JP3648164B2 - Transmitter, receiver and digital communication system using parallel concatenated convolutional coding - Google Patents

Description

【００１１】
α_k-1（ｓ_k-1）＝ｐ（ｓ_k-1，ｙ_j<k）は、ｋ−１の時点での状態がｓ_k-1である確率を、初期状態から再帰的に求めたものであり、β_k（ｓ_k）＝ｐ（ｙ_j>k｜ｓ_k）は、同じくｋの時点での状態がｓ_kである確率を、終了状態から求めたものである。
また、γ_k（ｓ_k-1，ｓ_k）は、ｋ−１時点での状態ｓ_k-1からｋ時点の常態ｓ_kへ遷移する確率で、
【数２】

となる。
【００１２】
これは、
【数３】

で求めることができる。
【００１３】
また、α_k（ｓ_k）およびβ_k（ｓ_k）は、
【数４】

【数５】

によりそれぞれ再帰的に求めることができる。
【００１４】
これらを用いて、Ｌ（ｕ_k）は、
【数６】

で求めることができる。
【００１５】
ここで、分子のΣ（ｓ＋）は、ｕ_k＝１となる場合の総和を示し、分母のΣ（ｓ−）は、ｕ_k＝−１となる場合の総和を示す。
さらに、次の組織畳み込み符号復号器に送られる外部尤度情報Ｌ^e（ｕ_k）はこの復号結果Ｌ（ｕ_k）から重み付けされた受信信号Ｌ_cｙ³ _kおよび、この復号器に入力された外部尤度情報Ｌ^e（ｕ_k）を引いて得られる。
【００１６】
【発明が解決しようとする課題】
誤り訂正符号は、情報系列を符号化して冗長ビットを付加することにより訂正能力を得るため、冗長ビットの付加により広い伝送帯域が必要となる。一般的には、冗長度の大きい（符号化率の小さい）符号化は伝送速度の低下を招くが、スペクトル拡散による符号分割多元接続(ＣＤＭＡ)との組合わせではもともと広帯域の信号を用いているので、帯域拡散に対してあまり考慮する必要がなかった。
【００１７】
しかしながら、誤り訂正技術や干渉除去技術、あるいはアダプティブアレーアンテナなどの空間信号処理の進歩に伴って、特定区域内に収容可能な符号の数が増加してくると、互いに相関特性の良い拡散符号の組み合わせが限られてくる。すなわち、ある特定数以上のユーザを収納するには、相互相関がある程度大きな組み合わせを用いざるをえない状況がある。
【００１８】
本発明は、上述した事情に鑑み提案されたもので、特定のユーザに対して相互相関のある複数の符号を用いて多重化して送受信する方法を用い、これらの符号間における相互相関値を用いて復号処理を行うことにより、干渉による劣化を低減することが可能な並列連接畳み込み符号化を用いた送信機、受信機およびデジタル通信システムを提供することを目的とする。
【００１９】
【課題を解決するための手段】
上述した目的を達成するため、本発明の並列連接畳み込み符号化を用いた受信機は、情報系列を並列連接畳み込み符号化して通信するデジタル通信方式に用いる受信機であって、
それぞれ異なる複数の拡散符号を用いて逆拡散を行うための逆拡散手段と、各拡散符号間の相互相関値を生成するための相互相関値生成手段と、送信機側で拡散前に行った逆インターリーバに対応するインターリーブ処理を行うためのインターリーバと、逆拡散およびインターリーブ処理を行った受信信号を入力してターボ復号処理を行うためのターボ復号手段とを備え、
前記ターボ復号手段では、前記相互相関値生成手段により得られた拡散符号間の相互相関を用いて復号処理を行うことを特徴とするものである。
【００２２】
ここで、相互相関値生成手段は、各拡散符号間の相互相関値を計算で求めたり、あるいは予め計算された相互相関値を記憶しておくものである。
【００２３】
本発明の並列連接畳み込み符号化を用いたデジタル通信システムは、前記送信機および前記受信機を用いて、前記受信機では、受信に際して、前記ターボ復号手段において、前記相互相関値生成手段により得られた拡散符号間の相互相関を用いて復号処理を行うことをものである。
【００２５】
また、本発明の並列連接畳み込み符号化を用いた受信機は、情報系列を並列連接畳み込み符号化して通信するデジタル通信方式に用いる受信機であって、それぞれ異なる複数の拡散符号を用いて逆拡散を行うための逆拡散手段と、パリティ部に対応する前記逆拡散手段の出力を第１パリティ部と第２パリティ部に分離するための手段と、拡散符号間の相互相関値に基づいて、組織部と前記第１パリティ部、および組織部と前記第２パリティ部の間の干渉量を生成するための干渉量生成手段と、送信機側で拡散前に行った逆インターリーバに対応するインターリーブ処理を行うためのインターリーバと、逆拡散およびインターリーブ処理を行った受信信号を入力してターボ復号処理を行うためのターボ復号手段とを備え、
前記ターボ復号手段では、前記干渉量生成手段により得られた干渉量を用いて復号処理を行うことを特徴とするものである。
【００２６】
ここで、干渉量生成手段は、前記拡散符号間の干渉量を計算で求めたり、あるいは予め計算された干渉量を記憶しておくものである。
【００２７】
また、本発明の並列連接畳み込み符号化を用いたデジタル通信システムは、前記送信機および前記受信機を用いて、前記受信機では、受信に際して、前記第１の複素拡散手段および前記第２の複素拡散手段において拡散変調された各信号を合成し、前記受信機では、送信に際して、前記ターボ復号手段において、前記干渉量生成手段により得られた干渉量を用いて復号処理を行うことを特徴とするものである。
【００２８】
【発明の実施の形態】
以下、本発明の並列連接畳み込み符号化を用いた送信機、受信機およびデジタル通信システムを、図面に示す具体的な実施例に基づいて説明する。
【００２９】
＜実施例１：送信機＞
図１は、本発明の並列連接畳み込み符号化を用いた送信機の具体的な実施例１を示すブロック図である。
実施例１の送信機は、図１に示すように、ターボ符号器４と、逆インターリーバ５と、乗算器６，７，８，１０と、加算器９と、無線搬送波発生器１１とからなる。ターボ符号器４は、再帰的組織畳み込み符号器（ＲＳＣＣ）１，２と、インターリーバ３とからなる。再帰的組織畳み込み符号器は、１を第１の符号器、２を第２の符号器とする。
【００３０】
この送信機１は、ターボ符号器４の出力のうち、入力データと同じ部分を組織部（ｉ）、第１の符号器１（ＲＳＣＣ１）で付加されるパリティ（ｐ）を第１パリティ部、第２の符号器２（ＲＳＣＣ２）で付加されるパリティ（ｐ）を第２パリティ部とする。なお、入力データの情報ビット系列をインターリーバ３によりインターリーブして得られる系列が、第２の符号器２（ＲＳＣＣ２）に入力される。
【００３１】
第２の符号器２で符号化された第２パリティ部は、逆インターリーバ５を通過することにより並び替えが行われる。そして、組織部、第１パリティ部および第２パリティ部は、乗算器６，７，８において、それぞれ異なる拡散符号系列であるＰＮ１，ＰＮ２，ＰＮ３との積をとって拡散変調され、加算器９により合成され、さらに乗算器１０において、無線搬送波発生器１１からの搬送波が乗算されて、キャリア変調がなされて送信される。
【００３２】
なお、ターボ符号器４は、上述した従来のターボ符号器（図５参照）と同様の構成であってもよい。また、さらに、第３パリティ部、第４パリティ部を含んで構成することも可能である。
【００３３】
＜実施例１：受信機＞
図２は、本発明の並列連接畳み込み符号化を用いた受信機の具体的な実施例１を示すブロック図である。
実施例１の受信機は、乗算器１２，１５，１６，１７と、無線搬送波発生器１３と、ローパスフィルタ（ＬＰＦ）１４と、逆拡散部１８，１９，２０と、ターボ復号器３０とからなる。ターボ復号器３０は、第１の復号器３１と、第２の復号器３２と、インターリーバ３３と、逆インターリーバ３４，３５，３６とからなる。
【００３４】
この受信機は、乗算器１２において、受信信号に対して無線搬送波発生器１３からの搬送波を乗算してキャリア変調し、ローパスフィルタ（ＬＰＦ）１４を通過させた後、逆拡散部１１８，１１９，１２０で逆拡散する。逆拡散部１１８，１１９，１２０は、乗算器１５，１６，１７において、受信信号にそれぞれの拡散符号ＰＮ１，ＰＮ２，ＰＮ３を乗算して、積分ダンプフィルタ１８，１９，２０を通過させることにより逆拡散させる。この際、それぞれの信号には、相互相関の大きさによって干渉成分が残る。
【００３５】
逆拡散された信号は、それぞれ組織部、第１パリティ部、第２パリティ部としてターボ復号器３０に送られる。ここで、組織部の送信系列をＸ（ｎ）、第１パリティ部の送信系列をＹ₁（ｎ）、第２パリティ部の送信系列をＹ₂（ｎ）とし、ＰＮ１，ＰＮ２，ＰＮ３のそれぞれで逆拡散した受信シンボルを、それぞｒ₁（ｎ），ｒ₂（ｎ），ｒ₃（ｎ）とすると、
ｒ₁（ｎ）＝Ｘ（ｎ）＋θ_1,2（ｎ）Ｙ₁（ｎ）＋θ_1,3（ｎ）Ｙ₂（ｎ）＋Ｎ₁（ｎ）
ｒ₂（ｎ）＝θ_1,2（ｎ）Ｘ（ｎ）＋Ｙ₁（ｎ）＋Ｎ₂（ｎ）
ｒ₃（ｎ）＝θ_1,3（ｎ）Ｘ（ｎ）＋Ｙ₂（ｎ）＋Ｎ₃（ｎ）
となる。
【００３６】
ここで、Ｎ₁，Ｎ₂，Ｎ₃は、雑音成分である。また、ここでは、ＰＮ２とＰＮ３は、互いに直交する系列としている。例えば、ＰＮ１を生成多項式１＋ｘ⁷＋ｘ¹⁸で生成されるｍ系列、ＰＮ２を生成多項式１＋ｘ⁵＋ｘ⁷＋ｘ¹⁰＋ｘ¹⁸で生成されるｍ系列とするとき、ＰＮ３をＰＮ２に対して1，−１，１，−１，…を乗じて得られる系列とすると、ＰＮ２とＰＮ３は２の巾乗の拡散率において直交する。
【００３７】
ここでは、相互相関値θ_1,2（ｎ）およびθ_1,3（ｎ）を用いて、遷移確率を求めるために、上記式(3)を次式のように変形する。
【数７】

ただし、θとして、ＲＳＣＣ１の復号の際はθ_1,2を用い、ＲＳＣＣ２の復号の際はθ_1,3を用いる。
【００３８】
ｒ₁（ｎ）は第１の復号器３１と逆インターリーバ３５に入力され、ｒ₂（ｎ）は第１の復号器３１に入力され、ｒ₃（ｎ）は逆インターリーバ３６に入力される。第１の復号器３１は、ｒ₁（ｎ）、ｒ₂（ｎ）及び相互相関値から、外部尤度情報Ｌｅ⁽¹⁾（ｕ）と復号結果Ｌ（ｕ）を出力する。ここで、相互相関値は、ｒ₁（ｎ）とｒ₂（ｎ）との相互相関を取ったもので、図示していない相互相関生成部により求められたものである。
外部尤度情報Ｌｅ⁽¹⁾（ｕ）はインターリーブ３３により並べ替えられて第２の復号器に入力される。第２の復号器３２は、外部尤度情報Ｌｅ⁽¹⁾（ｕ）、及び逆インターリーブされたｒ₁（ｎ）とｒ₃（ｎ）により、外部尤度情報Ｌｅ⁽²⁾（ｕ）と復号結果Ｌ（ｕ）を出力する。
外部尤度情報Ｌｅ⁽²⁾（ｕ）は、逆インターリーバ３４により逆インターリーブされて再び第１の復号器３１に入力される。
【００３９】
ここでは、ＰＮ２とＰＮ３が直交するように拡散符号を生成したが、これは必ずしも必要条件ではなく、ＰＮ１，ＰＮ２，ＰＮ３に対して、それぞれが異なるランダムな系列であってもかまわない。
【００４０】
＜実施例２：送信機＞
第１パリティ部と第２パリティ部を直交させる方法としては、ＱＰＳＫ変調を用いて、それぞれをＩ層、Ｑ層に割り当てる方法を採用することもできる。この際、組織部は情報変調をＢＰＳＫとして、複素拡数系列を用いたＱＰＳＫ拡散変調を行うことによって、第１パリティ部と第２パリティ部に対して均等に相互相関が生じるようになる。
【００４１】
図３は、本発明の並列連接畳み込み符号化を用いた送信機の具体的な実施例２を示すブロック図である。なお、上述した実施例１の送信機と同様の機能を有する部分には、同一の符号を付して説明を行う。
実施例２の送信機は、図３に示すように、データ系列をターボ符号化して組織部、第１パリティ部、第２パリティ部に分離し、第２パリティ部を逆インターリーバ５に通すところまでは、上述した実施例１と同様である。
【００４２】
その後、組織部は、拡散変調部６において、複素拡散符号ＰＮ１で拡散変調され、第１パリティ部および第２パリティ部は、マッピング部２１において複素平面上にマッピングされた後、乗算器７において、異なる複素拡散符号ＰＮ２で拡散変調され、加算器９において組織部の信号と合成される。さらに、乗算器１０において、無線搬送波発生器１１からの無線搬送波が乗算されてキャリア変調がなされ、送信される。
【００４３】
＜実施例２：受信機＞
図４は、本発明の並列連接畳み込み符号化を用いた受信機の具体的な実施例１を示すブロック図である。なお、上述した実施例１の受信機と同様の機能を有する部分には、同一の符号を付して説明を行う。
【００４４】
実施例２の受信機は、図４に示すように、乗算器１２において、受信信号に対して無線搬送波発生器１３からの搬送波を乗算してキャリア変調し、ローパスフィルタ（ＬＰＦ）１４を通過させた後、逆拡散部１２２，１２３で逆拡散する。逆拡散部１２２，１２３は、乗算器１５，１６において、それぞれの拡散符号ＰＮ１，ＰＮ２を乗算して、積分ダンプフィルタ２２，２３を通過させることにより逆拡散させる。このように受信信号はそれぞれの拡散符号を用いて逆拡散される。
【００４５】
この際、組織部の成分は、ＰＮ１で逆拡散した後、実数抽出部２４により実数部として抽出する。また、第１パリティ部は、ＰＮ２で逆拡散した信号から実数抽出部２５により実数部として取り出す。また、第２パリティ部は、ＰＮ２で逆拡散した信号から虚数抽出部２６から虚数部成分として取り出す。
ここで、完全な同期検波を仮定すると、第１パリティ部と第２パリティ部の間には干渉成分は生じない。しかし、組織部との間には、それぞれ、ＰＮ１とＰＮ２の相互相関に対応して干渉が生じる。
【００４６】
ＰＮ１（ｕ＋ｊｖ）とし、ＰＮ２を（ｒ＋ｊｓ）で表すと、系列間の相互相関は以下のように求まる。ただし、ｊは、虚数単位（ｊ＝（−１）^1/2）である。
（ｕ＋ｊｖ）（ｒ−ｊｓ）＝（ｕｒ＋ｖｓ）＋ｊ（ｖｒ−ｕｓ）
すなわち、組織部と第１パリティ部の間の干渉量は、ｕｒ＋ｖｓとなり、組織部と第２パリティ部の間の干渉量は、ｖｒ−ｕｓとなる。この値は図示しない干渉量生成部により求められる。
【００４７】
そして、逆拡散され、分離されたそれぞれの成分はターボ復号器３０に送られる。ここで得られた干渉量を、上述した実施例１における相互相関値の代りに用いて復号を行う。
ここで、ＰＮ２は、複素数値系列であるとしたが、これは実数値系列であってもかまわない。
【００４８】
＜誤り特性＞
次に、上述した実施例１の構成における誤り率特性を説明する。
なお、ターボ符号は、拘束長を「３」、ブロック長を「１０００」とする。また、繰返し復号の回数は、「６」とする。また、変調方式は、情報変調、拡散変調ともにＢＰＳＫで行い、拡散率は「４」、拡散符号は「２¹⁸−１」の周期のＭ系列をフレーム長でトランケートしている。また、通信路は、スタティックである。
【００４９】
図５に、誤り率特性の比較結果を示す。
比較対象は、実施例１の構成と同様な符号化と符号多重を行った信号に対して、従来のＭＡＰによる復号処理を行った場合の特性である。
図５から明らかなように、実施例１の構成によれば、従来の復号方式と比較して、より誤り率特性が改善することがわかる。
【００５０】
【発明の効果】
本発明の並列連接畳み込み符号化を用いた送信機、受信機およびデジタル通信システムは、上述した構成を備えているため、以下の効果を奏することができる。すなわち、本発明の並列連接畳み込み符号化を用いた送信機では、畳み込み符号化処理を行う第１の畳み込み符号器と、第２の畳み込み符号器とを、所定規則に基づいて入力データの並び替え操作を行うインターリーバを介して並列に接続した連接畳み込み符号器により、情報信号をターボ符号化した後に、第２の畳み込み符号器から得られる第２のパリティ系列を前記インターリーバの逆操作を行う逆インターリーバを用いて並び替えを行い、入力データと同一系列となる組織部、第１の畳み込み符号器から出力される第１のパリティ部、逆インターリーバから出力される第２のパリティ部をそれぞれ第１の拡散符号系列、第２の拡散符号系列、および第３の拡散符号系列で拡散処理し、多重化して送信する。
【００５１】
また、本発明の並列連接畳み込み符号化を用いた受信機では、前記それぞれの拡散符号系列を用いて逆拡散処理を行うことにより得られる、干渉成分を含む受信シンボルに対して、干渉成分の大きさを表す各拡散符号間の相互相関値を用いて繰り返し復号処理を行う。
【００５２】
また、本発明の並列連接畳み込み符号化を用いたデジタル通信システムは、前記送信機および前記受信機を用いてデジタル通信を行う。
したがって、干渉による誤りの発生を低減することができ、高速かつ高品質に通信を行うことが可能となる。
【図面の簡単な説明】
【図１】本発明の並列連接畳み込み符号化を用いた送信機の具体的な実施例１を示すブロック図である。
【図２】本発明の並列連接畳み込み符号化を用いた受信機の具体的な実施例１を示すブロック図である。
【図３】本発明の並列連接畳み込み符号化を用いた送信機の具体的な実施例２を示すブロック図である。
【図４】本発明の並列連接畳み込み符号化を用いた受信機の具体的な実施例２を示すブロック図である。
【図５】実施例１の送信機および受信機を用いた場合の誤り率特性の説明図である。
【図６】従来のターボ符号の符号器のブロック図である。
【図７】従来のターボ符号の復号器のブロック図である。
【符号の説明】
１ 第１の符号器（ＲＳＣＣ１）
２ 第２の符号器（ＲＳＣＣ２）
３ インターリーバ
４ ターボ符号器
５ 逆インターリーバ
６，７，８ 拡散変調部
９ 合成部
１０，１２ キャリア変調部
１１，１３ 無線搬送波発生器
１４ ローパスフィルタ（ＬＰＦ）
１５，１６，１７ 逆拡散部
１８，２０ ターボ復号器
１９ マッピング部
５１ 第１の符号器（ＲＳＣＣ１）
５２ 第２の符号器（ＲＳＣＣ２）
５３ インターリーバ
６１ 第１の符号器（ＲＳＣＣ１）
６２ 第２の符号器（ＲＳＣＣ２）
６３ インターリーバ
６４ 逆インターリーバ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a transmitter, a receiver, and a digital communication system used in the field of digital communication using an error correction technique, and more particularly, to a transmitter, a receiver, and a digital communication system using parallel concatenated convolutional coding.
[0002]
[Prior art]
In recent years, digital communication industries such as the Internet and digital mobile phones have been rapidly developed. In such a technical field, basic research for performing high-speed and high-quality communication is actively conducted when important information is communicated. In particular, the error correction technique is an indispensable technique for performing high-quality communication in an unstable communication channel such as digital mobile communication.
In error correction technology, a turbo code is an excellent error correction code that enables high-quality data communication with less energy, and has attracted widespread attention in recent years.
[0003]
Hereinafter, the turbo code will be described.
Parallel concatenated convolutional codes, so-called turbo codes, are constructed by combining two (or more) systematic convolutional codes (C. Berrou et al., "Near Shnannon limit error-correcting coding: Turbo codes", Procedings of ICC93 or "Near Optimum error-correcting coding: Turbo codes" IEEE Transactions on Communications, Vol. 44, no. 10, 1996).
[0004]
The turbo code decoding process is characterized in that errors are reduced by sequentially repeating the decoding of these two systematic convolutional codes. Here, the configuration and decoding method of the turbo encoder and decoder will be briefly described. For details, see Isaka et al., “Signpost to Shannon Limit:“ pralle1 concated (Turbo) coding ”,“ Turbo (iterative) decoding ”and its surroundings”, IEICE Technical Report IT98-51, J. Hagenauer, "Iterative Decoding of Binary Block and Convolutional Codes", IEEE Transactions on Information Theory, Vol.42, N0.2, 1996. Etc. are described.
[0005]
FIG. 6 shows a block diagram of a turbo code encoder, and FIG. 7 shows a block diagram of a turbo code decoder.
In the turbo code encoder, as shown in FIG. 6A, first and second recursive systematic convolutional encoders 51 and 52 (RSCC 1 and 2) are arranged in parallel, and an interleaver 53 is arranged therebetween. Has been. In this turbo encoder, a sequence obtained by interleaving the original information bit sequence i is input to the second encoder 52.
[0006]
The recursive systematic convolutional encoders 51 and 52 include shift registers 54 and 55 and adders (exclusive ORs) 56 and 57 as shown in FIG. The polynomial G (D) in the parity p generated by the encoders 51 and 52 is
G (D) = [1 (D ² +1) / (D ² + D + 1)]
It becomes.
Assuming that the coding rate of the recursive systematic convolutional encoders 51 and 52 is ½, the parity p obtained from each encoder is added to the original information bit i to obtain the coding rate. A 1/3 turbo code is obtained.
[0007]
As shown in FIG. 7, the turbo code decoder includes a first decoder 61 and a second decoder 62 arranged in series, and an interleaver 63 and a deinterleaver 64 arranged therebetween. The function of the turbo code decoder will be briefly described. After the received system is decoded by the first decoder 61 with the first systematic convolutional code, the decoding result and the reliability information are output. The second decoder 62 decodes the second systematic convolutional code using the reliability information from the first decoder 61 and the received sequence. Furthermore, using this result, the first decoder 61 performs decoding of the first systematic convolutional code again. By performing iterative decoding in this way, the reliability of decoded symbols gradually increases and errors are reduced.
This is because the interleaver 63 performs processing in the second decoder 62 in accordance with the processing order in the second encoder 52. Further, the deinterleaver 64 is for returning the rearrangement of the interleaver 63 to the reliability information output from the second decoder 62 and inputting it to the first decoder.
[0008]
In the decoding of the systematic convolutional code performed in the middle of the turbo code, a decoding result to which soft decision information is added is necessary for use in the subsequent decoding. The decoder 61 corresponding to the RSCC 1 uses a received symbol weighted in proportion to the SNR (L _c y) and external information as input signals, and a decoding result L (u) with a log likelihood ratio. And external likelihood information Le ⁽¹⁾ (u) to be passed to the next decoder.
[0009]
There are several decoding methods for obtaining the soft decision result. Here, MAP decoding will be described.
MAP (Maximum A Posteriori Probability: maximum a posteriori probability) decoding, under conditions that the received signal sequence y is given, for each information symbol u _k, seeks as a decoding result u _k that maximizes P (u _k) It is.
For this purpose, an LLR (Log Likelihood Ratio) L (u _k ) is _obtained based on the following equation.
[0010]
[Expression 1]

[0011]
α _k−1 (s _k−1 ) = p (s _k−1 , y _{j <k} ) recursively obtains the probability that the state at the time point _k−1 is s _k−1 from the initial state. Β _k (s _k ) = p (y _{j> k} | s _k ) is also obtained from the end state for the probability that the state at time _k is s _k .
Γ _k (s _k−1 , s _k ) is a probability of transition from the state s _k−1 at the time point _k−1 to the normal state s k at the time point _k .
[Expression 2]

It becomes.
[0012]
this is,
[Equation 3]

Can be obtained.
[0013]
Α _k (s _k ) and β _k (s _k ) are
[Expression 4]

[Equation 5]

Respectively can be obtained recursively.
[0014]
Using these, L (u _k ) is
[Formula 6]

Can be obtained.
[0015]
Here, Σ (s +) of the numerator indicates the sum when u _k = 1, and Σ (s−) of the denominator indicates the sum when u _k = −1.
Further, the external likelihood information L ^e (u _k ) sent to the next systematic convolutional code decoder is input to the received signal L _c y ³ _k weighted from the decoding result L (u _k ) and the decoder. It is obtained by subtracting the external likelihood information L ^e (u _k ).
[0016]
[Problems to be solved by the invention]
The error correction code obtains a correction capability by encoding an information sequence and adding redundant bits, so that a wide transmission band is required by adding redundant bits. In general, coding with high redundancy (low coding rate) causes a decrease in transmission speed, but a wideband signal is originally used in combination with code division multiple access (CDMA) by spread spectrum. Therefore, it was not necessary to consider much about the band spread.
[0017]
However, as the number of codes that can be accommodated in a specific area increases with the progress of spatial signal processing, such as error correction technology, interference cancellation technology, or adaptive array antenna, Combinations are limited. That is, in order to store a certain number of users or more, there is a situation in which a combination having a certain degree of cross-correlation must be used.
[0018]
The present invention has been proposed in view of the above-described circumstances, and uses a method of multiplexing and transmitting / receiving using a plurality of codes having a cross-correlation for a specific user, and using a cross-correlation value between these codes. It is an object of the present invention to provide a transmitter, a receiver, and a digital communication system using parallel concatenated convolutional coding that can reduce deterioration due to interference by performing decoding processing.
[0019]
[Means for Solving the Problems]
In order to achieve the above-described object, a receiver using parallel concatenated convolutional coding according to the present invention is a receiver used in a digital communication system that communicates by performing parallel concatenated convolutional coding of an information sequence,
Despreading means for performing despreading using a plurality of different spreading codes, crosscorrelation value generating means for generating crosscorrelation values between spreading codes, and despreading performed before spreading on the transmitter side An interleaver for performing an interleave process corresponding to the interleaver, and a turbo decoding means for performing a turbo decoding process by inputting a reception signal subjected to the despreading and the interleaving process,
The turbo decoding means is characterized in that decoding processing is performed using the cross-correlation between spreading codes obtained by the cross-correlation value generating means .
[0022]
Here, the cross-correlation value generating means obtains a cross-correlation value between the spreading codes by calculation, or stores a cross-correlation value calculated in advance.
[0023]
A digital communication system using parallel concatenated convolutional coding according to the present invention is obtained by the cross-correlation value generating means in the turbo decoding means at the time of reception in the receiver using the transmitter and the receiver. The decoding process is performed using the cross-correlation between the spread codes.
[0025]
The receiver using parallel concatenated convolutional coding according to the present invention is a receiver used in a digital communication system in which an information sequence is communicated by parallel concatenated convolutional coding, and is despread using a plurality of different spreading codes. Based on the cross-correlation value between the spreading code and the despreading means for performing the despreading means, the means for separating the output of the despreading means corresponding to the parity part into the first parity part and the second parity part Interfering processing corresponding to the deinterleaver performed before spreading on the transmitter side, and the interference amount generating means for generating the amount of interference between the unit and the first parity unit, and the organization unit and the second parity unit And an interleaver for performing de-spreading and interleaving processing, and a turbo decoding means for performing turbo decoding processing by inputting a received signal subjected to despreading and interleaving processing,
The turbo decoding means is characterized by performing decoding processing using the interference amount obtained by the interference amount generating means.
[0026]
Here, the interference amount generation means obtains the interference amount between the spread codes by calculation, or stores the interference amount calculated in advance.
[0027]
Also, the digital communication system using parallel concatenated convolutional coding according to the present invention uses the transmitter and the receiver, and the receiver uses the first complex spreading means and the second complex spreading unit upon reception. The signals spread-modulated by the spreading means are combined, and the receiver performs a decoding process using the interference amount obtained by the interference amount generating means in the turbo decoding means at the time of transmission. Is.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a transmitter, a receiver, and a digital communication system using parallel concatenated convolutional coding according to the present invention will be described based on specific embodiments shown in the drawings.
[0029]
<Example 1: Transmitter>
FIG. 1 is a block diagram showing a specific example 1 of a transmitter using parallel concatenated convolutional coding according to the present invention.
As shown in FIG. 1, the transmitter according to the first embodiment includes a turbo encoder 4, a deinterleaver 5, multipliers 6, 7, 8, 10, an adder 9, and a radio carrier generator 11. Become. The turbo encoder 4 includes recursive systematic convolutional encoders (RSCC) 1 and 2 and an interleaver 3. In the recursive systematic convolutional encoder, 1 is a first encoder and 2 is a second encoder.
[0030]
The transmitter 1 includes, in the output of the turbo encoder 4, the same part as the input data, the organization part (i), the parity (p) added by the first encoder 1 (RSCC1), the first parity part, The parity (p) added by the second encoder 2 (RSCC2) is set as the second parity part. Note that a sequence obtained by interleaving an information bit sequence of input data by the interleaver 3 is input to the second encoder 2 (RSCC2).
[0031]
The second parity part encoded by the second encoder 2 is rearranged by passing through the deinterleaver 5. The system unit, the first parity unit, and the second parity unit are spread and modulated in multipliers 6, 7, and 8 by taking products of PN 1, PN 2, and PN 3 that are different from each other. Further, the multiplier 10 multiplies the carrier wave from the wireless carrier wave generator 11, performs carrier modulation, and transmits the result.
[0032]
The turbo encoder 4 may have the same configuration as the conventional turbo encoder (see FIG. 5) described above. Furthermore, it is also possible to include a third parity part and a fourth parity part.
[0033]
<Example 1: Receiver>
FIG. 2 is a block diagram showing a specific example 1 of a receiver using parallel concatenated convolutional coding according to the present invention.
The receiver according to the first embodiment includes multipliers 12, 15, 16 and 17, a radio carrier generator 13, a low-pass filter (LPF) 14, despreading units 18, 19 and 20, and a turbo decoder 30. Become. The turbo decoder 30 includes a first decoder 31, a second decoder 32, an interleaver 33, and deinterleavers 34, 35, and 36.
[0034]
In the receiver 12, the multiplier 12 multiplies the received signal by the carrier wave from the radio carrier generator 13 to perform carrier modulation, passes the low-pass filter (LPF) 14, and then despreads 118, 119, Despread at 120. The despreading units 118, 119, and 120 are multiplied by the respective spreading codes PN1, PN2, and PN3 in the multipliers 15, 16, and 17 and passed through the integral dump filters 18, 19, and 20, respectively. Spread. At this time, an interference component remains in each signal depending on the magnitude of the cross-correlation.
[0035]
The despread signals are sent to the turbo decoder 30 as a system unit, a first parity unit, and a second parity unit, respectively. Here, the transmission sequence of the organization part is X (n), the transmission sequence of the first parity part is Y ₁ (n), the transmission sequence of the second parity part is Y ₂ (n), and each of PN1, PN2, and PN3 If the received symbols despread at r ₁ (n), r ₂ (n), and r ₃ (n) are
r ₁ (n) = X (n) + θ _1,2 (n) Y ₁ (n) + θ _1,3 (n) Y ₂ (n) + N ₁ (n)
r ₂ (n) = θ _1,2 (n) X (n) + Y ₁ (n) + N ₂ (n)
r ₃ (n) = θ _1,3 (n) X (n) + Y ₂ (n) + N ₃ (n)
It becomes.
[0036]
Here, N ₁ , N ₂ , and N ₃ are noise components. Here, PN2 and PN3 are sequences orthogonal to each other. For example, when PN1 is an m sequence generated by a generator polynomial 1 + x ⁷ + x ¹⁸ and PN2 is an m sequence generated by a generator polynomial 1 + x ⁵ + x ⁷ + x ¹⁰ + x ¹⁸ , PN3 is 1, -1, .. PN2 and PN3 are orthogonal to each other in the power of 2 power.
[0037]
Here, in order to obtain the transition probability using the cross-correlation values θ _1,2 (n) and θ _1,3 (n), the above equation (3) is modified as the following equation.
[Expression 7]

However, as θ, θ _1,2 is used when RSCC1 is decoded, and θ _1,3 is used when RSCC2 is decoded.
[0038]
r ₁ (n) is input to the first decoder 31 and the deinterleaver 35, r ₂ (n) is input to the first decoder 31, and r ₃ (n) is input to the deinterleaver 36. The The first decoder 31 outputs external likelihood information Le ⁽¹⁾ (u) and a decoding result L (u) from r ₁ (n), r ₂ (n) and the cross-correlation value. Here, the cross-correlation value is obtained by taking a cross-correlation between r ₁ (n) and r ₂ (n), and is obtained by a cross-correlation generating unit (not shown).
The external likelihood information Le ⁽¹⁾ (u) is rearranged by the interleave 33 and input to the second decoder. The second decoder 32 uses the external likelihood information Le ⁽¹⁾ (u) and the deinterleaved r ₁ (n) and r ₃ (n) to generate the external likelihood information Le ⁽²⁾ (u) and The decoding result L (u) is output.
The external likelihood information Le ⁽²⁾ (u) is deinterleaved by the deinterleaver 34 and input to the first decoder 31 again.
[0039]
Here, the spreading codes are generated so that PN2 and PN3 are orthogonal to each other. However, this is not necessarily a necessary condition, and PN1, PN2, and PN3 may be different random sequences.
[0040]
<Example 2: Transmitter>
As a method of making the first parity part and the second parity part orthogonal, a method of assigning them to the I layer and the Q layer by using QPSK modulation may be employed. At this time, the systematic unit performs QPSK spread modulation using a complex expansion sequence using information modulation as BPSK, so that the first parity part and the second parity part are evenly correlated.
[0041]
FIG. 3 is a block diagram showing a specific example 2 of a transmitter using parallel concatenated convolutional coding according to the present invention. In addition, the same code | symbol is attached | subjected and demonstrated to the part which has the same function as the transmitter of Example 1 mentioned above.
As shown in FIG. 3, the transmitter of Embodiment 2 turbo-codes a data sequence and separates it into a systematic part, a first parity part, and a second parity part, and passes the second parity part through the deinterleaver 5. Up to this point, the process is the same as in the first embodiment.
[0042]
After that, the systematic part is spread and modulated by the complex modulation code PN1 in the spread modulation part 6, and the first parity part and the second parity part are mapped on the complex plane in the mapping part 21, and then in the multiplier 7. The signal is spread-modulated with a different complex spread code PN2, and is combined with the tissue portion signal in the adder 9. Further, the multiplier 10 multiplies the radio carrier from the radio carrier generator 11 to perform carrier modulation, and transmits the result.
[0043]
<Example 2: Receiver>
FIG. 4 is a block diagram showing a specific example 1 of a receiver using parallel concatenated convolutional coding according to the present invention. In addition, the same code | symbol is attached | subjected and demonstrated to the part which has the function similar to the receiver of Example 1 mentioned above.
[0044]
In the receiver of the second embodiment, as shown in FIG. 4, the multiplier 12 multiplies the received signal by the carrier from the radio carrier generator 13 to modulate the carrier, and passes the low-pass filter (LPF) 14. After that, despreading is performed by the despreading units 122 and 123. The despreading units 122 and 123 multiply the respective spreading codes PN1 and PN2 in the multipliers 15 and 16, and despread them by passing through the integral dump filters 22 and 23. In this way, the received signal is despread using each spreading code.
[0045]
At this time, the components of the tissue part are despread by PN1 and then extracted as a real part by the real number extraction unit 24. The first parity part is extracted as a real part by the real number extraction part 25 from the signal despread by PN2. The second parity part is extracted from the imaginary number extraction unit 26 as an imaginary part component from the signal despread by PN2.
Here, assuming perfect synchronous detection, no interference component occurs between the first parity part and the second parity part. However, interference occurs between the tissue sections corresponding to the cross-correlation between PN1 and PN2.
[0046]
When PN1 (u + jv) is assumed and PN2 is represented by (r + js), the cross-correlation between sequences is obtained as follows. However, j is an imaginary unit (j = (− 1) ^1/2 ).
(U + jv) (r−js) = (ur + vs) + j (vr−us)
That is, the amount of interference between the tissue unit and the first parity unit is ur + vs, and the amount of interference between the tissue unit and the second parity unit is vr-us. This value is obtained by an interference amount generation unit (not shown).
[0047]
The despread and separated components are sent to the turbo decoder 30. The interference amount obtained here is used for decoding instead of the cross-correlation value in the first embodiment.
Here, PN2 is a complex value series, but it may be a real value series.
[0048]
<Error characteristics>
Next, error rate characteristics in the configuration of the first embodiment described above will be described.
The turbo code has a constraint length of “3” and a block length of “1000”. The number of iterative decoding is “6”. As the modulation method, both information modulation and spread modulation are performed by BPSK, and an M sequence having a cycle of “4” and a spread code of “2 ¹⁸ −1” is truncated by the frame length. The communication path is static.
[0049]
FIG. 5 shows a comparison result of error rate characteristics.
The comparison target is a characteristic when a conventional MAP decoding process is performed on a signal that has been encoded and code-multiplexed in the same manner as in the first embodiment.
As can be seen from FIG. 5, according to the configuration of the first embodiment, the error rate characteristics are further improved as compared with the conventional decoding method.
[0050]
【The invention's effect】
Since the transmitter, receiver, and digital communication system using parallel concatenated convolutional coding according to the present invention have the above-described configuration, the following effects can be achieved. That is, in the transmitter using parallel concatenated convolutional coding according to the present invention, the first convolutional coder and the second convolutional coder that perform the convolutional coding processing are rearranged based on a predetermined rule. After the information signal is turbo-encoded by a concatenated convolutional encoder connected in parallel via an interleaver that performs the operation, the second parity sequence obtained from the second convolutional encoder is inversely operated by the interleaver. Reordering is performed using an inverse interleaver, and a systematic part that is the same series as the input data, a first parity part output from the first convolutional encoder, and a second parity part output from the inverse interleaver Each of the first spreading code sequence, the second spreading code sequence, and the third spreading code sequence is spread, multiplexed and transmitted.
[0051]
Further, in the receiver using the parallel concatenated convolutional coding of the present invention, the magnitude of the interference component is larger than the received symbol including the interference component obtained by performing the despreading process using each spreading code sequence. The decoding process is repeatedly performed using the cross-correlation value between the spreading codes representing the length.
[0052]
The digital communication system using parallel concatenated convolutional coding according to the present invention performs digital communication using the transmitter and the receiver.
Therefore, the occurrence of errors due to interference can be reduced, and high-speed and high-quality communication can be performed.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a specific example 1 of a transmitter using parallel concatenated convolutional coding according to the present invention.
FIG. 2 is a block diagram showing a specific example 1 of a receiver using parallel concatenated convolutional coding according to the present invention;
FIG. 3 is a block diagram showing a specific example 2 of a transmitter using parallel concatenated convolutional coding according to the present invention.
FIG. 4 is a block diagram showing a specific example 2 of a receiver using parallel concatenated convolutional coding according to the present invention.
FIG. 5 is an explanatory diagram of error rate characteristics when the transmitter and receiver according to the first embodiment are used.
FIG. 6 is a block diagram of a conventional turbo code encoder.
FIG. 7 is a block diagram of a conventional turbo code decoder.
[Explanation of symbols]
1 First encoder (RSCC1)
2 Second encoder (RSCC2)
3 Interleaver 4 Turbo encoder 5 Inverse interleaver 6, 7, 8 Spreading modulation unit 9 Combining unit 10, 12 Carrier modulation unit 11, 13 Radio carrier generator 14 Low pass filter (LPF)
15, 16, 17 Despreading unit 18, 20 Turbo decoder 19 Mapping unit 51 First encoder (RSCC1)
52 Second encoder (RSCC2)
53 Interleaver 61 First encoder (RSCC1)
62 Second encoder (RSCC2)
63 Interleaver 64 Reverse interleaver

Claims (4)

A receiver used in a digital communication system for communicating information sequences by parallel concatenated convolutional coding,
  Despreading means for performing despreading using a plurality of different spreading codes, cross-correlation value generating means for generating cross-correlation values between the spreading codes,
  An interleaver for performing interleave processing corresponding to the deinterleaver performed before spreading on the transmitter side;
  Turbo decoding means for inputting a received signal subjected to despreading and interleaving processing and performing turbo decoding processing,
  A receiver using parallel concatenated convolutional coding in which the turbo decoding means performs a decoding process using the cross-correlation between spreading codes obtained by the cross-correlation value generating means. A digital communication system for communicating with a parallel concatenated convolutional coding the information sequence,
Multi-code transmission is performed using different spreading code sequences for each of the parity part generated by the system part of the code series and at least two or more encoders including the first and second encoders. Transmitters each provided with a deinterleaver for performing reordering to have a temporal relationship with the sequence of the organization unit for parity units other than the parity unit generated by one encoder;
Despreading means for performing despreading using a plurality of different spreading codes, crosscorrelation value generating means for generating crosscorrelation values between spreading codes, and despreading performed before spreading on the transmitter side Using an interleaver for performing an interleave process corresponding to the interleaver, and a receiver including turbo decoding means for performing a turbo decoding process by inputting a reception signal subjected to despreading and interleaving process, In the transmitter, after performing the rearrangement by the deinterleaver at the time of transmission, spread spectrum modulation is performed,
In the receiver, at the time of reception, the turbo decoding means uses parallel concatenated convolutional coding, in which decoding processing is performed using the cross-correlation between spreading codes obtained by the cross-correlation value generating means. Digital communication system. A receiver used in a digital communication system for communicating information sequences by parallel concatenated convolutional coding,
  Despreading means for performing despreading using a plurality of different spreading codes, and means for separating the output of the despreading means corresponding to the parity part into a first parity part and a second parity part,
  An interference amount generating means for generating an interference amount between the systematic part and the first parity part, and the systematic part and the second parity part, based on a cross-correlation value between spreading codes;
  An interleaver for performing interleave processing corresponding to the deinterleaver performed before spreading on the transmitter side;
  Turbo decoding means for inputting a received signal subjected to despreading and interleaving processing and performing turbo decoding processing,
  A receiver using parallel concatenated convolutional coding, wherein the turbo decoding means performs a decoding process using the interference amount obtained by the interference amount generating means. A digital communication system for communicating information series by performing parallel concatenated convolutional coding,
  The code sequence is separated into a systematic part, a first parity part generated by a first encoder, and a second parity part generated by a second encoder, and systematized with respect to the second parity part With a deinterleaver for performing rearrangement that has a temporal relationship with the series of parts,
  Parity for mapping on the QPSK signal point using one bit each in the second parity part after the rearrangement by the deinterleaver and the first parity part not in the rearrangement Partial mapping means, first complex spreading means for spreading a code sequence of the tissue part as a signal on a BSPK signal point on a complex plane using a complex-valued spreading code, and mapping on the QPSK signal point Organized parity part signal A transmitter comprising a second complex spreading means 2 for spreading on a complex plane using a complex-valued or real-valued spreading code different from the part sequence;
  Despreading means for performing despreading using a plurality of different spreading codes, means for separating the output of the despreading means corresponding to the parity part into a first parity part and a second parity part, and spreading Based on cross-correlation values between codes, an interference amount generating means for generating an interference amount between the system unit and the first parity unit, and between the system unit and the second parity unit, and before spreading on the transmitter side A receiver including an interleaver for performing interleave processing corresponding to the deinterleaver performed in step 1 and turbo decoding means for inputting a reception signal subjected to despreading and interleaving processing and performing turbo decoding processing; Using,
  In the receiver, upon reception, the signals modulated by the first complex spreading means and the second complex spreading means are combined,
  The digital communication system using parallel concatenated convolutional coding, wherein in the receiver, the turbo decoding unit performs decoding processing using the interference amount obtained by the interference amount generation unit at the time of transmission.

Images