JPWO2003036895A1 - COMMUNICATION SEMICONDUCTOR INTEGRATED CIRCUIT, MODEM / DEMODULATION DEVICE, AND COMMUNICATION STATE DIAGNOSIS METHOD - Google Patents

Abstract

A diagnostic program that can be executed by the CPU (340) in the modem LSI is loaded into the memory (370) in the modem device, and the diagnostic program is executed by the on-chip CPU to be provided in the modem LSI. The values of the registers (REG1 to REG4) that hold the information relating to the communication state are output to the external diagnostic computer (150, 200, 400).

Description

デローテータ３３２５の出力は、各種の信号処理によって再生された信号であって、スライサ３３２６ａ，３３２６ｂに入力される信号ｄｅｒｏｔ−ｉ／ｑはそれぞれ例えば８ビットの信号で表現されている。変調方式が１６ＱＡＭの場合、Ｉ／Ｑはそれぞれ２ビットの情報を持っているので、８ビット２５６階調の信号から等分に２ビット４階調の信号を取り出すのが、スライサ３３２６ａ，３３２６ｂの動作である。
復調回路は、復調結果とその期待値とを比較することによって誤差を計算し、誤差成分をフィードバックすることによって最適な復調条件に収束させるように構成されており、この実施例の復調回路ではスライサの出力を期待値として用いている。上記説明では、理解を容易にするため、変調方式が１６ＱＡＭの場合を取り上げたが、ケーブルモデムからヘッドエンドへの下り回線では６４ＱＡＭまたは２５６ＱＡＭを用いることが規定されている。この場合、ＩとＱそれぞれは、６４ＱＡＭで３ビット、２５６ＱＡＭで４ビットにスライスされる。
デマッパ３３３は、スライサ３３２６ａ，３３２６ｂによって弁別されたＩ／Ｑシンボルを誤り訂正の単位である８ビットもしくは７ビットに変換する機能を有する。そして、変換されたビット列に対して、誤り訂正回路３３４が予め定められた誤り訂正処理を施して出力する。
上記デローテータ３３２５の出力信号ｄｅｒｏｔ−ｉ／ｑはそれぞれ８ビットの信号で、Ｉ／Ｑの２次元の平面上にプロットすると、図７のようになる。これがコンステレーションと呼ばれるものである。図７は、理解を容易にするため、変調方式が１６ＱＡＭの場合を示したものである。なお、Ｉ／Ｑに何ビットを与えるかは設計思想に依存するもので、ここで８ビットとしたのは一例にすぎない。
伝送線路（ケーブル）上で雑音の影響を受けなければ、受信信号は図７に●印で示されている点に集まり、デマッパ３３３によって括弧内に示されているような４ビットの２進信号に変換される。ただし、実際には、伝送線路上で雑音の影響を受けて、図７に×印で示されているように理想点（●）から離れた信号となって現われる。
図８の（ａ）には理想的な場合のコンステレーションが、また（ｂ）には白色雑音を受けた場合、（ｃ）には位相誤差を含む場合のコンステレーションがそれぞれ示されている。受信信号は白色雑音を受けた場合、雑音が小さい理想点に近いものほど出現確率が高く振幅の小さいものを中心に正規分布するので、各受信信号は理想点を中心としその回りに確率的に分布し、コンステレーションは図８（ｂ）のように濃淡の模様となって現われる。一方、位相誤差を含む場合は、図８（ｃ）のように、全体が回転したようなコンステレーションとなる。
位相誤差θと雑音電力ＮＰは、図９に示すようにコンステレーションから計算で求めることができる。すなわち、位相誤差θはＩ／Ｑの０点と理想点（●印スライス結果）とを結ぶ線と、０点と着目点（×印信号処理結果）とを結ぶ線のなす角として、また雑音電力ＮＰは理想点から各点までの距離ｅの２乗の和（Σｅ^２）として求められる。本実施例では、復調回路内でこの計算を行ない、結果はそれぞれ対応するレジスタに格納される。位相誤差の計算は、デローテータ３３２５における回転補正角として、また雑音電力は適応等化フィルタ３３２４における係数を求めるためにもともと算出しているもので、本発明の通信障害診断のためにわざわざ追加する必要のある機能ではない。
次に、本発明の実施例における診断用コンピュータ４００を利用した通信障害診断の手順の一例を、図１１のフローチャートと図１２のデータフローチャートを用いて説明する。
エンドユーザにおいて障害が発生すると、連絡を受けたケーブル運営会社はエンジニアを派遣する。派遣されたエンジニアは、エミュレータ４１０などを使用して、診断用コンピュータ４００を障害が発生しているケーブルモデム装置３３０に接続する（ステップＳ１）。そして、診断用コンピュータ４００からケーブルモデム装置１３０内のモデム用ＬＳＩ３３０のプロセッサ３４０へ診断用プログラムを転送する（ステップＳ２）。すると、プロセッサ３４０は受信した診断用プログラムをＣＰＵに接続されているメモリ３７０内に格納する。このときのデータの流れが図１２に符号▲１▼で示されている。
その後、エンジニアがキーボードなどの入力装置４０１から起動指令や観測条件を入力すると、診断用コンピュータ４００がプロセッサ３４０へ診断用プログラムを起動させるコマンドを与える（図１１のステップＳ３、図１２の符合▲２▼）。すると、プロセッサ３４０は診断用プログラムを実行し、観測対象の設定や、観測周期（タイマ割込み周期）等の観測条件の設定、チップ内部の回路からの割込み条件の設定、レジスタの読出し情報を処理して出力する前処理内容の設定などの診断処理の具体的な内容を設定する（図１１のステップＳ４、図１２の符合▲３▼等）。割込み条件としては、例えばレジスタＲＥＧ１の訂正フラグが立ったら割込みを発生する等がある。
続いて、エンジニアがコンソール４０１から計測開始指令を入力すると、診断用コンピュータ４００がプロセッサ３４０へ計測開始コマンドを与える（図１１のステップＳ５、図１２の符合▲２▼）。すると、プロセッサ３４０はステップＳ４で設定された内容に従って、各レジスタの状態を順次読み出してメモリ３７０に格納する（図１１のステップＳ５、図１２の符合▲４▼）。それから、前処理が設定されている場合は、設定された前処理を行なって観測データをモニタの画面上に表示させたときに通信状態を判断し易い形態等に処理する（ステップＳ６）。
次に、プロセッサ３４０は、ステップＳ５で取得した観測データおよびステップＳ６で前処理したデータを診断用コンピュータ４００へ転送する（図１１のステップＳ７、図１２の符合▲５▼）。この観測データの転送は、診断用コンピュータ４００からの転送要求コマンドに行なっても良いし、一連の観測データが得られたときにプロセッサ３４０が転送を実行するようにしても良い。観測データが転送されると、診断用コンピュータ４００は、受信した観測データに対して例えば許容範囲を超えているデータは赤色で強調するなどの後処理を行なう（ステップＳ８）。それから、診断用コンピュータ４００は、後処理データに基づいて観測結果をモニタ４０２の画面上に表示させる（図１１のステップＳ９、図１２の符合▲６▼）。
その後、エンジニアがこの観測結果の表示を見て通信状態を診断してモデム装置の適切な調整を行なう（ステップＳ１０）。そして、再び、ステップＳ４へ戻って観測を行ない、所望の通信状態が得られるまで、上記処理を繰り返す。調整の結果、良好な通信状態を実現できたなら、エンジニアはキーボートから診断終了指令を入力すると、診断用コンピュータ４００は、モデム用ＬＳＩ内のプロセッサ３４０に対して診断プログラムの停止コマンドを与える（図１１のステップＳ１１、図１２の符合▲２▼）。
さらに、診断用コンピュータ４００は、プロセッサ３４０に対して診断プログラムの削除コマンドを送る（図１１のステップＳ１２、図１２の符合▲１▼）。すると、プロセッサ３４０はメモリ３７０から診断プログラムを削除して診断処理を終了する。なお、診断用コンピュータ４００からプロセッサ３４０に与える診断プログラムの削除コマンドは、停止コマンドに続けて自動的に与えても良いが、エンジニアがキーボードから削除指令を入力したことを条件とすることができる。診断プログラム削除後にエンジニアは、診断診断用コンピュータ４００をモデム装置１３０から取り外して作業を終了する。
診断プログラムは、上記のように削除しても良いが、そのままエンドユーザのモデム装置内に残すようにしてもよい。削除することによりプロセッサ３４０が利用可能なメモリ領域を減るのを回避でき、また削除せずに残しておくことにより、次回の通信障害発生時における診断を容易にすることができる。例えば、ユーザー自身が自己のネットワーク端末により通信状態を診断して、適切な処置を施すことができる可能性がある。
図１３は、診断用コンピュータ４００の制御手順を示すフローチャートである。
診断用コンピュータ４００は、キーボードなどから診断プログラムの転送指令が入力されると、モデム用ＬＳＩ３３０内のプロセッサ３４０へ診断プログラムを転送する（ステップＳ１１）。その後、キーボードからコマンド入力があると、そのコマンドを解析して終了コマンドであれば、プロセッサ３４０へ診断プログラムの停止を、また削除コマンドのときはプログラムの削除を指令する（ステップＳ１４→Ｓ２０）。終了コマンドと削除コマンド以外のコマンドのときは、対応する制御コマンドをプロセッサ３４０へ送って、観測制御を行なう（ステップＳ１５）。
ここで、観測制御には、診断プログラムの起動や停止、観測対象の設定、観測周期の設定、前処理内容の設定などが含まれる。また、ステップＳ１５で観測内容の変更を行なったときは、その変更に応じて診断用コンピュータ側の結果表示の制御内容を変更する（ステップＳ１６）。さらに、診断用コンピュータ４００は、コマンドの有無に関わらずモデム用ＬＳＩ１３０に観測結果の転送要求を行なう（ステップＳ１７）。そして、転送されてきた観測データに対して後処理を施し、モニタ４０２の画面上に、例えば図１８に示すように、グラフ等の形式で観測結果を表示させる（ステップＳ１８，Ｓ１９）。
図１８において、符号Ａは受信信号に高速フーリエ変換を施した結果を横軸に周波数をとって示したもの、符号Ｂはシンボルの頻度分布を解析することで得られたコンステレーション、符号Ｃはビット誤り率の時間的変化を示したもの、符号Ｄは位相誤差の時間的変化を示したもの、符号Ｅは雑音電力の時間的変化を示したものである。この実施例では、上記受信信号に対する高速フーリエ変換は、モデム用ＬＳＩ１３０のプロセッサ３４０によって前処理として行なわれる。
なお、ステップＳ１７において行なわれる診断用コンピュータ４００からモデム用ＬＳＩ１３０に対する転送要求は、タイマ割込みで定期的に行なうようにすることができる。また、診断用コンピュータ４００からの転送要求でなく、モデム用ＬＳＩ１３０側からの一方的なデータ転送の場合には、受信割込み等によってデータを受け取るようにすることができる。
図１４〜図１７には、プロセッサ３４０によって実行される診断プログラムの制御手順を示すフローチャートが示されている。
このうち図１４はコマンド処理ルーチンである。診断用コンピュータ４００からコマンド入力があると、プロセッサ３４０はそのコマンドを解析する（ステップＳ２１）。解析の結果、入力コマンドがプログラム削除コマンドであったときは、メモリ３７０内に格納されている診断プログラムを削除する。また、入力コマンドが観測対象設定コマンドであったときは、観測対象項目に応じてソフトウェア上に設けられているフラグを設定することにより、測定指示を与える（ステップＳ２３）。
入力コマンドが観測周期設定コマンドであったときは、タイマ３４５にタイマ割込みの周期を設定する（ステップＳ２４）。入力コマンドが前処理内容設定コマンドであったときは、実行すべき前処理の内容を設定する（ステップＳ２５）。さらに、入力コマンドがデータ転送要求コマンドであったときは、観測の結果、メモリ３７０にストックされた測定結果データを読み出して診断用コンピュータ４００へ出力する（ステップＳ２６）。
図１５は、タイマ３４５から出力される割込み信号により実行される例えば１秒のような長い周期の割込み処理ルーチンである。
このルーチンでは、図１４のステップＳ２３で設定されたフラグをチェックして観測対象項目を把握し、フラグが立っている観測対象項目のレジスタの値を順次読み出してメモリ３７０に格納する。ただし、このうち適応等化フィルタの係数のように変化しないものについては、コマンドが１回発行されたときに１回だけ係数レジスタＲＥＧ３ａ，ＲＥＧ３ｂからの読出しを実行して転送し、フラグを解除するようにしている（ステップＳ３１〜Ｓ３４）。適応等化フィルタの係数は、引き込み動作中は変化するが、一旦収束すると伝送路の状態が変化しない限り一定であるためである。
一方、適応等化フィルタの係数以外の例えば位相誤差、周波数オフセット、雑音電力、ビット誤り率、シンボル誤り率、訂正フラグ等は、１秒ごとにレジスタＲＥＧ１，ＲＥＧ４から読み出してメモリ３７０に格納するようにしている（ステップＳ３５〜Ｓ４６）。このように、まとめてレジスタの内容を読み出してメモリに格納することにより、取得したデータを処理する場合にもデータの取得の同時性を高めることができる。
つまり、あるデータを取得したらそれに対する処理を行なってメモリに記憶してから次のデータの取得、処理に移るよりも、上記のようにまとめて一旦レジスタの内容をメモリに格納した後でデータの処理を行なう方が、取得したデータの同時性が高くデータ相互間の関係を調べるときの精度が高くなる。
なお、上記レジスタからのデータの読出し周期は１秒に限定されるものでなく、自由に設定することができる。周期が短いほど単発的な伝送路の障害を捕らえることができる可能性が高くなる一方、プロセッサ３４０の処理能力やメモリ３７０の記憶容量に対する負荷が大きくなり、通常処理が遅くなるので、プロセッサ３４０等の性能等に応じて周期を決定してやれば良い。
図１５のルーチンに従うと、プロセッサ３４０によるレジスタＲＥＧ１，ＲＥＧ４からの位相誤差、周波数オフセット、雑音電力、ビット誤り率、シンボル誤り率、訂正フラグ等の読出しは、順次行なわれることになるが、これらのレジスタの値は急激に変化するものでないので、１秒周期で読み出すことは実質的に同時観測しているとみなすことができる。
また、上記プロセッサ３４０は、同種のＬＳＩにおける平均的な性能のものであれば、モデム用ＬＳＩ本来の機能を実現する通常プログラムの実行と並行して上記診断プログラムを実行してもそれほど負担にならずプロセッサ３４０の処理能力を圧迫することにならない。また、通常プログラムと並行して診断プログラムを実行して、ＪＴＡＧなどのシリアルインタフェースを通して測定結果を出力しても、プロセッサ３４０の実動作に深刻な影響を与えることはない。
図１６および図１７は、Ｉ／Ｑシンボルを観測して前処理を施してコンステレーションを示すデータを生成するルーチンと、それを出力するルーチンである。このうち図１６のルーチンは例えば１ｍ秒のような短い周期で実行されるタイマ割込み処理とされ、図１７のルーチンは例えば１秒のような長い周期で実行されるタイマ割込み処理とされている。
図１６に示されている割込み処理が開始されると、先ずＩ／Ｑシンボルの観測対象フラグが立っているか判定し、立っているときはレジスタＲＥＧ２ａ，ＲＥＧ２ｂからシンボルを読み出す（ステップＳ５２）。そして、後述のようなシンボルの頻度分布解析をしてその解析結果をメモリ３７０へ格納する（ステップＳ５３，Ｓ５４）。
このような処理を１ｍ秒ごとに行ない、メモリに格納された頻度分布データを図１７の割込み処理で１秒ごとに診断用コンピュータ４００へ転送してから、メモリ上の累積頻度をクリアするようにしている（ステップＳ６１，Ｓ６２）。
これによって、コンステレーションの変化を診断用コンピュータのモニタ４０２によりリアルタイムで観測することができる。また、レジスタＲＥＧ２ａ，ＲＥＧ２ｂから読み出されたＩ／Ｑシンボルデータをそのままメモリ３７０に記憶するのではなく、頻度分布に置き換えてから記憶するようにしているため、メモリ３７０の使用量は非常に少なくて済むとともに、データを診断用コンピュータ４００へ転送する場合にもデータ量が少ないので転送速度を下げることができる。
上記ステップＳ５２におけるシンボルの頻度分布の解析は、プロセッサ３４０が、Ｉ／Ｑの２次元平面を例えば図１０に示すように各々８×８の領域（４×４の領域等でも可）に仮想的に分割して、レジスタＲＥＧ２ａ，ＲＥＧ２ｂから読み出したＩ／Ｑシンボルが分割されたいずれの領域に属するか判定し、各領域ごとに、出現したシンボル数を累積し、その累積値をメモリに記憶することで行なわれる。
そして、診断用コンピュータ４００が各領域の累積値に応じて濃淡のパターンを生成することで図１８に符号Ｂで示すようなコンステレーションがモニタ上に表示される。なお、図１０は、図９のＩ／Ｑ平面における破線で分割された中央の１枡部分を拡大して示したものであり、他の部分について同様に領域分割と、領域ごとのシンボル数を累積が行なわれる。
従来の障害診断方式は、観測したいレジスタの内容をチップ外部へ出力するためのピンをモデム用ＬＳＩに設けておいて、そのピンからのデータをモニタに表示させる方式であったため、１度に１種類のデータしか観測できなかったが、この実施例の障害診断装置によれば、図１８のように複数の種類のデータに関する観測結果が１つの画面上に表示されるため、例えばあるデータがある変化したときに他のデータはどのように変化しているかを知ることができる。その結果、総合的な判断が行ない易くなるという利点がある。
また、従来の障害診断方式では、観測したいデータを増やしたい場合にはピンを増加させなくてはならず、チップサイズが増大するという不具合もあったが、本発明方式ではＪＴＡＧポートを設けるだけで良い。しかも、このＪＴＡＧポートはモデム用ＬＳＩの通常プログラムをデバッグするためのエミュレータを接続するためにも利用できるので、極めて都合が良い。
次に、本発明の他の実施例を、図１９〜図２１を用いて説明する。
このうち、図１９の実施例は、モデム用ＬＳＩ３３０のチップ内に、ＡＤ変換回路３３１の出力やナイキストフィルタ３３２２の出力、適応等化フィルタ３３２４ａ，３３２４ｂの出力、デローテータ３３２５の出力、スライサ３３２６ａ，３３２６ｂの出力など、モデム用ＬＳＩ内の各部の受信データをサンプリングしてメモリ３６０に直接転送するＤＭＡコントローラ３８０を設けたものである。また、図１９には、受信データの処理中に所定の割込み条件が成立したときにプロセッサ３４０に対して割込みをかける割込み制御回路３９０の一例が示されている。
この実施例における割込み制御回路３９０は、ビット誤り率のしきい値を保持するレジスタ３９１やシンボル誤り率のしきい値を保持するレジスタ３９２、レジスタＲＥＧ１内のビット誤り率と上記誤り率のしきい値とを比較する比較回路３９３、レジスタＲＥＧ１内のシンボル誤り率と上記シンボル誤り率のしきい値とを比較する比較回路３９４、これらの比較回路３９３、３９４の出力とレジスタＲＥＧ１内のエラー訂正フラグなどに基づいて割込み要因を選択し割込み信号を生成したり割込みにマスクをかけたりする割込み要因選択／マスク回路３９５などから構成されている。
上記ＤＭＡコントローラ３８０を割込み制御回路３９０による割込みに連動して動作させることで少ないメモリを使用して効率の良い診断を行なうことができる。例えば、エラー訂正フラグが訂正不能を示したりビット誤り率が所定のしきい値を越えたら受信データのメモリへの転送を開始したり、逆にエラー訂正フラグが訂正不能を示したりビット誤り率が所定のしきい値を越えたら受信データのメモリへの転送を停止させたりすることで、ビット誤りが増加する現象が発生する直前の受信データや直後の受信データをメモリに格納することができる。
これにより、まれにしか発生しないような障害についてもその現象を捉えて診断を行なうことができるようになる。なお、ある現象が発生する直前の受信データをメモリに格納する場合、メモリの利用できる領域が少ないときは容量をオーバーした段階で順次古いデータの上に新しいデータを上書きして行き、常に最新のデータが残るようにすればよい。
図２０の実施例は、ケーブルモデム装置１３０に接続された診断用コンピュータ４００を、一般の電話回線などケーブルテレビのケーブル以外の他の伝送媒体を介して基地局のヘッドエンド１００と連繋されている制御用コンピュータ２００と接続して、ヘッドエンド１００と診断用コンピュータ４００とを同期、もしくは連動させて動作させるようにしたものである。この実施例に従うと、例えばヘッドエンドから基準となる信号や標準パターンなどを送信して、それをケーブルモデム装置が受信している状態を診断用コンピュータ４００で観測するようなことができる。
なお、図４の実施例と同様に、エンドユーザのケーブルモデム装置１３０に接続されているパソコンなどのローカルネットワーク端末１５０を診断用コンピュータとして使用し、このローカルネットワーク端末１５０の有している通信手段を利用して電話回線などを介して基地局の制御用コンピュータ２００と接続して診断を行なわせることも可能である。
この場合には、ケーブル運用会社のコンピュータ２００から電話回線等を介してエンドユーザのローカルネットワーク端末１５０に診断用プログラムを送信し、また観測データをローカルネットワーク端末１５０からケーブル運用会社のコンピュータ２００へ送信することでエンジニアを派遣することなく診断を行なうことができる。さらに、モデム内部を調整するデータもケーブル運用会社のコンピュータ２００からローカルネットワーク端末１５０へ送信してモデムＬＳＩ内部のプロセッサ３４０により内部制御状態を調整させるようにすることも可能である。
図２１の実施例は、ローカルネットワーク端末１５０を診断用コンピュータとするとともに、ケーブルモデム装置１３０のネットワークインタフェース３３６に備えている音声モデムやＵＳＢなどの通信機能を利用して、電話回線などを介して基地局の制御用コンピュータ２００と接続して診断を行なうにしたシステムの例である。なお、図２０において、２１０は電話回線の途中に設けられた中継装置である。
以上本発明者によってなされた発明を実施例に基づき具体的に説明したが、本発明は上記実施例に限定されるものではなく、その要旨を逸脱しない範囲で種々変更可能であることはいうまでもない。例えば、実施例においては、ケーブルテレビのケーブルが同軸ケーブルまたは光ファイバであるとして説明したが、ケーブル網が電話回線や電力線、無線通信網その他で構成されている場合にも適用することができる。本発明は双方向通信が可能な通信網におけるモデム用ＬＳＩに利用することができる。
また、実施例においては、診断用プログラムをエンドユーザに派遣されたエンジニアがモデム装置のメモリにロードさせる場合を説明したが、予めフラッシュＲＯＭ３６０などのメモリ内に格納しておくようにしてもよい。さらに、エンドユーザ側の診断用コンピュータもしくはネットワーク端末と基地局の制御用コンピュータとを接続するケーブル以外の通信媒体は電話回線に限定されずどのような通信手段であってもかまわない。
産業上の利用可能性
以上の説明では主として本発明者によってなされた発明をその背景となった利用分野であるケーブルテレビのケーブル網を利用してデータ通信を行なうケーブルモデム用ＬＳＩに適用した場合について説明したが、本発明はそれに限定されるものでなく、通信衛星を利用した衛星放送および衛星通信システムにおけるデータ通信用のモデムにも利用することができる。
【図面の簡単な説明】
図１は、本発明を適用して好適なケーブルテレビのシステム構成例を示すブロック図である。
図２は、ケーブルモデム装置の構成例を示すブロック図である。
図３は、図２のケーブルモデム装置を構成するモデム用ＬＳＩの復調回路の構成例を示すブロック図である。
図４は、本発明の他の実施例が適用されるモデム用ＬＳＩとこれを使用した障害診断システムを示すブロック図である。
図５は、１６ＱＡＭ変調の原理を示す説明図である。
図６は、１６ＱＡＭの復調の原理を示す説明図である。
図７は、１６ＱＡＭの復調データをＩ／Ｑ２次元平面上にプロットしたコンステレーションを示す説明図である。
図８は、理想的な受信データのコンステレーションと外乱を受けた受信データのコンステレーションを示す説明図である。
図９は、位相誤差と雑音電力の大きさを示す説明図である。
図１０は、Ｉ／Ｑシンボルの頻度分布を解析するための領域分割の仕方の一例を示す説明図である。
図１１は、本発明の実施例における診断用コンピュータを利用した通信障害診断の手順の一例を示すフローチャートである。
図１２は、図１１のフローチャートに従った通信障害診断におけるデータの流れを示すデータフローチャートである。
図１３は、本発明の通信障害診断方式における診断用コンピュータ側の制御手順の一例を示すフローチャートである。
図１４は、本発明の通信障害診断方式におけるモデム用ＬＳＩ側のコマンド処理手順の一例を示すフローチャートである。
図１５は、本発明の通信障害診断方式におけるモデム用ＬＳＩ側のタイマ割込み処理手順の一例を示すフローチャートである。
図１６は、本発明の通信障害診断方式におけるモデム用ＬＳＩ側の観測データの前処理手順の一例を示すフローチャートである。
図１７は、モデム用ＬＳＩで前処理された観測データの転送手順の一例を示すフローチャートである。
図１８は、本発明の通信障害診断方式における診断用コンピュータのモニタ上に表示される観測データの表示例を示す表示画面図である。
図１９は、ケーブルモデム装置を構成するモデム用ＬＳＩの復調回路の他の構成例を示すブロック図である。
図２０は、本発明の通信障害診断方式の他の実施例を示すブロック図である。
図２１は、本発明の通信障害診断方式のさらに他の実施例を示すブロック図である。Technical field
The present invention relates to a technique that is effective when applied to a diagnosis of a communication failure of a semiconductor integrated circuit and a communication modem (MODEM: modulation / demodulation device) using the semiconductor integrated circuit, and for example, data using a cable network laid for cable television. The present invention relates to a cable modem device that performs communication and a technology that is effective when used for an LSI for a cable modem that constitutes the modem device.
Background art
A cable modem is used for communication to provide network services using a cable network laid for cable television, and is a head end in a base station via a coaxial cable or optical fiber constituting the cable network. Is a device for connecting a local network terminal such as a personal computer (hereinafter abbreviated as a personal computer) of each home (end user) and modulating / demodulating transmission / reception signals.
According to the current cable television facility standards, the headend constantly monitors the cable network status, and parameters such as the frequency and strength of the transmitted / received signal and the data transfer rate for each end-user modem depend on the network status. It is determined to maintain communication quality by avoiding interference and collision between cable modems.
However, the diagnosis by monitoring the state of the cable is a so-called uplink communication state from each cable modem to the head end, and a failure that occurs in downlink communication from the head end to each home cable modem. It cannot be diagnosed. Therefore, the procedure for inquiring received power, bit error rate, etc. to individual cable modems using SNMP (Simple Network Management Protocol), which is a general network management protocol, is also defined in the cable television standard. .
However, in order for the diagnosis by this procedure to function effectively, it is a major premise that the head end and the cable modem of each household are connected so that communication by the protocol can be established. Therefore, in a state where communication using such a protocol cannot be established, it is not possible to diagnose a communication failure on the downlink according to the above standard. Further, the bit error rate of transmission data is exclusively used as an indicator of the communication state of the cable in the conventional fault diagnosis. However, the bit error rate is only a result of data transmission affected by the cable condition, and is not an index directly representing the cable condition.
For this reason, if there is a problem with the analog signal modulated on the modem side, a dedicated observation terminal is provided in advance in the modem LSI and an observation device is connected so that the signal can be observed directly. Thus, a method for diagnosing a communication failure has been taken. Such a communication failure diagnosis method has a problem that the size of the chip is increased or the failure of the chip is caused because a dedicated observation terminal must be provided in the modem LSI. Moreover, if there are a plurality of signals to be observed, a plurality of observation terminals must be provided.
In the conventional failure diagnosis method, a special engineer must be dispatched to a home with a communication failure. The dispatched engineer brings an observation device, removes the case of the modem device, and observes it at the LSI observation terminal. It is necessary to connect the equipment for observation. In addition, if there are a plurality of signals to be observed, it is necessary to reconnect each time in order to observe the signals at a plurality of observation terminals in order. For this reason, there is a problem that it takes much time and time to diagnose the reception state and eliminate the failure. Furthermore, when a failure occurs rarely, it is necessary to make an engineer wait until the failure occurs, and there is a problem that it takes a long time to obtain a diagnosis result.
An object of the present invention is to provide a communication semiconductor integrated circuit device (modem LSI) capable of acquiring information indicating a communication state without providing an observation terminal for observing a signal inside a chip.
Another object of the present invention is to provide a communication semiconductor integrated circuit device capable of acquiring information indicating a more detailed communication state and performing highly accurate communication failure diagnosis.
Another object of the present invention is to provide a semiconductor integrated circuit device for communication that can shorten the time required to issue a diagnosis result relating to a communication failure.
Another object of the present invention is to provide a communication semiconductor integrated circuit device that allows an engineer dispatched to an end user's home with a communication failure to easily connect the diagnosis device and obtain information necessary for communication failure diagnosis. There is to do.
Another object of the present invention is to provide a diagnostic technique capable of obtaining necessary information using a general-purpose personal computer or the like as a diagnostic apparatus without using a dedicated observation apparatus.
Still another object of the present invention is to obtain information indicating a communication state in a communication device (modem) of each end user without sending an engineer to the end user's home having a communication failure and to diagnose the communication failure. It is to provide a diagnostic technique that can be used.
The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.
Disclosure of the invention
The outline of a typical invention among the inventions disclosed in the present application will be briefly described as follows.
That is, a register for holding information reflecting the communication state is provided in the modem LSI, and the processor in the modem LSI chip is configured to be readable, and the diagnostic program is externally stored in the memory in the modem device. By loading and executing the diagnostic program by the on-chip processor, the value of a register that is provided inside the modem LSI and holds information reflecting the communication state is output from a predetermined port to the outside. It is what I did.
According to the above-described means, it is possible to obtain a modem LSI capable of acquiring information indicating a communication state without providing an observation terminal for observing a signal inside the chip. In addition, since there is no need to provide an observation terminal, a plurality of information can be read out from one port, so that a highly accurate communication failure diagnosis can be performed and dispatched to an end user's home with a communication failure. An engineer can easily perform a task of obtaining information necessary for communication failure diagnosis by connecting a diagnosis device, and can shorten the time until a diagnosis result relating to communication failure is output.
In addition, a local network terminal device connected to a local area network such as an Ethernet included in a modem LSI can be used as a diagnostic computer, thereby diagnosing a communication failure without using a dedicated observation device. Can be done.
In addition, by using another communication means represented by a telephone line, connecting a diagnostic computer or a local network terminal device to a control computer on the base station side, and transmitting data read from the LSI for the modem, It is possible to obtain information indicating a communication state in a device in each home without dispatching an engineer to an end user's home having a communication failure and to diagnose the communication failure.
BEST MODE FOR CARRYING OUT THE INVENTION
Preferred embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 shows an example of a system configuration of a cable television suitable for applying the present invention. In FIG. 1, reference numeral 100 denotes a head end as a television broadcast program distribution device provided in the base station, 110 denotes a coaxial cable connecting the head end 100 of the base station and a home (end user) having a cable TV contract, It is a cable like an optical fiber. The set box 120 and the cable modem device 130 are connected to the other end of the cable 110. A television receiver 140 is connected to the set box 120, and a network terminal device 150 such as a personal computer is connected to the cable modem device 130 via a local area network such as Ethernet.
A frequency band of 50 to 860 MHz is allocated to transmission of signals from the head end 100 to each end user (downstream: downstream), and it is specified to use either 64QAM (Quadrature Amplitude Modulation) or 256QAM modulation system. ing. Moreover, a frequency band of 5 to 42 MHz is allocated to transmission of signals from each end user to the head end 100 (upstream: upstream), and a modulation scheme of either QPSK (Quadrature Phase Shift Keying) or 16QAM is used. Is stipulated. QPSK is a modulation system equivalent to 4QAM.
FIG. 2 shows a configuration example of the cable modem device 130 in FIG. The cable modem device 130 amplifies a tuner circuit 310 that extracts a signal of a designated channel (frequency band) from a signal transmitted via the cable 110 and a signal that is transmitted to the head end 100 via the cable 110. High-frequency amplifier 320 for output, modem LSI 330 having modulation / demodulation function of transmission / reception signals, read-only flash memory 360 for storing programs, initial setting data, and the like, data generated by modem LSI 330 for reading and writing at any time A memory (RAM) 370 and the like.
The modem LSI 330 is formed on a single semiconductor substrate such as single crystal silicon, an AD conversion circuit 331 that converts a received analog signal into a digital signal, and a demodulator that demodulates received data and reproduces I and Q signals. A circuit 332 is provided. Also, the modem LSI 330 generates an I / Q symbol from the reproduced I signal and Q signal, converts the bit configuration, and reproduces the original data, and receives data based on the error correction code. An error correction circuit 334 that corrects errors in the network, a MAC (Medium Access Control) circuit 335 that controls access to the local area network, a user such as Ethernet, USB (Universal Serial Bus), and SLIC (Subscriber Line Interface Circuit) are available. Network interface 336 having ports, CP for network processing such as initial data setting and protocol control, signal processing such as compression and decompression of voice data, and overall control inside LSI U (central processing unit, hereinafter referred to as a processor) 340, a data bus 338 connecting the processor 340 and the memories 360 and 370, and a data interface 337 for transferring data between the network interface 336, based on transmission data An ECC generation circuit 341 that generates an error correction code, a mapping circuit 342 that converts the transmission data into bits and generates an I / Q symbol, and performs QPSK or 16QAM digital modulation processing based on the generated I and Q signals A modulation circuit 343, a DA conversion circuit 344 that converts the modulated data into an analog signal, and the like are provided.
FIG. 3 is a block diagram showing in more detail the demodulation circuit 332 of the modem LSI 330 constituting the cable modem device in FIG.
As shown in FIG. 3, the demodulation circuit 332 reproduces the carrier wave from the output of the AD conversion circuit 331 and generates the sampling clock for the AD conversion circuit, and the output signal of the AD conversion circuit 331. A Nyquist filter 3322 for providing band limitation and removing interference between codes, a mixer (multiplier) 3323a for branching the output signal of the Nyquist filter 3322 and mixing a sine wave component of a carrier wave to reproduce an I signal, and a Nyquist filter A mixer 3323b that reproduces the Q signal by mixing the cosine wave component of the carrier wave with the other branch signal of the output signal of 3322, and performs a filtering process (noise removal) by multiplying the reproduced I signal and Q signal by respective appropriate coefficients. Performing adaptive equalization filters 3324a and 3324b, rotating the phase of the filter output A derotator 3325 that corrects phase distortion received on the transmission path (cable), slicers 3326a and 3326b that reproduce I / Q symbols that have been subjected to gradation conversion by discriminating signals corrected by the derotator 3325 at appropriate levels, and the slicers Based on the output signal of filter coefficient generation circuits 3327a and 3327b and slicers 3326a and 3326b that generate filter coefficients in the adaptive equalization filters 3324a and 3324b from the signals before and after slicing by the signals 3326a and 3326b. An arithmetic circuit 3328 for calculating a carrier frequency offset and noise power is formed.
Although not shown in FIG. 2, a timer circuit 345 for giving a timer interrupt to the processor 340, a bit error rate and symbol error rate detected by the error correction circuit 334, a flag indicating whether or not error correction has been performed, and the like. A register REG1 for holding, registers REG2a and REG2b for holding reproduced I / Q symbols, registers REG3a and REG3b for holding filter coefficients, and the like are provided. Further, the arithmetic circuit 3328 is provided with a register REG4 that holds a phase error, a frequency offset value, noise power, and the like, which are calculation results. The values of the registers REG1 to REG4 can be read by the processor 340 via the register access bus 346.
The registers REG1 to REG4 are configured to have a port for outputting to the register access bus 346 in addition to an original signal output port. The output port on the bus side is a tri-state port to avoid collision between the outputs of each register. When any of the registers connected to the bus outputs retained data to the bus, The output port on the bus side of the register is in a high impedance state.
As a method for selectively outputting the value of each register to the bus 346, for example, an identification code is assigned to each register, and a decoder is provided in the vicinity of each register, and the identification code of the register that the processor 340 wants to read is output. Then, there is a method in which the decoder decodes this to generate a control signal and opens the output port on the bus side of the corresponding register with this control signal. The register identification code can be any address not included in the address range of the memories 360 and 370.
In the modem LSI 330 of the embodiment of FIG. 3 to which the present invention is applied, a port 351 using an input / output port defined by a standard called JTAG (Joint Test Action Group) is provided. Signals can be directly input to and output from the processor 340 from the outside of the chip. This port is not a port that is open to allow the user to expand the system, but is generally a port for connecting an emulator or the like for debugging a program executed by an on-chip processor.
In this embodiment, a communication failure diagnosis computer 400 (hereinafter referred to as a diagnosis computer) 400 is connected to the port 351 via a data transfer means 410. The diagnostic computer 400 stores a program for giving a command to the processor 340 in the modem LSI, loading a diagnostic program to the processor 340, post-processing data received from the processor 340, displaying an observation result, and the like. ing.
The data transfer means 410 is a device having a function of transferring a command or a diagnostic program given from the diagnostic computer 400 to the processor 340 in the modem LSI, or transferring data to be output from the processor 340 to the diagnostic computer 400. It is. A device having such a function may be configured as a dedicated device. For example, if an emulator for debugging the program in the flash ROM 360 by controlling the processor 340 in the modem LSI from the outside is provided. Since it is possible to use, there is no need to newly design.
Under such a configuration, the diagnostic computer 400 loads a diagnostic program to the processor 340 and gives a command, whereby the processor 340 reads information held in the registers REG1 to REG4 in the chip and sends it to the external diagnostic computer 400. To be sent. The registers REG1 to REG4 hold information (bit error rate, I / Q symbol, adaptive equalization filter coefficient, phase error, carrier frequency offset, etc.) indicating the current reception state. By reading this value, the reception state can be grasped more accurately. Moreover, since most of the registers are registers provided in the modem LSI, the amount of hardware increased by applying the present invention can be extremely small.
FIG. 4 shows a modem LSI 330 to which another embodiment of the present invention is applied and a fault diagnosis system using the same. In FIG. 4, the same or equivalent circuits as in FIG.
In this embodiment, a diagnostic program is sent to the processor 340 by a terminal device 150 such as a personal computer connected to the network interface 336 provided in the modem LSI 330 to enable services such as the Internet using the cable network of cable television. A command is given by loading, and the processor 340 is configured to be able to read the information held in the registers REG1 to REG4 in the chip and transmit it to the terminal device 150.
As the terminal device 150, a local network terminal used by the end user can be used as well as a personal computer brought in by an engineer dispatched to the end user. The diagnostic program can be given from a storage medium such as a CD-ROM.
In addition, the modem LSI 330 of this embodiment shows a JTAG port 351 to which a diagnostic computer 400 can be connected as in the above embodiment, but this port can be omitted. However, by providing the JTAG port 351, the diagnostic computer 400 is connected without changing the user system configuration, the processor 340 is loaded with a diagnostic program, and a command is given to the processor 340. The information held in REG4 can be read out and transmitted to the external diagnostic computer 400, and the communication state can be observed.
Further, the modem LSI 330 of this embodiment differs from the modem LSI 330 of the embodiment of FIG. 2 in the bus configuration, and the MAC circuit 335 is not a network interface 336 but a system bus interface (corresponding to the data interface of FIG. 2). 338.
Although not shown in FIG. 4, in the modem LSI of the embodiment of FIG. 4 as well, ports such as Ethernet, USB, and SLIC are connected to the network interface 336 in the same manner as in the embodiment. Further, in FIG. 4, the modulation unit and the demodulation unit are simplified and shown as one block, but these modulation units and the demodulation unit are also configured in the same manner as the circuits shown in FIGS.
Next, the modulation / demodulation method in the modem LSI of the above embodiment will be described by taking 16QAM as an example. In 16QAM modulation, data to be transmitted is assigned to a signal called a 2-bit symbol for each I / Q. Specifically, as shown in FIG. 5A, four I / Q two-bit four states are assigned to +1, -1, +3, and -3.
Each I / Q is mixed and added by carrier waves sin (wt) and cos (wt) whose phases are different by 90 °. Since the carrier phase differences are orthogonal, the I / Qs do not interfere with each other. Since the I / Q symbol is discrete, the frequency band of the modulated signal is infinitely widened, and is transmitted after being band-limited by a Nyquist filter. At this stage, the transmission signal is a baseband signal. When the frequency characteristic of this signal is represented on a complex plane, it is discrete at 16 points as shown in FIG. That is, these 16 points correspond to the input I / Q total 4 bits 16 types. This is why it is called 16QAM. When 1 bit is given to each I / Q, QPSK (4QAM), when 3 bits are given, 64 QAM is given, and when 4 bits are given, 256 QAM is given. A point appearing on the complex plane in FIG. 5B is called a constellation. A carrier wave having a predetermined frequency is modulated by the baseband signal having such frequency characteristics, further converted into an analog signal, and transmitted to the transmission line.
On the demodulation side, as shown in FIG. 6, the received signal is converted into a digital signal by the AD conversion circuit 331, the frequency and phase of the carrier wave are recovered by the carrier wave recovery circuit 3321, and the band is limited by the Nyquist filter 3322. Thereafter, carrier waves sin (wt) and cos (wt) whose phases are different from each other by 90 ° are mixed with each other, thereby reproducing the I / Q symbol before modulation.
The adaptive equalization filter 3324 has a function of correcting distortion received by the cable 110 serving as a transmission path. In this embodiment, the signal derot-i / q before discrimination in the slicer 3326 and the signal slice- after discrimination. An adaptive algorithm that updates the filter coefficient by obtaining an error from i / q is employed.
The derotator 3325 has a function of rotating the phase by calculating the following determinant (1) with respect to the output eq−i / q of the adaptive equalization filter 3324, and the rotation angle θ is discriminated by the slicer 3326. It is obtained by calculating the phase error from the previous signal derot-i / q and the signal slice-i / q after discrimination.

The output of the derotator 3325 is a signal reproduced by various signal processing, and the signals derot-i / q input to the slicers 3326a and 3326b are each expressed by an 8-bit signal, for example. When the modulation method is 16QAM, each I / Q has 2-bit information, and therefore, it is the slicer 3326a, 3326b that takes out 2-bit 4-gradation signals equally from 8-bit 256-gradation signals. Is the action.
The demodulation circuit is configured to calculate an error by comparing the demodulation result and its expected value, and to converge to an optimal demodulation condition by feeding back the error component. In the demodulation circuit of this embodiment, the slicer Is used as the expected value. In the above description, in order to facilitate understanding, the case where the modulation system is 16QAM is taken up. However, it is specified that 64QAM or 256QAM is used in the downlink from the cable modem to the head end. In this case, each of I and Q is sliced into 3 bits at 64QAM and 4 bits at 256QAM.
The demapper 333 has a function of converting the I / Q symbols discriminated by the slicers 3326a and 3326b into 8 bits or 7 bits which are error correction units. Then, the error correction circuit 334 performs a predetermined error correction process on the converted bit string and outputs the result.
The output signal derot-i / q of the derotator 3325 is an 8-bit signal, and is plotted on a two-dimensional plane of I / Q as shown in FIG. This is called constellation. FIG. 7 shows a case where the modulation method is 16QAM for easy understanding. It should be noted that how many bits are given to the I / Q depends on the design concept, and the 8 bits here is only an example.
If the signal is not affected by noise on the transmission line (cable), the received signal gathers at the point indicated by ● in FIG. 7, and a 4-bit binary signal as shown in parentheses by the demapper 333 Is converted to However, in actuality, the signal appears as a signal away from the ideal point (●) as shown by a cross in FIG. 7 due to the influence of noise on the transmission line.
FIG. 8A shows an ideal constellation, FIG. 8B shows white noise, and FIG. 8C shows a constellation including a phase error. When the received signal receives white noise, the closer it is to the ideal point where the noise is smaller, the higher the appearance probability and the normal distribution centering on the one with the smaller amplitude, so each received signal is stochastically around the ideal point. The constellation appears as a shaded pattern as shown in FIG. 8B. On the other hand, when the phase error is included, the constellation is rotated as shown in FIG. 8C.
The phase error θ and the noise power NP can be obtained by calculation from the constellation as shown in FIG. That is, the phase error θ is an angle formed by a line connecting the 0 point of the I / Q and the ideal point (● marked slice result), and a line connecting the 0 point and the point of interest (× marked signal processing result), and noise. The power NP is the sum of the squares of the distance e from the ideal point to each point (Σe ² ). In the present embodiment, this calculation is performed in the demodulation circuit, and the results are stored in the corresponding registers. The phase error is calculated as the rotation correction angle in the derotator 3325, and the noise power is originally calculated to obtain the coefficient in the adaptive equalization filter 3324, and it is necessary to add it for the communication failure diagnosis of the present invention. It is not a certain function.
Next, an example of a communication failure diagnosis procedure using the diagnosis computer 400 in the embodiment of the present invention will be described with reference to the flowchart of FIG. 11 and the data flowchart of FIG.
When a failure occurs in the end user, the cable management company that receives the notification dispatches an engineer. The dispatched engineer uses the emulator 410 or the like to connect the diagnostic computer 400 to the cable modem device 330 in which the failure has occurred (step S1). Then, the diagnostic program is transferred from the diagnostic computer 400 to the processor 340 of the modem LSI 330 in the cable modem device 130 (step S2). Then, the processor 340 stores the received diagnostic program in the memory 370 connected to the CPU. The data flow at this time is indicated by reference numeral (1) in FIG.
After that, when the engineer inputs a start command or observation condition from the input device 401 such as a keyboard, the diagnostic computer 400 gives a command for starting the diagnostic program to the processor 340 (step S3 in FIG. 11, symbol ▲ 2 in FIG. 12). ▼). Then, the processor 340 executes the diagnostic program and processes the setting of the observation target, the setting of the observation condition such as the observation cycle (timer interrupt cycle), the setting of the interrupt condition from the circuit inside the chip, and the register read information. Specific contents of the diagnostic process such as setting of preprocess contents to be output are set (step S4 in FIG. 11, reference numeral 3 in FIG. 12, etc.). As an interrupt condition, for example, an interrupt is generated when a correction flag of the register REG1 is set.
Subsequently, when the engineer inputs a measurement start command from the console 401, the diagnostic computer 400 gives a measurement start command to the processor 340 (step S5 in FIG. 11, reference numeral 2 in FIG. 12). Then, the processor 340 sequentially reads out the state of each register in accordance with the contents set in step S4 and stores it in the memory 370 (step S5 in FIG. 11, reference numeral 4 in FIG. 12). Then, if pre-processing is set, the set pre-processing is performed, and the observation data is displayed on the monitor screen, so that the communication state can be easily determined (step S6).
Next, the processor 340 transfers the observation data acquired in step S5 and the data preprocessed in step S6 to the diagnostic computer 400 (step S7 in FIG. 11, reference numeral 5 in FIG. 12). This observation data transfer may be performed in response to a transfer request command from the diagnostic computer 400, or the processor 340 may execute the transfer when a series of observation data is obtained. When the observation data is transferred, the diagnostic computer 400 performs post-processing such as emphasizing the data that exceeds the allowable range in red with respect to the received observation data (step S8). Then, the diagnostic computer 400 displays the observation result on the screen of the monitor 402 based on the post-processing data (step S9 in FIG. 11, reference numeral 6 in FIG. 12).
Thereafter, the engineer observes the display of the observation result and diagnoses the communication state to appropriately adjust the modem device (step S10). Then, the process returns to step S4 to perform observation, and the above processing is repeated until a desired communication state is obtained. If a satisfactory communication state can be realized as a result of adjustment, when the engineer inputs a diagnosis end command from the keyboard, the diagnosis computer 400 gives a stop command of the diagnosis program to the processor 340 in the modem LSI (see FIG. 11 step S11, sign (2) in FIG.
Furthermore, the diagnostic computer 400 sends a diagnostic program deletion command to the processor 340 (step S12 in FIG. 11, reference numeral 1 in FIG. 12). Then, the processor 340 deletes the diagnostic program from the memory 370 and ends the diagnostic process. The diagnostic program deletion command given from the diagnostic computer 400 to the processor 340 may be automatically given following the stop command, but can be made on condition that the engineer inputs a deletion command from the keyboard. After deleting the diagnostic program, the engineer removes the diagnostic diagnostic computer 400 from the modem device 130 and finishes the operation.
Although the diagnostic program may be deleted as described above, it may be left in the end user's modem device as it is. By deleting, it is possible to avoid reducing the available memory area of the processor 340, and leaving it without deleting it can facilitate diagnosis when the next communication failure occurs. For example, there is a possibility that the user himself / herself can diagnose the communication state with his / her network terminal and take appropriate measures.
FIG. 13 is a flowchart showing a control procedure of the diagnostic computer 400.
When a diagnostic program transfer command is input from a keyboard or the like, the diagnostic computer 400 transfers the diagnostic program to the processor 340 in the modem LSI 330 (step S11). Thereafter, when a command is input from the keyboard, the command is analyzed and if it is an end command, the processor 340 is instructed to stop the diagnostic program, and if it is a delete command, the program is deleted (steps S14 → S20). If the command is other than the end command and the delete command, the corresponding control command is sent to the processor 340 to perform observation control (step S15).
Here, the observation control includes starting and stopping of a diagnostic program, setting of an observation target, setting of an observation period, setting of preprocessing contents, and the like. When the observation content is changed in step S15, the control content of the result display on the diagnostic computer side is changed in accordance with the change (step S16). Furthermore, the diagnostic computer 400 requests the modem LSI 130 to transfer the observation result regardless of the presence or absence of a command (step S17). Then, post-processing is performed on the transferred observation data, and the observation result is displayed on the screen of the monitor 402 in the form of a graph or the like as shown in FIG. 18 (steps S18 and S19).
In FIG. 18, reference symbol A shows the result of applying fast Fourier transform to the received signal and shows the frequency on the horizontal axis, reference symbol B is a constellation obtained by analyzing the frequency distribution of symbols, and reference symbol C is The code error shows the temporal change of the bit error rate, the code D shows the temporal change of the phase error, and the code E shows the temporal change of the noise power. In this embodiment, the fast Fourier transform on the received signal is performed as preprocessing by the processor 340 of the modem LSI 130.
It should be noted that the transfer request from the diagnostic computer 400 to the modem LSI 130 performed in step S17 can be periodically made by a timer interrupt. In the case of unilateral data transfer from the modem LSI 130 instead of a transfer request from the diagnostic computer 400, data can be received by a reception interrupt or the like.
14 to 17 are flowcharts showing the control procedure of the diagnostic program executed by the processor 340.
Of these, FIG. 14 shows a command processing routine. When a command is input from the diagnostic computer 400, the processor 340 analyzes the command (step S21). As a result of the analysis, when the input command is a program deletion command, the diagnostic program stored in the memory 370 is deleted. When the input command is an observation target setting command, a measurement instruction is given by setting a flag provided on the software according to the observation target item (step S23).
If the input command is an observation cycle setting command, a timer interrupt cycle is set in the timer 345 (step S24). If the input command is a preprocessing content setting command, the preprocessing content to be executed is set (step S25). Further, when the input command is a data transfer request command, the measurement result data stored in the memory 370 as a result of the observation is read and output to the diagnostic computer 400 (step S26).
FIG. 15 shows an interrupt processing routine having a long cycle such as 1 second, which is executed by the interrupt signal output from the timer 345.
In this routine, the flag set in step S23 of FIG. 14 is checked to grasp the observation target item, and the register value of the observation target item with the flag set is sequentially read and stored in the memory 370. However, for those that do not change, such as the coefficient of the adaptive equalization filter, when the command is issued once, reading from the coefficient registers REG3a and REG3b is executed and transferred, and the flag is cleared. (Steps S31 to S34). This is because the coefficient of the adaptive equalization filter changes during the pull-in operation, but is constant unless the transmission path state changes once it converges.
On the other hand, for example, phase error, frequency offset, noise power, bit error rate, symbol error rate, correction flag, and the like other than the coefficients of the adaptive equalization filter are read from the registers REG1 and REG4 and stored in the memory 370 every second. (Steps S35 to S46). In this way, by simultaneously reading out the contents of the registers and storing them in the memory, it is possible to improve the data acquisition simultaneity even when the acquired data is processed.
In other words, instead of moving to the acquisition and processing of the next data after acquiring some data and storing it in the memory, the contents of the registers are temporarily stored in the memory and then stored in the memory. When processing is performed, the simultaneity of the acquired data is high, and the accuracy in examining the relationship between the data becomes high.
The period for reading data from the register is not limited to 1 second, and can be set freely. The shorter the cycle, the higher the possibility that a single transmission line failure can be caught. On the other hand, the load on the processing capacity of the processor 340 and the storage capacity of the memory 370 increases, and the normal processing is slowed down. What is necessary is just to determine a period according to the performance etc. of.
According to the routine of FIG. 15, the processor 340 sequentially reads out the phase error, frequency offset, noise power, bit error rate, symbol error rate, correction flag, etc. from the registers REG1 and REG4. Since the value of the register does not change abruptly, it can be considered that reading at a period of 1 second is substantially simultaneously observed.
Further, if the processor 340 has an average performance in the same type of LSI, even if the diagnostic program is executed in parallel with the execution of the normal program for realizing the original function of the modem LSI, the processor 340 is not so burdensome. Therefore, the processing capacity of the processor 340 is not compressed. Even if the diagnostic program is executed in parallel with the normal program and the measurement result is output through a serial interface such as JTAG, the actual operation of the processor 340 is not seriously affected.
16 and 17 are a routine for observing an I / Q symbol and performing preprocessing to generate data indicating a constellation, and a routine for outputting it. Of these, the routine of FIG. 16 is a timer interrupt process executed at a short cycle such as 1 msec, and the routine of FIG. 17 is a timer interrupt process executed at a long cycle such as 1 sec.
When the interrupt process shown in FIG. 16 is started, it is first determined whether or not the observation target flag of the I / Q symbol is set. If it is set, the symbol is read from the registers REG2a and REG2b (step S52). Then, the frequency distribution analysis of symbols as described later is performed and the analysis result is stored in the memory 370 (steps S53 and S54).
Such processing is performed every 1 ms, and the frequency distribution data stored in the memory is transferred to the diagnostic computer 400 every second by the interrupt processing of FIG. 17, and then the accumulated frequency in the memory is cleared. (Steps S61 and S62).
Thereby, the change of the constellation can be observed in real time by the monitor 402 of the diagnostic computer. Further, the I / Q symbol data read from the registers REG2a and REG2b is not stored in the memory 370 as it is, but is stored after being replaced with the frequency distribution, so that the amount of use of the memory 370 is very small. When the data is transferred to the diagnostic computer 400, the data amount is small and the transfer speed can be reduced.
In the analysis of the symbol frequency distribution in the above step S52, the processor 340 virtually maps the I / Q two-dimensional plane into 8 × 8 regions (4 × 4 regions or the like are also possible) as shown in FIG. It is determined whether the I / Q symbol read from the registers REG2a and REG2b belongs to which area is divided, the number of symbols that appear is accumulated for each area, and the accumulated value is stored in the memory. This is done.
Then, the diagnostic computer 400 generates a shading pattern in accordance with the accumulated value of each region, whereby a constellation as indicated by symbol B in FIG. 18 is displayed on the monitor. FIG. 10 is an enlarged view of the central part of the center divided by the broken line in the I / Q plane of FIG. 9. Similarly, the other parts are divided into regions and the number of symbols for each region is shown. Accumulation is performed.
The conventional fault diagnosis method is a method in which a pin for outputting the contents of a register to be observed to the outside of the chip is provided in the modem LSI, and data from the pin is displayed on the monitor. Although only types of data could be observed, according to the fault diagnosis apparatus of this embodiment, observation results relating to a plurality of types of data are displayed on one screen as shown in FIG. You can know how other data is changing when it changes. As a result, there is an advantage that comprehensive judgment can be easily made.
In addition, in the conventional fault diagnosis method, if there is a need to increase the data to be observed, the number of pins must be increased, and there is a problem that the chip size increases, but in the method of the present invention, only a JTAG port is provided. good. Moreover, this JTAG port can be used for connecting an emulator for debugging a normal program of a modem LSI, which is very convenient.
Next, another embodiment of the present invention will be described with reference to FIGS.
Of these, the embodiment of FIG. 19 includes the output of the AD converter circuit 331, the output of the Nyquist filter 3322, the output of the adaptive equalization filters 3324a and 3324b, the output of the derotator 3325, and the slicers 3326a and 3326b in the chip of the modem LSI 330. The DMA controller 380 for sampling the received data of each part in the modem LSI and directly transferring it to the memory 360 is provided. FIG. 19 shows an example of an interrupt control circuit 390 that interrupts the processor 340 when a predetermined interrupt condition is satisfied during processing of received data.
The interrupt control circuit 390 in this embodiment includes a register 391 that holds a bit error rate threshold, a register 392 that holds a symbol error rate threshold, and the bit error rate in the register REG1 and the threshold of the error rate. A comparison circuit 393 for comparing the values, a comparison circuit 394 for comparing the symbol error rate in the register REG1 and the threshold value of the symbol error rate, outputs of these comparison circuits 393 and 394, and an error correction flag in the register REG1 An interrupt factor selection / mask circuit 395 for selecting an interrupt factor based on the above and generating an interrupt signal or masking the interrupt is configured.
By operating the DMA controller 380 in conjunction with an interrupt from the interrupt control circuit 390, efficient diagnosis can be performed using a small amount of memory. For example, if the error correction flag indicates uncorrectable or the bit error rate exceeds a predetermined threshold, transfer of received data to the memory is started, or conversely, the error correction flag indicates uncorrectable or the bit error rate is By stopping the transfer of the received data to the memory when the predetermined threshold is exceeded, the received data immediately before the occurrence of the phenomenon of increasing bit errors and the received data immediately after can be stored in the memory.
This makes it possible to diagnose a fault that rarely occurs by capturing the phenomenon. When storing the received data immediately before a certain phenomenon occurs in the memory, if the available area of the memory is small, the new data is overwritten sequentially on the old data when the capacity is exceeded, and the latest data is always updated. Data should remain.
In the embodiment of FIG. 20, the diagnostic computer 400 connected to the cable modem device 130 is connected to the head end 100 of the base station via a transmission medium other than a cable of cable television such as a general telephone line. The headend 100 and the diagnosis computer 400 are connected to the control computer 200 and operated in synchronization or in conjunction with each other. According to this embodiment, for example, a signal or standard pattern as a reference is transmitted from the head end, and the state in which the cable modem apparatus receives the signal can be observed by the diagnostic computer 400.
Similar to the embodiment of FIG. 4, a local network terminal 150 such as a personal computer connected to the cable modem 130 of the end user is used as a diagnostic computer, and the communication means possessed by the local network terminal 150 is used. It is also possible to make a diagnosis by connecting to the control computer 200 of the base station via a telephone line or the like.
In this case, a diagnostic program is transmitted from the cable management company computer 200 to the end user's local network terminal 150 via a telephone line or the like, and observation data is transmitted from the local network terminal 150 to the cable management company computer 200. By doing so, diagnosis can be performed without dispatching engineers. Further, data for adjusting the inside of the modem can also be transmitted from the computer 200 of the cable operating company to the local network terminal 150 so that the internal control state is adjusted by the processor 340 inside the modem LSI.
The embodiment of FIG. 21 uses the local network terminal 150 as a diagnostic computer and uses a communication function such as a voice modem or USB provided in the network interface 336 of the cable modem device 130 via a telephone line or the like. This is an example of a system connected to a base station control computer 200 for diagnosis. In FIG. 20, reference numeral 210 denotes a relay device provided in the middle of the telephone line.
The invention made by the present inventor has been specifically described based on the embodiments. However, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the scope of the invention. Nor. For example, in the embodiments, the cable TV cable is described as being a coaxial cable or an optical fiber, but the present invention can also be applied to a case where the cable network is configured by a telephone line, a power line, a wireless communication network, or the like. The present invention can be used for a modem LSI in a communication network capable of bidirectional communication.
In the embodiment, the case where the engineer dispatched to the end user loads the diagnostic program into the memory of the modem device has been described. However, the diagnostic program may be stored in the memory such as the flash ROM 360 in advance. Furthermore, the communication medium other than the cable for connecting the diagnostic computer on the end user side or the network terminal and the control computer for the base station is not limited to a telephone line, and any communication means may be used.
Industrial applicability
In the above description, the case where the invention made mainly by the present inventor is applied to an LSI for a cable modem that performs data communication using a cable network of a cable television, which is a field of use that has become the background, has been described. However, the present invention is not limited thereto, and can also be used for a modem for data communication in satellite broadcasting using a communication satellite and a satellite communication system.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a system configuration example of a cable television suitable for applying the present invention.
FIG. 2 is a block diagram illustrating a configuration example of the cable modem device.
FIG. 3 is a block diagram showing a configuration example of a modem LSI demodulating circuit constituting the cable modem device of FIG.
FIG. 4 is a block diagram showing a modem LSI to which another embodiment of the present invention is applied and a fault diagnosis system using the same.
FIG. 5 is an explanatory diagram showing the principle of 16QAM modulation.
FIG. 6 is an explanatory diagram showing the principle of 16QAM demodulation.
FIG. 7 is an explanatory diagram illustrating a constellation in which 16QAM demodulated data is plotted on an I / Q two-dimensional plane.
FIG. 8 is an explanatory diagram showing an ideal received data constellation and a received data constellation subjected to disturbance.
FIG. 9 is an explanatory diagram showing the magnitude of the phase error and the noise power.
FIG. 10 is an explanatory diagram showing an example of a region division method for analyzing the frequency distribution of I / Q symbols.
FIG. 11 is a flowchart illustrating an example of a communication failure diagnosis procedure using the diagnosis computer according to the embodiment of the present invention.
FIG. 12 is a data flowchart showing a data flow in communication failure diagnosis according to the flowchart of FIG.
FIG. 13 is a flowchart showing an example of a control procedure on the diagnostic computer side in the communication failure diagnosis method of the present invention.
FIG. 14 is a flowchart showing an example of a command processing procedure on the modem LSI side in the communication failure diagnosis method of the present invention.
FIG. 15 is a flowchart showing an example of a timer interrupt processing procedure on the modem LSI side in the communication failure diagnosis method of the present invention.
FIG. 16 is a flowchart showing an example of a preprocessing procedure of observation data on the modem LSI side in the communication failure diagnosis method of the present invention.
FIG. 17 is a flowchart showing an example of the transfer procedure of observation data preprocessed by the modem LSI.
FIG. 18 is a display screen diagram showing a display example of observation data displayed on the monitor of the diagnostic computer in the communication failure diagnosis method of the present invention.
FIG. 19 is a block diagram showing another example of the configuration of a modem LSI demodulation circuit constituting the cable modem device.
FIG. 20 is a block diagram showing another embodiment of the communication failure diagnosis method of the present invention.
FIG. 21 is a block diagram showing still another embodiment of the communication failure diagnosis method of the present invention.

Claims (14)

Formed on one semiconductor substrate,
A demodulation circuit for demodulating a signal received via a cable modem communication network;
A register provided in the demodulation circuit and reflecting a reception state;
A processor capable of reading the value of the register;
A communication semiconductor integrated circuit comprising a port for outputting a value read from the register by the processor to the outside. An error correction circuit for correcting an error in received data demodulated by the demodulation circuit is further provided, and the register includes a register for holding information related to processing of the error correction circuit. 2. The semiconductor integrated circuit for communication according to 1. 2. The communication semiconductor integrated circuit according to claim 1, wherein the register includes a register that captures and holds the reception data demodulated by the demodulation circuit. The communication semiconductor integrated circuit according to claim 1, wherein the processor is configured to process a value read from the register and output the value to the outside from the port. 5. The communication semiconductor integrated circuit according to claim 1, wherein the port is a port different from a port to which a device constituting a network is connected. It is further provided with a memory circuit that can be read and written as needed, and the processor is configured to temporarily store a value read from the register or a result obtained by processing the value in the memory and then output the value to the outside from the port. The semiconductor integrated circuit for communication according to any one of claims 1 to 5. A semiconductor integrated circuit for communication according to any one of claims 1 to 5 and a semiconductor memory capable of reading and writing as needed, or a semiconductor integrated circuit for communication according to claim 6,
Signal extraction means for extracting a signal of a predetermined frequency from a signal supplied via a cable television communication network;
A modulation circuit for modulating the transmission signal;
Transmitting means for outputting a modulated signal to the communication network;
A network port to which a local network terminal can be connected;
A modulation / demodulation device comprising: A diagnostic method for diagnosing a communication state in the modem according to claim 7,
A diagnostic computer is connected to the port of the modem via a data transfer means, a diagnostic program is sent from the diagnostic computer to the processor, and the processor sends the received diagnostic program to the memory circuit or semiconductor By storing in the memory and executing the diagnostic program, the value of the register is read out and transferred to the diagnostic computer. The diagnostic computer displays the received values of the plurality of registers on the screen of the display device. A communication state diagnosis method characterized by displaying. The communication state diagnosis method according to claim 8, wherein the processor processes the value read from the register and outputs the value to the diagnostic computer from the port. The processor periodically processes the value read from the register and stores it in the memory circuit or semiconductor memory, reads the stored value from the memory circuit or semiconductor memory, and outputs the stored value from the port to the diagnostic computer. The communication state diagnosis method according to claim 8. The diagnostic computer is connected to a control device on the base station side of the cable television via a communication path other than the cable television communication network, and the data output from the modem device via the diagnostic computer is transmitted from the communication path. The communication state diagnosis method according to claim 8, wherein the communication state is transmitted to the control device on the base station side. A diagnostic method for diagnosing a communication state in the modem according to claim 7,
A diagnostic program is sent to the processor from a local network terminal connected to the modem via the network port, the processor stores the received diagnostic program in the memory circuit or semiconductor memory, and the diagnostic A communication state, wherein the register value is read out and transferred to the network terminal by executing a program for the network, and the network terminal displays the received values of the plurality of registers on the screen of the display device Diagnosis method. The network terminal is connected to a control device on the base station side of the cable television via a communication path other than the cable television communication network, and the data output from the modem device via the network terminal is transmitted to the base via the communication path. The communication state diagnosis method according to claim 12, wherein the communication state diagnosis method transmits to a control device on the station side. A diagnostic method for diagnosing a communication state in the modem according to claim 7,
The modem is connected to a control device on the base station side of the cable television via a communication path other than the cable television communication network, and a diagnostic program is transmitted from the control device to the modem via the communication path. A communication state diagnosis method comprising transmitting data output from the modem to the control device on the base station side via the communication path.

Images