Data transmission involving multiplexing and demultiplexing of embedded clock signals
25 claims: 6 independent, 19 dependent
- 1データ伝送システムであって、 第1ノード(100)であって、少なくとも2組の入力データ信号(d−in1、d−in2)を受信し、前記少なくとも2組は、クロック信号がそれぞれ組み込まれる少なくとも2つのデータ信号を含み、時間分割多重、TDM、構造による伝送のための前記組をフォーマットする、ように構成される、第1ノードと、 回線周波数を有するビットストリーム(bs、bs1、bs2)として前記TDMフォーマット信号を送信するように構成される伝送媒体(L、L1、L2)と、 第2ノード(200、201、202)であって、前記ビットストリーム(bs、bs1、bs2)を受信し、前記少なくとも2組の入力データ信号(d−in1、d−in2)が少なくとも一つの第1信号および少なくとも一つの第2信号を含み、前記少なくとも一つの第1信号は、前記少なくとも一つの第2信号が基づく同期ソースとは異なる同期ソースに基づくことを特徴とする、少なくとも2組の出力データ信号(d−out、d−out1、d−out2)へ前記ビットストリーム(bs、bs1、bs2)を逆多重化する、ように構成される、第2ノードと、 を備え、 前記第1ノード(100)は、 前記少なくとも一つの第1および第2信号から前記異なる同期ソースのそれぞれを示す各クロック信号(CLKe1、CLKex2)を抽出するように構成される少なくとも一つのクロック抽出モジュール(111、112)と、 前記回線周波数で同期されたサンプリング周波数(smp)に基づく各結果として生じたサンプリングクロック信号(CLKsp1、CLKsp2)へ前記抽出クロック信号(CLKe1、CLKex2)のそれぞれをサンプリングするように構成される少なくとも一つのサンプリングモジュール(131、132)と、 前記TDM構造において各独立信号として前記結果として生じたサンプリングクロック信号(CLKsp1、CLKsp2)のそれぞれを含むように構成される多重化モジュール(140)と、 を備え、 前記第2ノード(200、201、202)は、前記受信したビットストリーム(bs、bs1、bs2)を、少なくとも2組の出力データ信号(d−out、d−out1、d−out2)と、前記結果として生じたサンプリングクロック信号(CLKsp1、CLKsp2)を示す一連の逆多重化されたクロック信号(CLKdm、CLKdm1、CLKdm2)と、に逆多重化するように構成される逆多重化モジュール(210)と、前記一連の逆多重化されたクロック信号(CLKdm、CLKdm1、CLKdm2)における各信号に対して、所定のレベル以下に周波数ジッタの量を減少し、従って、前記一連の逆多重化されたクロック信号(CLKdm、CLKdm1、CLKdm2)における信号の同期品質よりも上位の同期品質を有する各クロック信号を生成するように構成される少なくともジッタ減衰手段(230、231、232、271、272)と、前記少なくとも2組の出力データ信号(d−out、d−out1、d−out2)における各データ信号を、その関連するクロック信号(CLKstb、CLKstb1、CLKstb2)と再結合させ、データ信号の組(d−res、d−res1、d−res2)において各結果として生じるクロック運搬データ信号を生成するように構成される少なくとも一つのインターフェースモジュール(220、221、222)と、 を備える、データ伝送システム。
- 2前記第2ノード(200、201、202)は、一連の逆多重化クロック信号における各信号において、周波数ジッタの量を所定のレベル以下に低減させ、従って、各安定化クロック信号(CLKstb、CLKstb1、CLKstb2)を生成するように構成される少なくとも一つの狭帯域ジッタ減衰手段(230、231、232)を含む、請求項1に記載のデータ伝送システム。
- 3前記少なくとも一つの狭帯域ジッタ減衰手段(230)は、問題の前記出力データ信号(d−out)と関係がある前記逆多重化クロック信号(CLKdm)によって定義される入力レートで前記出力データ信号(d−out)のビットを連続して受信し、前記逆多重化データ信号の所定の数のビットを一時的に格納し、その後、回線アダプタモジュール(320)へ前記ビット(bts)を送り出すように構成され、前記ビット(bts)は、問題の前記出力データ信号(d−out)と関係がある前記逆多重化クロック信号(CLKdm)に基づく前記安定化クロック信号(CLKstb)によって特定される最終出力レートで連続して送り出され、前記回線アダプタモジュール(320)は、前記最終出力レートで出力ターミナルへ前記出力ビット(bts)を連続して転送し、従って前記結果として生じるクロック運搬データ信号(d−res)を示すように構成される、バッファモジュール(310)と関係がある、請求項2に記載のデータ伝送システム。
- 4前記少なくとも一つの狭帯域ジッタ減衰手段(230、231、232)の少なくとも一つは、前記少なくとも一つのインターフェースモジュール(220、221、222)の少なくとも一つに組み込まれる、請求項3に記載のデータ伝送システム。
- 5前記第2ノード(200)は、 各逆多重化クロック信号(CLKdm1、CLKdm2)を受信し、 前記逆多重化クロック信号において周波数ジッタを閾値周波数を超えて低減させ、それによって、結果として生じる除去逆多重化クロック信号(CLKdm1’、CLKdm2’)を生成し、 前記少なくとも一つの狭帯域ジッタ減衰手段(230、231、232)へ前記結果として生じた除去逆多重化クロック信号(CLKdm1’、CLKdm2’)を転送する ように構成される、少なくとも一つの広帯域ジッタ減衰モジュール(271、272)を含む、請求項1〜4のいずれか1項に記載のデータ伝送システム。
- 6前記少なくとも一つの広帯域ジッタ減衰モジュール(271、272)の少なくとも一つは、前記周波数ジッタを閾値周波数を超えて低減させるように構成される位相ロックループ回路を含む、請求項5に記載のデータ伝送システム。
- 7前記少なくとも一つの広帯域ジッタ減衰モジュール(271、272)の少なくとも一つは、前記周波数ジッタを閾値周波数を超えて低減させるように構成される共振回路を含む、請求項5または6に記載のデータ伝送システム。
- 8前記少なくとも一つの広帯域ジッタ減衰モジュール(271、272)の少なくとも一つは、 各逆多重化クロック信号(CLKdm1、CLKdm2)を受信し、 多くのクロック周期を含む平均間隔で前記逆多重化クロック信号(CLKdm1、CLKdm2)のための平均周期時間を示す各平均周期長値を繰り返して生成し、 前記平均周期長値に基づいて前記除去逆多重化クロック信号(CLKdm1’、CLKdm2’)を生成する ように構成されるクロック再生器を含む、請求項5〜7のいずれか1項に記載のデータ伝送システム。
- 9前記逆多重化モジュール(210)は、前記異なる同期ソースのそれぞれについて読み出し信号(CLKo1、CLKo2)を生成するように構成され、前記読み出しクロック信号(CLKo1、CLKo2)は、クロックパルスの各列が前記TDM構造の各フレームにおいて含まれるビットの数と同等のクロックパルスの数を包含する、クロックパルスの列を含み、前記第2ノード(200)は、 読み出しクロック信号(CLKo1、CLKo2)によって特定される入力レートで連続して前記出力データ信号(d−out1、d−out2)のビットを受信し、 前記逆多重化データ信号(d−out1、d−out2)の前記受信したビットの所定の数を一時的に格納し、その後、 前記除去逆多重化クロック信号(CLKdm1’、CLKdm2’)によって特定される出力レートで連続して前記ビットを送り出す ように構成される少なくとも一つのバッファ手段(251、252)をさらに含む、請求項1〜8のいずれか1項に記載のデータ伝送システム。
- 10前記第1ノード(100)の前記少なくとも一つのサンプリングモジュール(131、132)は、問題の前記抽出クロック信号(CLKex1、CLKex2)の前記周波数と比べて1以上のオーバーサンプリングの係数を示すサンプリングしゅうはすうに基づいて前記抽出クロック信号(CLKex1、CLKex2)のそれぞれをサンプリングするように構成される、請求項1〜9のいずれか1項に記載のデータ伝送システム。
- 11前記第1ノード(100)は、各抽出クロック信号(CLKex1、CLKex2)を受信し、それに応えて、前記抽出クロック信号(CLKex1、CLKex2)の前記周波数の所定の割合を示す減少した周波数を有する結果として生じるダウンコンバート抽出クロック信号(CLKex1/F、CLKex2/G)を生成するように構成される少なくとも一つの分割モジュール(151、152)を含む、請求項10に記載のデータ伝送システム。
- 12前記第2ノード(200)は、前記クロック信号を前記インターフェースモジュール(221、222)へ送る前に、前記所定の割合(1/F、1/G)の反対(F、G)を示す係数を有する各逆多重化サンプリングクロック信号(CLK/F、CLK/G)、またはその除去バージョンを乗算するように構成される少なくとも一つの乗算器モジュール(261、262)を含む、請求項11に記載のデータ伝送システム。
- 13前記第1ノード(100)は、前記入力データ信号(d−in1、d−in2)のそれぞれのペイロード情報を示す各データ信号(d−ex1、d−ex2)を抽出するように構成される少なくとも一つのデータ抽出モジュール(121、122)を含み、 前記乗算モジュール(140)は、前記TDM構造における各独立信号として一連の前記抽出データ信号(d−ex1、d−ex2)において各データ信号を含むように構成される、 請求項1〜12のいずれか1項に記載のデータ伝送システム。
- 14データ伝送方法であって、第1ノード(100)において、少なくとも2組の入力データ信号(d−in1、d−in2)を受信することであって、前記少なくとも2組は、クロック信号がそれぞれ組み込まれ、時間分割多重TDM構造による伝送のための前記組をフォーマットする少なくとも2組のデータ信号を含む、受信することと、 前記第1ノード(100)から伝送媒体(L、L1、L2)へビットストリーム(bs、bs1、bs2)として前記TDMフォーマット信号を送信することであって、前記ビットストリーム(bs、bs1、bs2)は、回線周波数を有する、送信することと、 第2ノード(200、201、202)において前記ビットストリーム(bs、bs1、bs2)を、前記伝送媒体(L、L1、L2)を介して、受信することと、 少なくとも2組の出力データ信号(d−out、d−out1、d−out2)へ前記ビットストリーム(bs、bs1、bs2)を、第2ノード(200、201、202)において、逆多重化することと、 を含み、 少なくとも一つの第1信号および少なくとも一つの第2信号を含む前記少なくとも2組の入力データ信号(d−in1、d−in2)によって特徴付けられ、前記少なくとも一つの第1信号は、前記少なくとも一つの第2信号が基づく同期信号とは異なる同期信号に基づき、前記方法は、前記第1ノード(100)において、前記少なくとも一つの第1および第2信号から前記異なる同期信号を示す各クロック信号(CLKex1、CLKex2)を抽出することと、前記第1ノード(100)において、前記回線周波数と同期するサンプリング周波数(smp)に基づいて各結果として生じるサンプリングクロック信号(CLKsp1、CLKsp2)へ抽出クロック信号(CLKex1、CLKex2)をサンプリングすることと、前記第1ノード(100)において、前記TDM構造において各独立信号として前記結果として生じるサンプリングクロック信号(CLKsp1、CLKsp2)のそれぞれを多重化することと、前記第2ノード(200,201、202)において、前記受信したビットストリーム(bs、bs1、bs2)を、少なくとも2組の出力データ信号(d−out、d−out1、d−out2)と、前記結果として生じるサンプリングクロック信号(CLKsp1、CLKsp2)を示す一連の逆多重化クロック信号(CLKdm、CLKdm1、CLKdm2)と、に逆多重化することと、前記第2ノード(200、201、202)において、前記一連の逆多重化クロック信号(CLKdm、CLKdm1、CLKdm2)における各信号に対して、周波数ジッタの量を所定のレベル以下に低減させ、従って一連の逆多重化クロック信号(CLKdm、CLKdm1、CLKdm2)における前記信号の同期品質よりも上位の同期品質を有する各クロック信号を生成するようにジッタ減衰することと、前記第2ノード(200、201、202)において、前記少なくとも2組の出力データ信号(d−out、d−out1、d−out2)における各データ信号を、その関連するクロック信号(CLKstb、CLKstb1、CLKstb2)と再結合し、データ信号の組(d−res、d−res1、d−res2)における各結果として生じたクロック運搬データ信号を生成することと、 を含む、方法。
- 15前記第2ノード(200、201、202)において、周波数ジッタの量を所定のレベル以下に低減させ、従って各安定したクロック信号(CLKstb、CLKstb1、CLKstb2)を生成するように相対的に狭周波数帯において前記一連の逆多重化クロック信号(CLKdm、CLKdm1、CLKdm2)における各信号においてジッタ減衰することを含む、請求項14に記載の方法。
- 16前記狭帯域ジッタ減衰は、 前記出力データ信号(d−out)の連続して受信したビットの一時的な格納であって、前記ビットは、問題の前記出力データ信号(d−out)と関係がある前記逆多重化クロック信号(CLKdm)によって特定される入力レートで受信していた、ビットの一時的な格納と、その後、 回線アダプタモジュール(320)への前記ビット(bts)の送り出しであって、前記ビット(bts)は、問題の前記出力データ信号(d−out)と関係がある前記逆多重化クロック信号(CLKdm)に基づく前記安定化クロック信号(CLKstb)によって特定される最終出力レートで連続して送り出さる、前記ビットの送り出しと、 を含み、 前記回線アダプタモジュール(320)は、前記最終出力レートで出力ターミナルへ前記出力ビット(bts)を連続して転送し、従って、結果として生じるクロック運搬データ信号(d−res)を示すように構成される、 請求項15に記載の方法。
- 17周波数ジッタを前記一連の逆多重化クロック信号(CLKdm1、CLKdm2)における閾値周波数を超えて低減させ、それにより結果として生じる除去逆多重化クロック信号(CLKdm1’、CLKdm2’)を生成することと、 前記少なくとも一つの狭帯域ジッタ減衰手段(230、231、232)へ前記結果として生じる除去逆多重化クロック信号(CLKdm1’、CLKdm2’)を転送することと、 によって、前記第2ノード(200、201、202)において、比較的広周波数帯にける前記一連の逆多重化クロック信号(CLKdm1、CLKdm2)をジッタ減衰することを含む、請求項15または16のいずれか1項に記載の方法。
- 18周波数ジッタを前記一連の逆多重化クロック信号(CLKdm1、CLKdm2)における閾値周波数を超えて低減させることは、 多くのクロック周期を含む平均間隔で前記逆多重化クロック信号(CLKdm1、CLKdm2)のための平均周期時間を示す各平均周期長値を、繰り返し、生成することと、 前記平均周期長値に基づいて前記結果として生じる除去逆多重化クロック信号(CLKdm1’、CLKdm2’)を生成することと、 を含む、請求項17に記載の方法。
- 19前記逆多重化することは、 前記異なる同期ソースのそれぞれについて読み出しクロック信号(CLKo1、CLKo2)を生成することであって、前記読み出しクロック信号(CLKo1、CLKo2)は、クロックパルスの各列が前記TDM構造の各フレームにおいて含まれるビットの数と同等のクロックパルスの数を包含する、クロックパルスの列を含む、読み出しクロック信号を生成することと、 前記読み出しクロック信号(CLKo1、CLKo2)によって特定される入力レートで連続して前記出力データ信号(d−out1、d−out2)のビットを受信することと、 前記逆多重化データ信号(d−out1、d−out2)の前記受信したビットの所定の数を、一時的に、格納することと、その後、 前記除去逆多重化クロック信号(CLKdm1’、CLKdm2’)によって特定される出力レートで連続して前記ビットを送り出すことと、 を含む、請求項15〜18のいずれか1項に記載の方法。
- 20前記第1ノード(100)における前記サンプリングは、問題の前記抽出クロック信号(CLKex1、CLKex2)の前記周波数と比べて1以上のオーバーサンプリング係数を示すサンプリング周波数に基づいて前記抽出クロック信号(CLKex1、CLKex2)のそれぞれをサンプリングすることを含む、請求項15〜19のいずれか1項に記載の方法。
- 21各抽出クロック信号(CLKex1、CLKex2)に応えて前記第1ノード(100)において生成すること、前記抽出クロック信号(CLKex1、CLKex2)の前記周波数の所定の割合(1/F、1/G)を示す減少した周波数を有する結果として生じるダウンコンバート抽出クロック信号(CLKex1/F、CLKex2/G)を含む、請求項20に記載の方法。
- 22前記第2ノード(200)において、前記インターフェースモジュール(221、222)へ前記クロック信号を送る前に前記所定の割合(1/F、1/G)の反対(F、G)を示す係数を有する、各逆多重化サンプリングクロック信号(CLKex1/F、CLKex2/G)、またはその前記逆バージョンを多重化することを含む、請求項21に記載の方法。
- 23前記少なくとも2組の前記入力データ信号(d−in1、d−in2)において各データ信号のペイロード情報を示す各データ信号(d−ex1、d−ex2)を、前記第1ノード(100)において、抽出することと、 前記TDM構造における各独立信号として前記抽出データ信号(d−ex1、d−ex2)のそれぞれを、前記第1ノード(100)において、多重化することと、 を含む、請求項15〜22のいずれか1項に記載の方法。
- 24コンピュータのメモリに読み込み可能なコンピュータプログラムであって、前記プログラムが前記コンピュータ上で実行される場合に請求項14〜23のいずれかのステップを制御するためのソフトウェアを含む、コンピュータプログラム。
- 25コンピュータ可読媒体であって、記録されたプログラムを有し、前記プログラムは、前記プログラムが前記コンピュータに読み込まれる場合に請求項14〜23のいずれかのステップを前記コンピュータが制御するようにする、コンピュータ可読媒体。
Independent claims25
89 paragraphs, as filed
Background of invention and prior art The present invention generally relates to data transmission in which digital signals are multiplexed and demultiplexed. More specifically, the present invention relates to the data transmission system according to the preamble of claim 1 and the method according to the preamble of claim 14. The present invention also relates to the computer program of claim 24 and the computer readable medium of claim 25.
In a synchronous digital hierarchy (SDH) type communication system, the synchronous signal can be propagated by a so-called E1 signal (2.048 Mbit / s). Similarly, in a synchronous optical network (SONET) system, the synchronous signal can be propagated by a so-called T1 signal (1.544 Mbit / s). In both cases, it is the unique clock frequency of the signals E1 / T1 (ie 2.048 MHz and 1.544 MHz, respectively) that carry synchronization information through the system.
Communication systems typically use an atomic clock to generate a sync signal. These signals (eg, E1 type) are then propagated through the network, and signals far from the network may be input to a base station having a wireless interface. In such cases, for example, the synchronization signal sent by the base station for communication with the mobile phone controls the radio frequency. Therefore, even very small fluctuations in clock frequency may cause substantial performance problems. To avoid this type of problem, the wobble of the sync signal is described, for example, by ITU-TG. 823 and ITU-TG. It must be lower than a predetermined limit, as specified in 813.
Generally, each mobile operator distributes a separate clock signal in its network. The so-called backhaul operator may provide network resources to two or more mobile phone operators. This means that they may have to co-exist in given physical network signals originating from different clock sources. In addition, multiple mobile operators may sometimes share the site of a particular base station. Such a site is, for example, a place such as a tower where multiple base stations may be installed. The sharing situation may occur when one backhaul operator provides to a large number of mobile operators via one base station (or cell) site. Here, the technical problem of the backhaul operator is to transmit different synchronization signals together with the traffic signal to each mobile phone operator at the cell site as efficiently as possible.
It is assumed that the E1 signal is used. Each mobile operator then uses his own atomic clock to generate all the E1 signals in his system. A group of E1 signals from a given mobile operator may be referred to as a particular sync group, as all of these E1 signals come from the same source (most often an atomic clock).
For example, the 155.22 Mbps STM-1 signal may propagate up to the 63E1 signal, and all these E1 signals may belong to the same synchronization group. The atomic clock for this synchronization group controls the bit rate, i.e. the frequency corresponding to 155.52 MHz. The 155.52 MHz frequency will therefore be very accurate as it derives from the exact clock. The natural frequency of the E1 signal is 2.048 MHz, which is also believed to be very accurate as 2.048 MHZ is exactly 16 × 155.52 / 1215 = 2.048 MHz. It produces an exact clock frequency of 2.048 MHz, for example, if 155.52 MHz is multiplied by 16 (eg, phase-locked loop, PLL) and divided by 1215. Alternatively, the frequency 2.048 MHz can be generated directly from 155.52 MHz by so-called fractional division.
Nevertheless, it has proven very difficult to transmit more than one data or clock signal in multiplex transmission format with good synchronization quality / phase accuracy over a common medium. This is especially true if the signals have a nominal frequency, however, the signals show a slight frequency deviation when compared to each other. Specifically, the problem is, for example, ITU-TG. It is here that we encounter in demultiplexing processes where wobble is low enough to meet the given requirements of communication standards, such as 813. The term "wobble" is usually defined as low frequency jitter up to 10 Hz.
When transferring a single signal, so-called low factor oversampling may be employed to protect the phase information. Low factor Oversampling is understood to be a factor that is higher than 1 but generally lower than 2. U.S. Pat. Nos. 3,819,853, U.S. Pat. Nos. 4,920,545 and U.S. Pat. No. 6,009,109 show different solutions of this type. Unfortunately, neither of these approaches can be used to address the issues mentioned above.
In the prior art, the problem has instead been avoided by incorporating the required synchronization signal in the Ethernet (registered trademark) flow packet. Therefore, it is possible to reconstruct the original signal on the receiver side, thus emulating the so-called circuit connection through the use of packets. This strategy includes "Pseudo Wire" or the precision time synchronization protocol IEEE1588V. Sometimes called 2.
U.S. Pat. No. 4,873,684 describes a system for multiplexing, transmitting and demultiplexing signals with different frequencies. Here, reference sample signals obtained by multiplication of frequencies equal to or higher than the frequency of the highest frequency of the transmitted signals by the number of transmitted signals are used. Each signal transmitted is sampled based on a reference sample signal before being multiplexed into a time-divided format. Any empty time zone is filled with dummy signals. As a result, frequency requirements can be extreme and there is a risk of wasting substantial bandwidth resources.
U.S. Patent Application Publication No. 2008/0025346 describes solutions for asynchronous signal synchronization and multiplexing. Here, the so-called frame absorption phase is performed with respect to the inflow of asynchronous signals. As a result, the sync signal is generated because the pointer value is assigned to describe the asynchronous characteristic. The synchronization signal is then multiplexed through a process of changing the pointer value by the pointer transmitter.
U.S. Patent Application Publication No. 2002/0018493 discloses a data transmission system in which multiple data signals are incorporated into carrier signals using time division multiplexing (TDM) operations. Rate matching is performed between the data signal and the carrier signal by means of the location of the object. The packed data and reallocation management information is incorporated into the path layer overhead of the carrier signal superframe.
U.S. Pat. No. 6,888,826 discloses a solution that allows multiple clock signals to share processing resources. Here, the pointer is stored in a FIFO buffer, which in turn makes it possible to compensate for the time difference between the system clock and each leader clock signal so that the clock signal can be reconstructed on the receiving side. Put it in a state.
US Patent Application Publication No. 2005/0078683 describes a data communication system for transferring one or more payload stream data signals and backup data signals over a common medium. The spare data signals are organized as data packets, and the transmitted data for that format formats these packets into a stream data signal format. The signal is then multiplexed into a bitstream for transmission with the payload stream data signal.
<p num="0015"> Problems related to prior art Despite the various TDM-based approaches described above, if the signal is based on a different synchronization source, via a common medium with sufficiently high phase accuracy to meet the wobble requirements of today's most important communication standards. There is no conventional solution that allows the transmission of two or more data or clock signals in multiple forms.</p><p num="0016"> An object of the present invention is to solve the above problems and provide an effective and reliable means of multiplexing, transmitting and demultiplexing many signals controlled by at least two different clock sources. ..</p>
<p num="0017"> According to one aspect of the invention, the above object is achieved by the first described data transmission system, at least two sets of input data signals include at least one first signal and at least one second signal. The at least one first signal is based on a synchronization source that is different from the synchronization source on which the at least one second signal is based. Further, the first node includes at least one clock extraction module, at least one sampling module and a multiplexing module. The at least one clock extraction module is composed of at least one first and second signals so as to extract each clock signal indicating each of the different synchronization sources. At least one sampling module is configured to sample each of the extracted clock signals into the resulting sampled clock signal. Sampling is based on the sampling frequency, which is synchronized with the line frequency used to transmit the resulting bitstream on the transmission medium to the second node. The multiplexing module is configured to include each of the resulting sampled clock signals as separate signals in the TDM structure. The second node includes a demultiplexing module, at least one jitter attenuator and at least one interface module. The demultiplexing module is configured to demultiplex the received bitstream into at least two sets of output signals and one set of demultiplexed clock signals indicating the resulting sampled clock signal. At least one jitter damping means is configured on each signal in a series of demultiplexed clock signals to reduce the amount of frequency jitter to a lower predetermined level, and thus a series of demultiplexed clock signals. Generates a separate clock signal with a higher synchronization quality than that of the clock signal signal.</p><p num="0018"> This design also allows protection for very high timing accuracy when transmitting multiple data signals based on different sync sources with the same nominal frequency, but with slightly different drifts around this frequency. It is advantageous.</p><p num="0019"> According to a preferred embodiment of this aspect of the invention, the second node determines the amount of frequency jitter in each signal in the series of demultiplexed clock signals and in each signal in the series of demultiplexed clock signals. It includes at least one configured narrowband jitter damping means to reduce below the level, and thus each produces a stable clock signal.</p><p num="0020"> Preferably, at least one narrowband jitter attenuator includes a buffer module. This module is configured to continuously receive bits of the output data signal at the input rate specified by the demultiplexed clock signal associated with the output data signal in question. The buffer module temporarily stores a predetermined number of bits of the demultiplexed data signal, after which the module is configured to supply the bits to the line adapter module. The bits are contiguous at the final output rate specified by the phase-locked loop in which the stable clock signal is in turn controlled by the stable clock signal associated with the output data signal in question, based on the demultiplexed clock signal. Is supplied. The line adapter module is configured to continuously send output bits to the output terminal at the final output rate and thus represents the resulting clock carrying data signal from the second node. With this simple configuration, any frequency deviation in the resulting clock carrier data signal can be kept low enough to meet the wobble requirements of any existing communication standard.</p><p num="0021"> According to another preferred embodiment of this aspect of the invention, the second node is at least one narrowing for reducing the amount of frequency jitter below a predetermined level in each signal in a series of demultiplexed clock signals. Band jitter attenuation means are included, thus producing stable clock signals for each. Preferably, at least one of the at least one narrowband jitter attenuator is incorporated in at least one of the at least one interface module. This is advantageous as it provides effective design and circuit use.</p><p num="0022"> According to yet another embodiment of this aspect of the invention, the second node comprises at least one wide band jitter attenuation module. Each such module receives each demultiplexed clock signal and is configured to reduce frequency jitter beyond the frequency threshold in the demultiplexed clock signal, thereby resulting in a demultiplexed demultiplexed clock. Generate a signal. The wideband jitter attenuation module is then configured to send the resulting elimination demultiplexed clock signal to at least one narrowband jitter attenuation means. Therefore, jitter with a relatively wide bandwidth cannot be mixed down and then reaches at least one interface module.</p><p num="0023"> According to yet another preferred embodiment of this aspect of the invention, at least one of the wide band jitter attenuation modules is a phase-locked loop circuit configured to reduce frequency jitter beyond the frequency threshold. / Or includes a resonant circuit. Therefore, propagation of wide band jitter can be prevented by a simple method.</p><p num="0024"> According to a more preferred embodiment of this aspect of the invention, at least one of the at least one wide area jitter attenuator module comprises a clock regenerator. This unit is configured to receive each demultiplexed clock signal and repeatedly generate each average period length value indicating the average period time of the demultiplexed clock signal at average intervals including many clock periods. The clock regenerator is further configured to generate a demultiplexed elimination signal based on the average period length value. As a result, a very strong clock signal can be obtained. The design is even more advantageous as the clock regenerator can be implemented very effectively in terms of the required chip area, for example in an FPGA (Field Programmable Gate Array).</p><p num="0025"> According to another preferred embodiment of this aspect of the invention, the demultiplexing module is configured to generate a read clock signal for each of the different synchronization sources. The read clock signal includes a sequence of clock pulses, and each sequence of clock pulses contains a number of clock pulses equivalent to the number of bits contained in each frame of the TDM structure. The second node continuously receives the bits of the output data signal at the input rate specified by the read clock signal, temporarily stores a predetermined number of received bits of the demultiplexed data signal, and then removes it. It further comprises at least one buffer means configured to continuously deliver the bits at the output rate specified by the demultiplexed clock signal. Therefore, the interface module is provided on a high quality basis for producing the resulting clock carrying data signal.</p><p num="0026"> According to yet another preferred embodiment of this aspect of the invention, at least one sampling module of the first node is an extraction clock signal based on a sampling frequency that exhibits an oversampling coefficient with respect to the frequency of the extraction clock signal in question. It is configured to sample each of the above. The overlapping coefficient is at least 1 or more, preferably 1.25 or more. More preferably, the overlapping coefficient is around 1.5 to 1.75. As a result, jitter in the data is handled as in the case of clock signals, and at the same time the bandwidth of the transmission medium is efficiently utilized.</p><p num="0027"> According to a further preferred embodiment of this aspect of the invention, each node responds to the generation of a down-converted extraction clock signal resulting from having a reduced frequency relative to the frequency of the extraction clock signal. Includes at least one split module configured to receive the extraction clock signal. Here, the reduced frequency indicates a predetermined ratio of the frequency of the extraction clock signal. The predetermined ratio may be 1/2 such that the resulting down-converted extraction clock signal is a so-called half clock. However, technically, any other proportion is conceivable according to the present invention. In either case, the split module further relaxes the bandwidth requirement on the transmission medium.</p><p num="0028"> According to another preferred embodiment of this aspect of the invention, it is provided that the first node comprises at least one split module and the second node comprises at least one multiplication module. The module is configured to multiply each demultiplexed clock signal, or its elimination version, by a factor that indicates the inverse of the predetermined proportions described above before sending the clock signal to the interface module. Therefore, the extraction clock signal is reformed.</p><p num="0029"> According to yet another preferred embodiment of this aspect of the invention, the first node comprises at least one data extraction module configured to extract each data signal indicating the respective payload information of the input data signal. .. The multiplexing module is further configured to include each of the extracted data signals as independent signals in the TDM structure. Therefore, the TDM structure includes both the extracted data signal and the sampled clock signal.</p><p num="0030"> According to another aspect of the invention, the above object is achieved by the method originally described, at least two sets of input data signals include at least one first signal and at least one second signal. Here, at least one first signal is based on a synchronization source that is different from the synchronization source on which at least one second signal is based. The method results in extracting at the first node each clock signal indicating each of the different synchronization sources from the at least one first and second signal, and synchronizing at the line frequency, respectively, based on the sampling frequency. The step of sampling each of the extracted clock signals into the sampled clock signal generated as an independent signal in the first node and the resulting clock signal sampled as independent signals in the TDM structure are multiplexed in the first node. A step of demultiplexing and a step of demultiplexing a bit stream received into a series of demultiplexed clock signals indicating at least two sets of output data signals and the resulting sampled clock signal at the second node. In a series of demultiplexed clock signals to reduce the amount of frequency jitter below a predetermined level of clock signal with a higher sync quality than the signal sync quality in the series of demultiplexed clock signals. In each signal, at the second node, each data signal in at least two sets of output data signals having a step of jitter attenuation and the resulting clock carrying data signal and its associated stabilized clock signal is added to the second node. Further includes, in, the step of recombination. The advantages of this method, as well as its preferred embodiment, are evident from the above discussion with reference to the proposed data transmission system.</p><p num="0031"> According to a further aspect of the invention, the object is loadable directly into the memory of a computer and includes software applied to perform the proposed method when the program is run on the computer. Achieved by computer programs.</p><p num="0032"> According to another aspect of the invention, an object is achieved by a computer-readable recording medium having a recorded program for controlling a computer that performs the proposed method when the program is loaded into the computer. To.</p><p num="0033"> Further advantages, beneficial features and applications of the present invention will be apparent from the following description and dependent claims.</p><p num="0034"> The present invention will be described more clearly by reference to the means of the preferred embodiment and the accompanying drawings disclosed by way of example.</p>
<figref num="1">The schematic block diagram of the node composition to which this invention is applied is shown.</figref><figref num="2a">The block diagram of the data transmission system which concerns on 1st Embodiment of this invention is shown.</figref><figref num="2b">The block diagram of the data transmission system which concerns on 2nd Embodiment of this invention is shown.</figref><figref num="2c">The node configuration of FIG. 1 which further refers to the specific features of the present invention is shown.</figref><figref num="3">The block diagram of the interface module which concerns on one Embodiment of this invention is shown.</figref><figref num="4">A flow diagram shows the general method performed by the proposed sending node.</figref><figref num="5">A flow diagram shows the general method performed on the proposed receiving node.</figref>
FIG. 1 is first referred to, showing a block diagram showing a node configuration to which the present invention is applicable.
For clarity, FIG. 1 shows only the data traffic and synchronization propagating from the first node 100 to the two second nodes 201 and 202 via the first transmission line L1 and the second transmission line L2, respectively. Shown. However, in a real system, data traffic is also generally considered to flow in the opposite direction. Normally, however, it is sufficient to send an accurate synchronization signal in one direction, and here the flow from the first node 100 to the second nodes 201 and 202 is illustrated. This is especially true when the first node 100 is a so-called hub node and the second nodes 201 and 202 are access nodes for mobile communication systems, where the access nodes may indicate equipment at different cell sites. Although FIG. 1 shows only two access nodes, it should be further understood that in a real system, a larger number (such as 8) of access nodes may be connected to one hub node. ..
In FIG. 1, the set of input data signals includes STM-1 signals STM-1A, STM-1B, STM-1C and STM-1D and Gigabit Ethernet® signals GBE-A, GBE-B, GBE-C and Indicated by GBE-D. However, the present invention is not limited to these signal formats. On the contrary, many additional different types of signals may be transmitted from the first node 100 to one or more second nodes 201 and 202. The first and second transmission lines L1 and L2 may each include an optical fiber cable that interconnects various cell sites. In such cases, each transmission line L1 and L2 generally utilizes a single wavelength in this application. Therefore, one wavelength may carry many different sync signals, which is not possible according to previously known solutions.
In most cases, each mobile operator provides its own atomic clock signal as a synchronization source for the base station of each of their communication systems.
In FIG. 1, the first mobile operator uses the STM-1A signal to provide synchronization in its own system, and the second mobile operator uses the STM-1B to provide synchronization in its own system. Using the signal, the third mobile operator uses the STM-1C signal to provide synchronization in its system, and the fourth mobile operator uses the STM-1D to provide synchronization in its system. Suppose you want to use a signal.
The STM-1A signal may sequentially include up to 63 E1 signals called E1-A1 through E1-A63. As a result, in FIG. 1, the E1 signals E1-A1, E1-A2, E1-A3 and E1-A4 arise from the STM-1 signal STM-1A, and thus all of these signals have a frequency of 2.048 MHz. Based on the first sync source.
Similarly, the E1 signal E1-B1 originates from the second mobile operator's STM-1 signal STM1B, which is based on a second sync source having a frequency of 2.048 MHz, and the E1 signal E1-C1 has a frequency of 2. The origin is the STM-1 signal STM-1C of the third mobile operator, which is based on the third sync source having 048 MHz, and the E1 signal E1-D1 is based on the fourth sync source having a frequency of 2.048 MHz. 4 The STM-1 signal STM-1D of the mobile phone operator is the deadline.
In the example of FIG. 1, the E1 signal generated from the STM-1 signal is terminated at the second nodes 201 and 202. Nevertheless, assisted by a network management system, the unique E1 signal can be distributed from hub node to access node in any combination.
FIG. 2a shows a block diagram of a data transmission system according to the first embodiment of the present invention.
The system includes a transmission medium L that interconnects the first node 100, the second node 200, and the first and second nodes 100 and 200.
The first node 100 receives a set of input data signals d-in1 and d-in2 and is configured to format a signal for transmission by a TDM structure, or at least a subset of each set of signals. The pair of input data signals d-in1 and d-in2 includes at least two data signals, each of which incorporates a clock signal. Specifically, the pair of input data signals d-in1 and d-in2 includes at least one first signal and at least one second signal, and at least one first signal has at least one second signal. Based on a different synchronization source than the based synchronization source.
The transmission medium L is configured to transmit a TDM format signal as a bit stream bs. The bitstream bs, in turn, have a line frequency that meets the bandwidth requirements of the TDM format signal.
The bitstream bs may have 2.5 Gbps. This is sufficient to propagate from the first node 100 to the signal second node 200, and each sync source where the input signal is based on two different sync sources and sampled into the resulting sampled signal of 3.125 Mbps. , 16E1 signals, each having a maximum of 429.732 Mbps of 1015.625 Mbps (64B / 65B encryption) plus overhead data bit rates. Here, 64B / 65B encryption means a format, 64 bits are encoded in 65 bits, and the quantity 1015.625 is therefore derived as (64/65) x 1000.
In addition, the bitstreams that represent each of the synchronous sources are each nearly jitter-free 3.125 Mbps stream, or 2.048 Mbps stream (of the 101010 type) with a relatively large peak for peak jitter corresponding to the period of the oversampling frequency. Is considered to be either. This means that the maximum amplitude jitter will be 320 ns. Since the 3.125 Mbps bitstream bs has a clock frequency in a complete period with a line frequency, transmitting these bits in a TDM frame having the same number of bits per frame in each frame is a trivial task. ..
The line frequency, however, is not synchronized to the 2.048 MHz frequency. Therefore, when a TDM structure containing an E1 signal is used, the number of bits per E1 and per frame must vary due to the frequency shift between the line frequency and the frequency 2.048 MHz of the E1 signal. This, however, can be handled by the multiplexing module 140 at first node 100 by including information in each frame regarding the number of bits contained in the frame.
The resulting bitstream is then transmitted, for example, on the transmission medium L of the optical fiber cable. The bit rate may be 2.5 Gbps (ie 2500 Mbps), which is appropriate as the frequency 3.125 MHz may be generated as 2500/800 = 3.125.
The second node 200 is configured to receive the bitstream bs and demultiplex the bitstream bs into two sets of output data signals d-out1 and d-out2.
The first node 100 includes at least one sampling module 131 and 132 and a multiplexing module 140, respectively, of at least one clock extraction module 111 and 112. Preferably, the first node also includes at least one data extraction module 121 and 122.
At least one data extraction module 121 and 122 is configured to extract each set of data signals d-ex1 and d-ex2 indicating the payload information of each set of input data signals d-in1 and d-in2. ..
The multiplexing module 140 is configured to include each signal in the extracted set of data signals d-ex1 and d-ex2 as each independent signal in the TDM structure.
Each clock extraction module 111 and 112 is configured to extract each clock signal CLKex1 and CLKex2 indicating each of the different synchronization sources from at least one first and second signal.
At least one sampling module 131 and 132 is configured to sample the extraction clock signals CLKex1 and CLKex2, respectively, to the resulting sampling clock signals CLKsp1 and CLKsp2. Sampling is based on the sampling frequency sm, which is synchronized with the line frequency. Preferably, at least one sampling module 131 and 132 samples the extraction clock signals CLKex1 and CLKex2, respectively, based on a sampling frequency that exhibits an oversampling coefficient of 1 or more relative to the frequencies of the extraction clock signals CLKex1 and CLKex2 in question. It is configured as follows. More preferably, the oversampling coefficient is 1.25 or more, and most preferably about 1.5 to 1.75. Nevertheless, if the division Y / X indicating the oversampling factor does not have a low denominator ratio, for example, 3/2 or 4/3, is the edge of the irregularly distributed extraction clock signals CLKex1 and CLKex2. Therefore, it is advantageous to cause an out-of-phase display. As a result, oversampling coefficients such as 1.52 and 1.77 are better than 1.5 (ie 3/2) and 1.75 (ie 7/4), respectively. By selecting an appropriate oversampling coefficient, the degree of jitter in the data signals d-ex1 and d-ex2 and the extraction clock signals CLKex1 and CLKex2 can be processed. At the same time, the bandwidth of the transmission medium L is saved.
In this context, frequency f<sub>CLK</sub>The Hz clock has a bit rate of 2f<sub>CLK</sub>It is worth mentioning that it corresponds to a data signal with an endless 01010 pattern in bps. Therefore, when sampling a clock signal, the oversampling coefficient OF is the sampling frequency f.<sub>smp</sub>Is determined as a ratio between and the rate of the data signal corresponding to the clock signal to be sampled, i.e. OF = f<sub>smp</sub>/ (2f<sub>CLK</sub>).
The multiplexing module 140 is further configured to include each of the resulting sampling clock signals CLKsp1 and CLKsp2 as each independent signal in the TDM structure.
The second node 200 includes a demultiplexing module 210, at least one jitter attenuator (eg, represented by narrowband jitter attenuator 231 and 232 in FIG. 2a), and at least one interface module 221 and 222.
The demultiplexing module 210 reverses the bit stream bs received by the pair of output data signals d-out1 and d-out2, respectively, of the pair of demultiplexed clock signals CLKdm1 and CLKdm2 indicating the resulting sampling clock signals CLKsp1 and CLKsp2. It is configured to be multiplexed. Here, each set of the output data signals d-out1 and d-out2 usually includes a plurality of (8, etc.) data signals, and the set of the demultiplexed clock signals CLKdm1 and CLKdm2 is a series of input data signals d. The number of clock signals equivalent to the number of synchronization sources included in in1 and d-in2 is shown.
The at least one narrowband jitter attenuating means 231 and 232 are configured to reduce the amount of frequency jitter below a predetermined level in each signal in the series of demultiplexed clock signals CLKdm1 and CLKdm2. Therefore, as a result, the stabilized clock signals CLKstb1 and CLKstb2 are generated. Preferably, at least one of the at least one narrowband jitter attenuator 231 and 232 is associated with the respective data feed modules 241 and 242 and is integrated with at least one of the at least one interface modules 221 and 222. The module. That is, it provides an effective overall design and circuit usage. Further details regarding the interface module are presented below with reference to FIG.
At least one interface module 221 and 222 has a series of output data having each of the stabilized clock signals CLKstb1 and CLKstb2 associated with the resulting clock carrier data signal in each of the series of data signals d-res1 and d-res2. The signals d-out1 and d-out2 are configured to recombine each data signal.
FIG. 2b shows a block diagram of a data transmission system according to a second embodiment of the present invention. In FIG. 2b, all units, signals and values carrying the same reference code as those that also occur in FIG. 2a specify values that refer to the same unit, signal and FIG. 2a, respectively, and are therefore repeated below. Absent.
According to one preferred embodiment of the invention, the second node 200 includes at least one wideband jitter attenuation module. In the embodiment shown in FIG. 2b, each of the two wideband jitter attenuation modules 271 and 272 is included in the second node 200.
Each module 271 and 272 receives each set of demultiplexed clock signals CLKdm1 and CLKdm2 and reduces frequency jitter beyond the threshold frequency in this set of demultiplexed clock signals, thus resulting in elimination reverse. It is configured to generate each of the multiplexed clock signals CLKdm1'and CLKdm2'. Each of the wideband jitter attenuation modules 271 and 272 then results to at least one narrowband jitter attenuation means 231 and 232 (see FIG. 2a), which are assumed to be incorporated into the interface modules 221 and 222 respectively in FIG. 2b. Elimination Demultiplexed clock signals CLKdm1'and CLKdm2' are transferred.
An important objective of the wideband jitter attenuation modules 271 and 272 is to reduce the effect of any jitter where the critical spectrum is downmixed. A relatively well-known problem is the fact that, in addition to reducing the frequency of the signal that the frequency divider results from, such as the frequency divider modules 51 and 52 at the first node 100, it also mixes jitter or phase modulation. Regarding. For example, it is well known that when the clock frequency of 2.048 MHz is connected so as to be divided by a 64 divider, the output frequency becomes 32 kHz (that is, 2.048 MHz / 64). What is not well known is that the 2.048 MHz signal of the same example is phase modulated with 3200 Hz, and the output from the divider has only 1 Hz jitter (or phase modulation). This signal, in turn, may be within the PLL bandwidth where it is difficult to remove the signal component in question. Similarly, 31999Hz jitter also causes 1Hz at the output of the divider.
Applicants perform simulations and measurements showing that in the above example, high frequency jitter is downmixed to low frequency baseline jitter when the high frequency jitter has frequencies 32 kHz, 54 kHz, 96 kHz, 128 kHz, etc. I came. Simulations and measurements have further shown that such down-mixing causes a large amount of unwanted wobbling that is unacceptable when transferring synchronization signals such as sampling clock signals CLKsp1 and CLKsp2.
For example, if filtering is not performed in advance, the wobbling in the signal caused by downmixing may result in wobbling greater than 5 ns at intervals of -15 ppm to 15 ppm. To reduce this effect, the applied prefilter must have substantial frequency jitter attenuation according to the following equation. M × f / N Here, M = 1, 2, 3, 4,. .. .. f = source signal frequency
Therefore, there should be substantial jitter attenuation at 32 kHz, 64 kHz, 96 kHz ... For N = 64 and f = 2.048 MHz. This type of pre-filtering prevents high frequency jitter from being downmixed to low frequency jitter or wobbling.
Broadband jitter attenuation modules 271 and 272 therefore have this filtering function. Modules 271 and 272 may be implemented through different techniques. According to embodiments of the present invention, these modules 271 and 272 (or clock prefilters) may include a PLL, resonant circuit or so-called clock regenerator.
In particular, at least one of the at least one wideband jitter attenuation modules 271 and 272 may include a PLL circuit configured to reduce frequency jitter above the threshold frequency.
Alternatively, or as an aid thereof, at least one of the at least one wideband jitter attenuation modules 271 and 272 may include a resonant circuit configured to reduce frequency jitter above the threshold frequency.
Yet another alternative, or as an aid, at least one of the at least one wideband jitter attenuation modules 271 and 272 may include a new type of clock regenerator. This unit receives each demultiplexed clock signal CLKdm1 and CLKdm2 and repeats each average period length value indicating the average period for the demultiplexed clock signals CLKdm1 and CLKdm2 at averaging intervals including many clock periods. , And are configured to generate the elimination demultiplexed clock signals CLKdm1'and CLKdm2' based on the average period length value. The result of this averaging at frequency jitter above the threshold is reduced at each of the signals CLKdm1'and CLKdm2'.
To clarify the operating principle of the proposed new type of clock regenerator, refer to Table 1 below. Table 1 shows how much average period length PL<sub>avg</sub>Is the input signal CLK<sub>in</sub>Determined by the rolling averaging interval of 4 cycles of, and based on how much the stabilized output clock signal CLK<sub>out</sub>Indicates whether is generated. It should be noted that the 4-cycle averaging interval is only selected to provide an example. In practice, much larger intervals are preferred, for example, the input clock signal CLK.<sub>in</sub>Used to indicate 128 cycles of.
The second column of Table 1 shows the input clock CLK.<sub>in</sub>Indicates the value of the period of. As you can see, the input clock signal CLK<sub>in</sub>Has a period length of decimal value 8 between the first 4 columns (ie, corresponds to 8 cycles of the sampling clock). After that, between columns 5 and 11, the cycle length is increased to the decimal value 14, and finally, as in column 12, the cycle length is lowered to the decimal value 4. Instead of this example, the input clock signal CLK<sub>in</sub>Since there are approximately 10 sampling clock cycles for each cycle of, in practice the sampling clock frequency is the input clock signal CLK.<sub>in</sub>It should be further noted that there are about 100 sampling clock cycles for each cycle of.
The third column of Table 1 shows the rolling total of the four-cycle length PL. Note that we assume that there is a one-column delay in the calculation. This becomes apparent, for example, in the fifth column when the period length PL changes from 8 to 14 and the rolling total changes first in the sixth column.
The fourth column of Table 1 shows the so-called modified rolling totals. The first modification here occurs in row 7, where the error term is added from the previous column (ie, column 6) in column 7. The error term in column 7 is the calculated average period length PL.<sub>cavg</sub>And average cycle length PL<sub>avg</sub>Compensate for any differences between and, with reference to column 7 and further discussed below.
The fifth column of Table 1 is the number of period length PLs for which averaging is performed to determine the average period length PLavg, the modified rolling obtained by dividing the value in column 4 by 4. Shows the average value of the total.
The sixth column of Table 1 shows the values in the fifth column, rounded down to an integer. Output clock signal CLK<sub>out</sub>Note the numerical values in the sixth column corresponding to the cycle length values of, and these cycle length values are, in order, the input clock signal CLK.<sub>in</sub>Can be regarded as a low-pass filter version of the period length PL of.
Column 7 of Table 1 shows the error term obtained by subtracting the value in column 5 from the value in column 6 and then multiplexing this value with 4 in each row (ie, the average interval). Show).
Each time is the input clock signal CLK<sub>in</sub>It means that each start cycle is obtained by accumulating the cycle length value PL in the second column. The eighth column of Table 1 shows the sampling clock CLK received from time 0.<sub>smp1</sub>The points of these times are shown, which also correspond to the number of clock pulses in.
The ninth column shows, in particular, the points in the pulse time of the output clock signal CLKout. The value in column 8 is simply obtained by accumulating the numbers in column 2 up to a given row.
In the fifth row, the number of inputs in the eighth column is 46 (ie, a higher number), while the number of outputs in the ninth column is 40. This is an example of so-called non-causal behavior that cannot occur in an actual system. The problem behind this phenomenon is the output clock signal CLK at the point at time t = 40.<sub>out</sub>The information required to generate the pulse of is not available by the point at time t = 46.
This non-causal behavior is that the input cycle length PL increases from 8 to 14, but the input clock signal CLK.<sub>in</sub>Is revealed by the decrease in the frequency of. A large enough offset number to eliminate non-causal behavior is the output clock signal CLK.<sub>out</sub>Is added to the calculated point time when the pulse of is generated. Here, the 15 offsets shown in column 10 of Table 1 have been selected. As you can see, some input points in time (8th column) are the same as some output points in time (10th column). This means that the system is on the verge of non-causal behavior. Therefore, here, 15 is certainly the minimum possible offset value.
Preferably, the number of offsets (relatively large) is the output clock signal CLK.<sub>out</sub>Is added to the initial calculated points at the time of. This allows any non-causal behavior caused by lower frequency, jitter and phase modulation to be handled by the clock regenerator.
In the first row of columns 10 and 9, the output clock signal CLK<sub>out</sub>It is clear that the pulse is a 15 sampling clock pulse generated after a point in time if the clock regenerator "recognizes" that the clock pulse in question should be created.
This is the input clock signal CLK<sub>in</sub>As long as has a period length of 8, it means that there is a margin for 15 (ie, up to 4th line) acausal behavior. However, when the frequency is reduced, the margin has been reduced to 0 so that the cycle length is 14.<tables num="1"><img id="000002" he="139" wi="159" file="0005592001.tif" img-format="tif" img-content="drawing" /></tables>
According to a preferred embodiment of the invention, the demultiplexing module 210 is configured to generate read clock signals CLKo1 and CLKo2, respectively, for each of the different synchronization sources. Each read clock signal CLKo1 and CLKo2 includes a sequence of clock pulses, and each sequence of clock pulses includes a number of clock pulses equivalent to the number of bits contained in each frame of the TDM structure.
In this embodiment, the second node 200 further includes at least one buffer means 251 and 252. Each of the buffer means 251 and 252 is configured to continuously receive the bits of the output data signal set d-out1 and d-out2 at the input rate specified by the applicable read clock signals CLKo1 and CLKo2. .. Each of the buffer means 251 and 252 temporarily stores a predetermined number of received bits of the demultiplexed data signal sets d-out1 and d-out2, and then removes them from the wideband jitter attenuation modules 271 and 272, respectively. These bits are continuously sent out at the output rate specified by the demultiplexed clock signals CLKdm1'and CLKdm2'.
According to another preferred embodiment of the invention, the first node 100 includes at least one divider module. FIG. 2b shows two such modules, 151, configured to receive the respective extraction clock signals CLKex1 and CLKex2 and to generate the resulting down-converted extraction clock signals CLKex1 / F and CLKex2 / G in response. And 152 are shown respectively. The down-converted extraction clock signals CLKex1 / F and CLKex2 / G have reduced frequencies indicating predetermined proportions 1 / F and 1 / G of the frequencies of the extraction clock signals CLKex1 and CLKex2, respectively.
The predetermined ratio may be 1/2. In such a case, the first node 100 generates, for example, a so-called half clock in which 1.024 MHz is transmitted via the system instead of the 2.048 MHz signal. Of course, the bandwidth is therefore conserved. Note that the half clock also corresponds to 2.048 Mbps with the data pattern 101010 and the like. It should also be noted that according to the present invention, other predetermined proportions 1 / F and 1 / G may be used in the same manner as, for example, 1/3 or 1/4.
In any case, if the first node 100 includes at least one divider module 151 and / or 152, the second node 200 will send the clock signal to the interface modules 221 and 222, respectively, at a predetermined ratio of 1. At least one matching multiplier module 261 configured to multiply each demultiplexed sampling clock signal CLK / F and CLK / G with coefficients indicating the opposite F and G of / F and 1 / G respectively. And / 262 must be included.
Nevertheless, the particular order between the multiplier modules 261 and 262 and the wideband jitter attenuation modules 271 and 272 in the processing chain is irrelevant. This means that the multiplier modules 261 and 262 may operate similarly in place of the elimination demultiplexed clock signals CLKdm17F and CLKdm27G (not shown in FIG. 2b), respectively.
FIG. 2c shows the node configuration of FIG. 1 which further references the particular features of the present invention. However, in contrast to FIG. 1, for clarity of explanation, the signal indicating the GBE signal is not shown in FIG. 2c. Nevertheless, in FIG. 2c, all units, signals and values having the same reference code as used in FIGS. 2a and / or 2b are similar to those described above with reference to FIGS. 2a and / or 2b. References are made to units, signals and values respectively and are not repeated below.
Similar to the example shown in FIG. 1, in FIG. 2c, the first node 100 is connected to the two second nodes 201 and 202 via the first and second transmission lines L1 and L2. It is further assumed that the pair of input data signals d-in1 and d-in2 contains signals based on two different synchronization sources and signals from each of these groups are transmitted to the second nodes 201 and 202 respectively. .. Here, in the resulting clock carrier signal, d-res1 and d-res3 are designated as d-res2 and d-res4, respectively. As a result, the sampling clock signals CLKsp1 and CLKsp2 indicating both synchronization sources are the first bitstream transmitted via the first transmission line L1 as well as the second bitstream bs2 transmitted via the first transmission line L2. Must be included in bs1. Further, the second nodes 201 and 202 are respectively used to recombine each output data signal in the set of data signals d-out1, d-out2, d-out3 and d-out4 regarding the demultiplexed clock signals CLKdm1 and CLKdm2. Interface modules 221, 222, 223 and 224 must be included.
With reference to FIG. 3, the block diagram of the interface module 220 according to the embodiment of the present invention is referred to (see modules 221 and 222 of FIGS. 2a and 2b, and modules 221 to 224 of FIG. 2c). In this embodiment, the narrowband jitter attenuating means 230 is incorporated into the interface module 220 together with the data sending module 240. The narrowband jitter attenuating means 230 includes a PLL circuit 331, 332, 340, 360, and the data sending module 240 includes a buffer module 310 and a line adapter module 320.
The buffer module 310 is configured to continuously receive the bits of the output data set d-out at the input rate specified by the demultiplexed clock signal CLKdm associated with the set d-out of the output data signal in question. To. The buffer module 310 is configured to temporarily store a predetermined number of bits of the demultiplexed data signal and then send the bits bts to the line adapter module 320. Bits bts are continuously sent out at the final output rate specified by the stabilizing clock signal CLKstb, which is related to the output data signal Seto d-out in question.
The stabilized clock signal CLKstb is sequentially generated by the narrowband jitter attenuating means 230. Specifically, the first frequency divider module 331 receives the demultiplexed clock signal CLKdm, and the second frequency divider module 332 receives the stabilized clock signal CLKstb. Both the first and second fractional period modules 331 and 332 divide their respective input signals CLKdm and CLKstb by an appropriate factor N to produce the resulting signal at the reduced frequency. The phase comparator 340 receives the reduced frequency signal and in response reduces the input to the loop filter 350. The loop filter-350, in turn, controls a voltage controlled oscillator 360 that produces the stabilized clock signal CLKstb. Therefore, the data bits stored in the buffer module 310 are sent out at the final output rate specified by the stabilized clock signal CLKstb, which in turn is based on the demultiplexed clock signal associated with the set d-out of the output data signal.
The line adapter module 320 is configured to continuously transfer the data bits bts output from the buffer module 310 to the output terminal at the final output rate. Therefore, the signal at the output terminal indicates the resulting clock carrier data signal d-res from the second node.
In summary, a general method for multiplexing, transmitting and demultiplexing a data signal according to the present invention will be described with reference to the flow diagrams in FIGS. 4 and 5. Here, FIG. 4 shows a procedure on the transmitter side, and FIG. 5 shows a procedure on the receiver side.
In FIG. 4, the first step 410 receives at least two sets of input data signals. The set of input data signals includes at least two data signals, each of which incorporates a clock signal. In particular, the set of input data signals includes at least one first signal and at least one second signal, and the at least one first signal is based on a different synchronization source than the synchronization source on which the at least one second signal is based. .. Step 420 then extracts each clock signal indicating each of the different synchronization sources from the input data signal. Subsequently, step 430 samples each of the extracted clock signals into the resulting sampling clock signals based on the sampling frequency. The sampling frequency, in turn, synchronizes with the line frequency used to transmit the signal in the form of a string of bits (see step 450 below). Step 440 then formats the sampling clock signal extracted from the set of input data signals for transmission by the TDM structure along with the extractor signal. Then, step 450 transmits the TDM format signal as a bit stream on the transmission medium. Finally, the procedure returns to step 410 again. Of course, in practice, the data is received continuously, so all steps 410-450 are performed simultaneously, however, with respect to different parts of the data.
In FIG. 5, the first step 510 receives the bitstream via the transmission medium. The bitstream has a line frequency and represents a TDM format signal containing an extracted data signal as well as a sampling clock signal indicating at least one first signal and at least one second signal. It is based on a different synchronization source than the synchronization source on which at least one second signal is based. Step 520 then demultiplexes the bitstream into at least two sets of output data signals, each of a series of demultiplexed clock signals indicating the resulting sampling clock signal. Subsequently, step 530 performs jitter attenuation for each signal in the series of demultiplexed clock signals in order to reduce the amount of frequency jitter below a predetermined level. Therefore, each clock signal is generated and has a higher synchronization quality than the simultaneous quality of the signals in a series of demultiplexed clock signals. Step 540 then recombines each data signal in a series of output data signals having its associated clock signal to each resulting clock carrier data signal, and finally the procedure returns to 510 again. Similar to the above, in practice, the data bits are received consecutively, so all steps 510-540 are performed simultaneously, but with respect to different parts of the data.
All steps, as well as steps in any subsequence, are described with reference to FIGS. 4 and 5, the above may be controlled by a programmed computer device. Further, the embodiments of the present invention described above with reference to the drawings include computer devices and processes performed in the computer devices, but the present invention is therefore also applied to carry out the present invention. It extends to computers, specifically to carrier computer programs. The program may be in the form of source code, object code, code intermediate source and object code, such as a partially compiled format or other format suitable for use in performing the procedures according to the invention. The program may be either part of the operating system or an independent application. The carrier may be any component or device capable of executing the program. For example, the carrier may include a flash memory, ROM, such as a recording medium such as a DVD, CD, EPROM, EEPROM, or a time magnetic recording medium such as a floppy (registered trademark) disk or hard disk. Further, the carrier may be a transmittable carrier such as an electrical or optical signal that can be propagated via an electrical or optical cable or by radio or other means. If the program is embodied in a signal that can be propagated directly by a cable or other device or means, the carrier may be configured by such cable or device or means. Alternatively, the carrier may be an integrated circuit into which the program is incorporated, and the integrated circuit is adapted for or for use in performing the relevant procedure.
As used herein, the term "comprises / comprising" is construed to identify the presence of defined features, integers, steps or components. However, the term does not preclude the existence or addition of one or more additional features, integers, steps or components or groups thereof.
Any prior art reference herein is not construed as a confirmation or any proposal that the referenced prior art forms part of common sense in Australia or other countries.
The present invention is not limited to the embodiments described in the drawings, and can be freely modified within the scope of the claims.
8 sheets
Sheet 1 Sheet 2 Sheet 3 Sheet 4 Sheet 5 Sheet 6 Sheet 7 Sheet 8
9 members in 5 offices
Priority claims4
| Document | Office | Kind | Date |
|---|---|---|---|
| 2010055613 | European Patent Office (EPO) | W | |
| 2010055613 | European Patent Office (EPO) | W | |
| 2010055613 | – | – | – |
| WO2010EP55613 | – | – | – |
Members9
| Document | Office | Kind | |
|---|---|---|---|
| WO2011134498A1 | World Intellectual Property Organization (WIPO) | A1 | |
| CN102859912A | China | A | |
| US2013039369A1 | United States of America | A1 | |
| EP2564530A1 | European Patent Office (EPO) | A1 | |
| JP2013526196A | Japan | A | |
| EP2564530B1 | European Patent Office (EPO) | B1 | |
| JP5592001B2This record | Japan | B2 | |
| US8923347B2 | United States of America | B2 | |
| CN102859912B | China | B |
17 legal events, as the office reported them to INPADOC
Over the term
Point at a mark for the eventEvents
| Event | Code | |
|---|---|---|
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Certificate of patent or registration of utility modelJAPANESE INTERMEDIATE CODE: R150R150 | R150 | |
| First payment of annual fees (during grant procedure)JAPANESE INTERMEDIATE CODE: A61A61 | A61 | |
| Written decision to grant a patent or to grant a registration (utility model)JAPANESE INTERMEDIATE CODE: A01A01 | A01 | |
| Decision of grant or rejection writtenTRDD | TRDD | |
| Request for written amendment filedJAPANESE INTERMEDIATE CODE: A523A521 | A521 | |
| Written permission of extension of timeJAPANESE INTERMEDIATE CODE: A602A602 | A602 | |
| Written request for extension of timeJAPANESE INTERMEDIATE CODE: A601A601 | A601 | |
| Notification of reasons for refusalJAPANESE INTERMEDIATE CODE: A131A131 | A131 |
Numbers
- Publication, DOCDB
- 5592001
- Publication, EPODOC
- JP5592001B
- Application
- 2013506496
- Application, DOCDB
- 2013506496
- Application, EPODOC
- JP20130506496
Titles
- English
- Data transmission including multiplexing and reverse multiplexing of an inclusion clock signal
Classification
- CPC, 5
- H04J3/1664
- H04L5/26
- H04L7/0083
- H04L5/14
- H04L1/0025
- IPC, 2
- H04J3 00
- H04L7 00
