 1 or 2,
When detecting the phase error and amplitude error of the quadrature demodulator, the output signal obtained by applying the known signal to the quadrature modulator is frequency-converted to the same frequency as the received signal input to the quadrature demodulator. A quadrature demodulation error compensation method, wherein the quadrature demodulator is input to the quadrature demodulator.

A quadrature demodulation error correction circuit in a transmission / reception apparatus comprising: a quadrature demodulator that inputs a modulated signal and outputs a baseband signal; and a quadrature modulator that inputs a baseband signal and outputs a modulated signal. ,
A first average value detection circuit for detecting an average value of a known baseband in-phase output signal of the quadrature demodulator as a first DC offset;
A second average value detection circuit for detecting an average value of known baseband quadrature output signals of the quadrature demodulator as a second DC offset;
A first adder for subtracting the first DC offset from the baseband in-phase output signal of the quadrature demodulator;
A second adder for subtracting the second DC offset from the baseband quadrature output signal of the quadrature demodulator;
When you enter multiple known signal to the quadrature demodulator, each of the base to be output from each of the mean square value and the second adder of each baseband in-phase output signal output from said first adder A mean-square / arithmetic circuit for obtaining a coefficient for compensating the phase error and the amplitude error of the quadrature demodulator from the mean square value of each of the band quadrature output signals;
The amplitude obtained with respect to the baseband in-phase output signal output from the first adder or the baseband quadrature signal output from the second adder when receiving a normal signal is determined by the coefficient obtained by the root mean square / arithmetic circuit. A phase / amplitude compensation circuit for removing the error and the phase error;
An orthogonal demodulation error compensation circuit comprising: