JP3773368B2 - Node connection method - Google Patents

Description

表１において、コネクト端子ＣＯＮ１とＣＯＮ２がともにＯＮとなるのは、当該ノードの左右両側のコネクタがともに接続されている状態を意味する。このような接続状態となるのは、ネットワーク２０の中間位置に存在するノード２２である。
【００７３】
また、ＣＯＮ１がＯＮで、ＣＯＮ２がＯＦＦとなるのは、当該ノードの左側コネクタが接続、右側コネクタが非接続の場合で、ネットワーク２０の右端位置に存在するノード２３の接続状態に相当する。
【００７４】
さらに、ＣＯＮ１がＯＦＦで、ＣＯＮ２がＯＮとなるのは、当該ノードの左側コネクタが非接続、右側コネクタが接続の場合で、ネットワーク２０の左端位置に存在するノード２１の接続状態に相当する。
【００７５】
なお、ＣＯＮ１とＣＯＮ２がともにＯＦＦとなるのは、当該ノードがネットワークに接続されていないことを意味する。
【００７６】
一般的には、本実施形態のノードによるこのようなネットワーク上での自身の位置の認識は、当該ノード内のスイッチ部における接続の選択と連動している。
【００７７】
たとえば、図６（Ｃ）および（Ｄ）、ならびに図４に示すように、ケーブル３０Ａ、３０Ｂと同様なケーブル３０Ｃを用いて右端位置のノード２３のさらに右側に新たなノード３３を接続することでネットワーク２０に対するノードの追加を行う場合、ノード２３は中間位置のノードとなり、当該新たなノード３３が右端位置のノードとなる。
【００７８】
当該追加の前後で、ノード２１〜２３に収容されている端末間の通信は、まったく影響を受けない。
【００７９】
また、図６（Ｃ）および（Ｄ）から明らかなように、ノード３３の追加の前後でパケットの伝送、処理動作が変化しているのは、右端ノード２３だけである。左端ノード２１および中間ノード２２の動作はまったく変化していない。
【００８０】
左端位置のノード２１の左側に新たなノードを追加する場合でも、これらの点は同じである。
【００８１】
換言するなら、各ノードは、ネットワーク上で自身に隣接する箇所で起きた変化（ノードの追加、削除、メンテナンス、障害発生など）にだけ対応すればよく、しかもその変化の検出はソフトウエア的にもハードウエア的にもシンプルなもので、ケーブルの接続、非接続を検出するだけで実行される。これは、後述するノードの削除、変更、障害検出の場合などにもあてはまる。
【００８２】
また、本実施形態によれば、従来の図６（Ｂ）に相当する過渡状態が存在せず、図６（Ｃ）の状態から直接的に同図(Ｄ）の状態に移行することができるので、ノードを追加するための操作が簡略である。これはノードの変更や削除の場合にもあてはまり、メンテナンス性、信頼性を向上する要因となる。
【００８３】
たとえば、図１の状態のネットワーク２０からノード２３を削除する場合、ノード２２の位置は中間位置から右端位置に変化するので、スイッチ部２６Ｂの接続選択の状態も、図示のノード２３内のスイッチ部２６Ｃの接続選択の状態と同じになるように変化する。
【００８４】
この場合も、当該ノード２３の削除の前後で、他のノードに収容されている端末間の通信はまったく影響を受けない。
【００８５】
この点は、図１の状態からノード２１を削除した場合でも同じである。
【００８６】
また、ネットワーク２０のノード数が８〜６程度であるケースで、中間位置のノードを１つ削除する場合、１つのネットワーク２０が過渡的には、たとえばノード数３〜４程度の２つのネットワークに分割されて運用されることになる。
【００８７】
この場合、分割で接続が断たれたネットワーク間の通信は影響を受け得るものの、同一のネットワーク内の通信は、当該削除の前後で、まったく影響を受けない。
【００８８】
なお、ノードの変更は、上述したノードの削除とノードの追加を合成した操作であるので、その説明は省略する。
【００８９】
また、以上ではケーブル側コネクタ３０Ａ、３０Ｂを着脱する場合について説明したが、たとえばケーブル３０Ａやノード２１の障害のため、前記の接続通知伝送路３１Ａの接続が断たれるようなことがあれば、ノード２２は自律的に、自身が左端位置に変更されたものと認識し、スイッチ部２６Ｂの接続選択の機能がはたらいて、ノード２１はネットワーク２０から切り離される。
【００９０】
このような場合、スイッチ部２６Ａ〜２６Ｃの接続選択の機能は、障害検出機能および障害の影響がネットワーク上を波及することを最小限に防止する機能として積極的にとらえることができる。一般に、接続通知伝送路の接続が断たれるような場合は、パケットの処理や伝送も正常に行うことができなくなっている可能性が高いからである。
【００９１】
すなわち、スイッチ部２６Ａ〜２６Ｃの接続選択機能はそのまま、非常に簡略な障害診断機能ともなっている。
【００９２】
次に、論理的にも物理的にもリング型のトポロジを有する従来のネットワーク１０と本実施形態のネットワーク２０で、ネットワーク中の任意の２つのノード間の直線的な最大距離を、ノード数を変更しながら比較する。
【００９３】
送信部Ｓの増幅度および出力パワーやケーブルの特性などで決まる１ノード間隔あたりの伝送距離能力の最大値をＤＬとし、ネットワーク中の全ノード数をＮとし、問題のネットワークの最大距離をネットワーク１０ではＭ１、ネットワーク４０ではＭ２とする。
【００９４】
そして、Ｍ１、Ｍ２をＮを変数とする関数とみて、Ｍ１をＭ１（Ｎ）、Ｍ２をＭ２（Ｎ）と書く。
【００９５】
ノード数Ｎ＝２の場合は、Ｍ１（２）＝Ｍ２（２）。
【００９６】
ここで、Ｍ１（２）＝Ｍ２（２）＝ＤＬ（定数）である。
【００９７】
Ｎが、Ｎ＝３，４，５，…と増加してゆくとＭ２（Ｎ）はＤＬずつ単調に増加してゆき、Ｍ２（Ｎ）＝（Ｎ−１）×ＤＬの関係が成立する。
これに対して、Ｎが、Ｎ＝３，４，５，…と増加してゆくとＭ１（Ｎ）はＮが偶数から奇数に変わるときは増加せず、Ｎが奇数から偶数に変わるときだけＤＬずつ増加するので、Ｍ１（Ｎ）は、Ｍ２（Ｎ）の半分、あるいはＭ２（Ｎ）の半分にＤＬ／２を加えた値を交互にとることになる。
【００９８】
この結果、一般的に、本実施形態のネットワーク２０では問題の最大距離は、従来のネットワーク１０のほぼ２倍となる。
【００９９】
（Ａ−３）第１の実施形態の効果
以上のように本実施形態によれば、ノードの追加、変更、削除、メンテナンス、および障害発生などのさい、各ノードは、ネットワーク上で自身に隣接する箇所で起きた変化にだけ対応すればよいので、その機能に比較して各ノードの規模は、ソフトウエア的にもハードウエア的にも小さく構成することができ、ノードおよびネットワークのメンテナンス性や信頼性が高い。
【０１００】
また、伝送距離、伝送路の特性、周囲環境などの諸条件がそれほど過酷でない場合、本実施形態のデイジーチェイン形式のネットワークでは、ネットワーク中の任意の２つのノード間の直線的な最大距離を、従来にくらべてほぼ２倍に設定することも可能となる。したがって、同じノード数で、従来よりも遠距離地点にノードを設置し、地理的に遠くに存在する端末をネットワークに収容することができるので、リソースの有効利用、ネットワークの効率的な運用が可能になる。
【０１０１】
（Ｂ）第２の実施形態
第１の実施形態の右方向の伝送路ＰＲでは、左端ノード２１から送信された信号は、いったん中間ノード２２の受信部２７Ｂ、送信部２８Ｂで、波形整形や増幅作用を受けてから、右端ノード２３まで伝送される。
【０１０２】
これに対し左方向の伝送路ＴＬでは、右端ノード２３内の送信部２８Ｃから送信された信号は、左端ノード２１内の受信部２７Ａで受信されるまで、ケーブル３０Ｂ、ノード２２、ケーブル３０Ａなどの伝送路を単に通過するだけなので、単調に減衰し劣化する。中間ノード２２での波形整形や増幅が行われないため、右方向の伝送路ＰＲにくらべて、減衰や劣化の程度は大きなものとなる。
【０１０３】
そしてこの減衰や劣化が、実際のノード間隔を、上述した１ノード間隔あたりの伝送距離能力の最大値ＤＬよりも小さく制限する要因となる。
【０１０４】
すなわち、伝送距離、伝送路の特性、周囲環境などによっては、受信した右端ノード２１で正常な処理を行うことができない可能性が高まり、最大値ＤＬまでノード間隔を広げることが困難になるため、第１の実施形態のネットワーク構成が本来もっている性能をフルに活用することができなくなる可能性がある。
【０１０５】
本実施形態はこのような点に対処するために第１の実施形態を改良したものである。
【０１０６】
（Ｂ−１）第２の実施形態の構成および動作
第２の実施形態のネットワーク４０を図５に示す。
【０１０７】
図５において、図１の各部と同一の符号で対応付けられた部分は、図１の各部と同一なので、その詳しい説明は省略する。
【０１０８】
すなわち、図５において、符号２４Ａ〜Ｃ，２５Ａ〜Ｃ，２６Ａ〜Ｃ，２７Ａ〜Ｃ，２８Ａ〜Ｃ，２９Ａ〜Ｃ，３０ＡおよびＢを付した各部分は、同一の符号を付した図１の各部分と同じである。
【０１０９】
上述したネットワーク２０と同様、ネットワーク４０では、ノード４１と４３のあいだの空間的な距離は、一般的な仕様の機器、ケーブルを前提として、最大で、数百ｍ〜数ｋｍ程度となる。したがって伝送路ＴＬでは、パケットは数百ｍ〜数ｋｍ程度の距離をなんらの処理も受けることなく伝送されることになり、伝送途中での信号の減衰や劣化が懸念される。
【０１１０】
図５において、ネットワーク４０の中間位置のノード４２のなかに図示されている信号増幅器（ＡＭＰ）４４は、スイッチ部２６Ｂが右側コネクタ２５Ｂの出力端子ＲＯの接続先として左側コネクタ２４Ｂの入力端子ＬＩを選択するとき、すなわち伝送路ＴＬ上で、信号を増幅し、波形を整えるための回路である。
【０１１１】
この信号増幅器４４は、第１の実施形態で述べた送信部２８Ａ、受信部２７Ａの内部に設けられている増幅回路と同じものであってよい。もちろん、増幅度を変えるなど、送信部２８Ａなどの増幅回路とは異なる仕様、異なる回路構成の回路としてもよい。
【０１１２】
そしてネットワーク内に中間ノードが複数存在する場合は、中間ノードごとに、信号増幅器が信号を増幅し、波形整形する。
【０１１３】
図５では、左端ノード４１および右端ノード４３の内部には信号増幅器が示されていないが、ノードの追加、削除、変更があり得る通常の条件下では、ネットワーク内のすべてのノードが中間ノードになる可能性があるので、右端位置および左端位置のノードにも信号増幅器を設けるようにするとよい。
【０１１４】
ただし信号増幅器の必要性の高さは、その性質上、伝送距離、伝送路の特性、周囲環境（ネットワーク内の不要輻射、外来雑音等のノイズ要因など）のような諸条件に大きく依存するので、これらの条件によっては、信号増幅器を搭載しているノードと搭載していないノードが同じネットワーク内に混在するような運用形態も現実的である。
【０１１５】
なお、ノードの削除や障害発生などでノード４２が右端ノードまたは左端ノードとなったときは、スイッチ部２６Ｂの接続選択も変わり、信号増幅器４４は使用されなくなる。
【０１１６】
(Ｂ−２）第２の実施形態の効果
以上説明したように、本実施形態によれば、第１の実施形態とまったく同じ効果を得ることができる。
【０１１７】
これに加えて、本実施形態では、中間ノード内の信号増幅器のはたらきによって、論理的な処理が行われない方向の伝送路（ＴＬ）でも波形整形や増幅を行うことで、伝送による信号の減衰や劣化を軽減することができ、伝送距離、伝送路の特性、周囲環境などの諸条件がネットワークにとって過酷な場合でも、通信品質を維持することが可能で、いっそうの信頼性向上が期待できる。
【０１１８】
さらに、本実施形態によれば、伝送路ＴＬでの信号の減衰、劣化の影響を抑制できるので、送信部Ｓの増幅度および出力パワーやケーブルの特性などで決まる１ノード間隔あたりの伝送距離能力の最大値ＤＬ付近まで、ノード間の距離を伸ばすことができ、ネットワーク設計の自由度が増大する。
【０１１９】
また、中間ノードの信号増幅器のはたらきによって、上述した諸条件が過酷な場合であっても、第１の実施形態よりも確実にノード間隔を最大値ＤＬに近づけることができ、ネットワーク中の任意の２つのノード間の直線的な最大距離を、従来にくらべてほぼ２倍とすることができる。
【０１２０】
このため、本実施形態（および第１の実施形態）のデイジーチェイン形式のネットワーク構成が本来もっている性能をフルに引き出すことができ、同じノード数で、従来よりも遠距離にノードを設置して地理的に遠くの地点に存在する端末をネットワークに収容することができるので、リソースの有効利用、ネットワークの効率的な運用を実現することができる可能性が高まる。
【０１２１】
（Ｃ）他の実施形態
以上の説明において、ノード間を接続するケーブルは、電気信号を伝送する同軸ケーブルであったが、光ファイバケーブルを使用するようにしてもよい。マルチモードの光ファイバを使用する場合、端部のノード２１と２３のあいだの空間的な距離は、一般的な仕様の機器、ケーブルを前提とすると、最大で、たとえば２００ｍ程度になる。同じ前提のもとにシングルモードを使用すると、この距離は最大で１ｋｍ程度となる。
【０１２２】
また、ノード間の接続を検出するための構成は、前記プルアップ抵抗や接地端子を利用したものにかぎらず、他の回路構成を用いてもよい。
【０１２３】
さらに、以上の説明では、たとえば１本のケーブルであるケーブル３０Ａのなかに左方向の伝送路ＴＬおよび右方向の伝送路ＰＲが存在するものとしたが、１方向につき１本のケーブルを使用し、たとえばケーブル３０Ａを２本のケーブルで置換するようにしてもよく、必要に応じて３本以上のケーブルを使用してもよい。
【０１２４】
なお、以上の説明では、左端位置のノード２１には右側コネクタＣＮＲにケーブルを接続し、右端位置のノード２３には左側コネクタＣＮＬにケーブルを接続するものとしたが、左右のコネクタＣＮＬ、ＣＮＲの区別は便宜的なものである。すなわち、左右のコネクタＣＮＬ、ＣＮＲの区別は、各ノードの内部で各ノードシステムが把握していれば十分であり、ノードの外部からみた場合これらを区別する意味はない。
【０１２５】
したがって、たとえば、図１の右端位置のノード２３の右側コネクタ２５Ｃにケーブル３０Ｂを接続したとしても、ノード２３およびネットワーク２０の動作上なんら矛盾は生じない。
【０１２６】
また、第１、第２の実施形態とも、論理的なリング型の経路は１重であったが、２重以上の多重ループとしてもよい。
【０１２７】
さらにまた、上述したネットワーク規模、すなわちノード数、ノードに収容される端末数、右端ノードと左端ノード間の空間的な距離の最大値などは、本発明の適用にあたって、好ましいと考えられる例を示しただけであって、本発明の適用範囲がこれらの数値に限定されるものではない。
【０１２８】
そして、以上の実施形態では、ネットワーク上を伝送される単位信号は、固定長のＡＴＭセルであったが、可変長のパケットを用いるようにしてもよい。
【０１２９】
すなわち、本発明は、少くとも論理的にはリング型のトポロジを有して、単位信号を伝送するネットワーク、このようなネットワークで使用され得るノード、およびノードの接続方法について、広く適用することができる。
【０１３０】
【発明の効果】
以上のように、本発明によれば、論理的トポロジはリング型、物理的トポロジは直列的なデイジーチェイン形式のネットワークを運用しているとき、コネクタに対するケーブル接続で隣接している各ノードが、ケーブルの接続状態に応じて、自ノードのなかでの前記単位信号の伝送経路を変更することにより、ネットワークの論理的、物理的トポロジを維持するので、各ノードは、ネットワーク上で自身に隣接する箇所で起きた変化にだけ対応すればよい。デイジーチェイン上の当該変化箇所に隣接するノード以外のノードは、当該変化以前の動作状態を維持すればよく、当該変化が起きたことを検出する必要もない。
【０１３１】
この変化のなかには、ノードの追加、変更、削除、メンテナンス、障害発生などが含まれ得るため、その機能に比較して各ノードの規模は、ソフトウエア的にもハードウエア的にも小さく構成することができ、ノードおよびネットワークのメンテナンス性、信頼性が高い。
【０１３２】
また、本発明のデイジーチェイン形式のネットワークによれば、ネットワーク中の任意の２つのノード間の直線的な最大距離が、従来にくらべてほぼ２倍となり、同じノード数で、従来よりも遠距離地点に存在する端末をネットワークに収容することができるので、リソースの有効利用、ネットワークの効率的な運用が可能になる。
【図面の簡単な説明】
【図１】第１の実施形態にかかるネットワークおよびノードの構成を示すブロック図である。
【図２】従来のネットワークおよびノードの構成を示すブロック図である。
【図３】第１の実施形態にかかるノードがケーブルの接続状態を検出するための構成を示すブロック図である。
【図４】第１の実施形態でノードを追加した状態を示す概略図である。
【図５】第２の実施形態にかかるネットワークおよびノードの構成を示すブロック図である。
【図６】第１の実施形態および従来のネットワークにおいて、ノードを追加する場合の動作変化を示す概略図である。
【符号の説明】
１０、２０、４０…ネットワーク、１１〜１３、２１〜２３、３３、４１〜４３…ノード、２４Ａ〜２４Ｃ…左側コネクタ、２５Ａ〜２５Ｃ…右側コネクタ、２６Ａ〜２６Ｃ…スイッチ部、２７Ａ〜２７Ｃ…受信部、２８Ａ〜２８Ｃ…送信部、２９Ａ〜２９Ｃ…パケット処理部、３０Ａ〜３０Ｃ…ケーブル、３１Ａ、３１Ｂ…接続通知伝送路、ＣＯＮ１、ＣＯＮ２…コネクト端子、ＴＬ…左方向の通信用伝送路、ＰＲ…右方向の通信用伝送路。[0001]
BACKGROUND OF THE INVENTION
The present invention Is used to connect nodes used in the network. It is related.
[0003]
Furthermore, the present invention relates to a method for connecting nodes used in such a network.
[0004]
[Prior art]
An example of a ring-type communication network having a logical topology is ATM-LAN described in the following document 1.
[0005]
Reference 1 "Realization of ATM-LAN for plant monitoring and control"
Authors Kazuyuki Kashima, Kazunori Kodaka, Junichi Hirota, Akihiko Kubo,
Ken Murakami, Tachiki Ichihashi
Source: IEICE Communication Society Conference (B-729, p.214, 1996)
FIG. 2 shows the basic configuration of the ring network disclosed in Document 1.
[0006]
In FIG. 2, the network 10 has three nodes 11 to 13. The nodes 11 to 13 are connected by two loops A and B so as to be logically doubled.
[0007]
Inside the node 11, there are an input side switch SW (switch) 14A for inputting an ATM (asynchronous transfer mode) cell, an R (reception unit) 15A, an S (transmission unit) 16A, and an output side SW 17A. Is provided.
[0008]
In addition, S, R15A, and 16A are S as a transmission unit and R as a reception unit in the state shown in the figure, but S16A functions as a transmission / reception unit when looping back. It is also possible to function.
[0009]
The other nodes 12 and 13 have the same configuration as that of the node 11, and the input side SW 14 B, R 15 B, S 16 B and the output side SW 17 B are provided in the node 12, and the input side SW 14 C, R 15 C, S 16 C are provided in the node 13. An output side SW17C is provided.
[0010]
The turn-back process with the function change of the above-described S16A to 16C and R15A to 15C, such as when a failure occurs or during maintenance, is performed at the two corresponding nodes.
[0011]
For example, between the nodes 11 and 12, the SW 17A in the node 11 and the SW 14B in the node 12 become an act system in which either loop A or B is used for communication, and the other becomes a standby system not used for communication. .
[0012]
Whether the loops A and B are used or not is determined independently between the nodes 12 and 13 and between the nodes 13 and 11.
[0013]
It is assumed that loop A is an act system and loop B is a standby system among all nodes.
[0014]
For example, if a failure occurs in the act loop A between the nodes 11 and 12, ATM cells are turned back in the nodes 11 and 12, and the transmission path between the nodes 11 and 12 is not used for both the loops A and B.
[0015]
Then, the cell transmitted from the node 13 and transmitted to the node 11 is looped back in the node 11 and transmitted to the node 13 using the loop B that has been a standby system until then.
[0016]
At this time, also in the node 12, the loop-back transmission of the cell is performed between the nodes 12 and 13 using the loop B that has been a standby system until then.
[0017]
Changes in the topology of the network 10 due to the return are as shown in FIGS. 6 (A) and 6 (B). The topology of the network 10 in FIG. 6A is a ring type logically and physically.
[0018]
The topology of the network 10 that was in the ring shape in FIG. 6A is, for example, when a new node 18 is added between the nodes 11 and 13, transiently due to the loopback processing in the nodes 11 and 13. 6 (B).
[0019]
The state of FIG. 6B is the same as the operation when a failure occurs in the node 18 assuming a ring network originally composed of the nodes 11 to 13 and the node 18.
[0020]
The ATM terminals accommodated in the nodes 11 to 13 can perform the same communication as before the addition of the node 18 or the occurrence of a failure even in the transient state of FIG. 6B.
[0021]
[Problems to be solved by the invention]
However, in the network 10 described above, when trying to add a new node to the network 10 or when deleting an existing node from the network, the operation of the entire network is necessarily stopped, Alternatively, the above-described loopback process must always be performed at two nodes. Since the suspension of operation significantly reduces the reliability of the network, let us consider a case where the network is turned back on while the network is in an operational state.
[0022]
The loopback processing in the network 10 must always be performed at two nodes adjacent to the problem location such as a maintenance location or a failure location on the network.
[0023]
In the case of a ring topology, any one node must process all ATM cells transmitted over the network, so each node is responsible for ensuring that the received cells are sent out to the next node. is there. In order to fulfill this responsibility, the two nodes must reliably execute the loopback process.
[0024]
In this loopback process, it is necessary to determine which node on the network is to be looped in which direction. However, assuming a distributed system, all nodes on the network, including nodes at intermediate positions, perform this determination. It is necessary to have a function to perform.
[0025]
Further, as apparent from FIGS. 6A and 6B, the operation performed by the node at the intermediate position where the folding process is not performed, that is, the node 12, is also changed during the folding. In FIG. 6A, the node 12 that has transmitted and processed the cell only in the counterclockwise direction must transmit the cell in the clockwise direction in addition to the counterclockwise direction in the transient state of FIG. 6B. .
[0026]
This means that when there is a return, all the nodes that make up the network must perform different actions than before.
[0027]
In addition to the above determination, all nodes on the network must be equipped with a fault diagnosis capability in order to execute an operation different from that before returning. The contents of the fault diagnosis ability are considered to be the ability to detect the occurrence of a fault and the ability to identify the location where the fault has occurred.
[0028]
On the other hand, it is always necessary to perform the loopback process on two nodes and perform a different operation from the loopback on all nodes on the network including the node at the intermediate position. It tends to be a factor that deteriorates operability and work efficiency.
[0029]
For example, when a new node is added between the nodes 11 and 12 in the network 10, a loopback process is performed in the nodes 11 and 12, so that neither the loop A nor B is used between the nodes 11 and 12, and the node In 13, it is necessary to start bidirectional transmission.
[0030]
This is the same regardless of where the node to be added is installed in the network 10, such as between the nodes 12 and 13, and is the same even when an existing node is deleted or changed.
[0031]
[Means for Solving the Problems]
To solve this problem , According to the present invention, there is provided a method for connecting nodes of a network that at least logically has a ring topology and transmits unit signals. (1) The network is at least physically configured in a serial topology. Daisy chain format Complete Each no Do , (1-1) Between nodes Connection notification that conveys the connection between the node and the transmission line that transmits the unit signal Connect the cables that make up the transmission path A grounding terminal and a pull-up resistor connecting terminal connected to a cable as a connection notification transmission line, and a voltage detection unit A pair of left and right connectors And (1-2) a switch unit that selectively selects the connection of a pair of left and right connectors; Set up The voltage detection unit recognizes the connection between the node and the connection destination node connected through the cable by detecting the voltage of the pull-up resistor connection terminal, and the switch unit detects the pull-up resistor connection detected by the voltage detection unit. Select the connector on the cable side to be connected to the terminal It is characterized by that.
[0034]
DETAILED DESCRIPTION OF THE INVENTION
(A) Embodiment
The node connection method according to the present invention The law The first and second embodiments of the present invention will be described using a case where the present invention is applied to ATM-LAN as an example.
[0035]
In the first and second embodiments, the network is logically ring-type and has a physical (ie, electrical, mechanical) serial daisy chain configuration. Each node constituting the network is characterized in that the selection of the connection of each part is changed in the own node according to the connection state of the cables to the left and right connectors.
[0036]
(A-1) Configuration of the first embodiment
The network 20 of this embodiment is shown in FIG.
[0037]
In FIG. 1, the network 20 includes three nodes 21 to 23. Each of the nodes 21 to 23 is a transmission device having an exchange function such as a router or an exchange. Although the node 21 is located at the left end of the daisy chain, the node 22 is located at the middle, and the node 23 is located at the right end, the positions on the network 21 are different, but the internal configurations of the nodes 21 to 23 are common. It is.
[0038]
The connectors in the node 21 are a device side connector (CNL) 24A on the left side and a device side connector (CNR) 25A on the right side. The right connector 25A includes an output terminal RO and an input terminal RI for the inside of the node 21. Similarly, the left connector 24A also includes an output terminal LO and an input terminal LI for the inside of the node 21.
[0039]
A switch unit (SW) 26A connected to the terminals RI, RO, LI, and LO of the connectors 24A and 25A by a signal line (not shown) according to the connection state of a cable side connector to be described later with respect to the device side connectors 24A and 25A. Thus, the electronic connection inside the switch unit 26 is switched.
[0040]
That is, as shown in the figure, when the cable 30A is connected to the connector 25A and there is no cable to be connected to the connector 24A, the switch unit 26A responds to such a connection state and the input terminal RI of the connector 25A. Are connected to the output terminal of the transmitting section (S) 28A, and the input terminal of the receiving section (R) 27A is connected to the output terminal RO of the connector 25A.
[0041]
Arrows (TL, PR) shown in FIG. 1 indicate logical transmission paths of packets inside and outside the nodes 21 to 23. Each of TL and PR corresponds to approximately half of the ring topology, and TL and PR are combined to form one ring.
[0042]
As a cable to be used, a coaxial cable or a twisted pair cable that transmits an electrical signal can be used. In this embodiment, the cable is a coaxial cable.
[0043]
Bidirectional transmission lines TL and PR are present inside the cable 30A, which is such a single coaxial cable, as shown in the figure.
[0044]
The transmitter 28A and the receiver 27A include an amplifier circuit having a waveform shaping unit. Using this amplification circuit, the transmitter 28A amplifies the signal to a sufficient strength in consideration of attenuation due to the transmission of the cable 30A and the like, and then arranges the waveform and sends out the signal. The waveform of the attenuated and deteriorated signal is adjusted, amplified, and supplied to the packet processing unit 29A.
[0045]
The packet processing unit 29A is a circuit that processes a signal supplied from the receiving unit 27 and a signal supplied from an ATM terminal accommodated in the node 21 via a user interface unit (not shown).
[0046]
When the packet is received from the receiving unit 27A, the packet processing unit 29A determines that the packet processing unit 29A is a packet having an address designating the ATM terminal accommodated by the node 21 according to the address in the packet header. The packet is sent to the corresponding ATM terminal via the interface unit, and if it is a packet of any other address, it is sent to the sending unit 28A as it is and relayed.
[0047]
On the other hand, for a signal supplied from an ATM terminal accommodated in the node 21 by designating a terminal accommodated in another node, the packet processing unit 29A converts the signal into a predetermined packet format. After the configuration, the data is sent to the transmission unit 28A.
[0048]
In addition, the packet processing unit 29A performs necessary processing such as transmission error check.
[0049]
In the internal configuration of the other nodes 22 and 23, portions corresponding to the respective portions in the node 21 are denoted by corresponding reference numerals, and detailed description thereof is omitted.
[0050]
That is, in the node 22, the connector 24B corresponds to the connector 24A, the connector 25B corresponds to the connector 25A, the switch unit 26B corresponds to the switch unit 26A, and the receiving unit 27B corresponds to the receiving unit 27A. The transmission unit 28B corresponds to the transmission unit 28A, and the packet processing unit 29B corresponds to the packet processing unit 29A.
[0051]
In the node 23, the connector 24C corresponds to the connector 24A, the connector 25C corresponds to the connector 25A, the switch unit 26C corresponds to the switch unit 26A, and the receiving unit 27C corresponds to the receiving unit 27A. The transmission unit 28C corresponds to the transmission unit 28A, and the packet processing unit 29C corresponds to the packet processing unit 29A.
[0052]
A cable 30B connecting the nodes 22 and 23 is the same cable as the cable 30A.
[0053]
In the transmission path for communication on the network 20, in the right direction, that is, the transmission path PR from the node 21 to the node 23 through the node 22, each packet processing unit 29A, 29B, 29C in the nodes 21 to 23 sequentially Although logical processing is performed, in the transmission path TL in the left direction, that is, from the node 23 to the node 21 through the node 22, there is neither a transmission unit nor a reception unit, and the nodes 21 to 23 simply pass signals. Just let it.
[0054]
Note that this transmission line TL corresponds to a half-round portion of a logical ring topology in which no transmission unit and reception unit exist.
[0055]
In the present embodiment, the spatial distance between the nodes 21 and 23 is about several hundreds m to several km at maximum, assuming a general specification of equipment and cables.
[0056]
FIG. 1 shows three nodes 21 to 23. As the scale of the network 20, for example, the number of nodes is about 8 to 6, and the number of ATM terminals accommodated in each node is about 20 to 30. Is preferred.
[0057]
The operation of the first embodiment having the above configuration will be described below.
(A-2) Operation of the first embodiment
In the network 20 of FIG. 1, a packet sent from the transmitting unit 28A of the node 21 at the left end position is transmitted to the node 22 via a path (that is, a transmission path PR) connecting the connector 25A, the cable 30A, and the connector 24B of the node 22. To the switch unit 26B.
[0058]
Since the node 22 has the cable 30A and the node 21 connected to the left connector 24B and the cable 30B and the node 23 connected to the right connector 25B, the switch unit 26B performs connection selection as shown in the figure. Yes.
[0059]
That is, the connector 2 is connected to the output terminal RO of the connector 25B. 4 The B input terminal LI is selected, the output terminal of the transmission unit 28B is selected as the connection destination of the input terminal RI of the connector 25B, and the input terminal of the reception unit 27B is selected as the connection destination of the output terminal LO of the connector 24B. .
[0060]
For example, the node 22 autonomously performs such connection selection based on the structure shown in FIG.
[0061]
In FIG. 3, the cable 30 </ b> A includes a connection notification transmission path 31 </ b> A for transmitting a connection between nodes in addition to transmission paths TL and PR for transmitting packets. The connection notification transmission path 31A is connected to the node (that is, the cable 30A). of By electrically connecting the node (first connected node) 21 and the node 22 which is the connection destination node, both nodes are notified of this connection.
[0062]
When the cable-side connector 32A is attached to the corresponding slot of the node 22, the cable connector 32A is connected to the node 22. In the cable-side connector 32A, the connection notification transmission path 31A can be formed in the same shell as a contact in the same manner as a contact such as a pin provided in the shell of the cable-side connector 32A for transmitting a packet.
[0063]
As shown in FIG. 3, the node 22 detects the connection of the node to the left connector 24B at the connection terminal CON1 of the switch unit 26B. Similarly, the connection of the node to the right connector 25B can be detected by the connection terminal CON2 of the switch unit 26B.
[0064]
However, for this detection to be performed on both sides, it is assumed that both the connection destination node and the connection source node are at least energized (a state in which packet processing is possible).
[0065]
In general, simply connecting a cable connector to a slot of a certain node makes it unclear whether the connection source node is in a state where packets can be processed. In the present embodiment, a configuration in which the connection source and the connection destination can detect the connection only in the energized state. If this is not the case, an operation of notifying both sides that the packet can be processed becomes necessary separately from the attachment of the cable connector.
[0066]
The specific configuration of the connection notification transmission path 31A and the connection terminal CON1 in FIG. 3 includes, for example, a signal line for connecting from the connection source to the connection destination (node 21 → node 22) as the connection notification transmission path 31A, and the connection destination. A signal line for the direction from the connection source (node 22 to node 21) is provided, and a grounded terminal and a terminal connected to the pull-up resistor are provided as the connection terminal CON1, and the pull-up is further performed. It is preferable to provide a circuit for detecting the terminal voltage of the resistor.
[0067]
If the connection terminal CON2 of the switch section 26B of the node 22 and the other nodes 21 and 23 have the same configuration, when the nodes 21 and 22 are connected by the cable 30A, the connection terminal CON1 By connecting one end of the pull-up resistor to the grounded terminal of the node 21, the terminal voltage of the pull-up resistor is changed. small The node 22 that has detected this can recognize the connection of the node 21 to the left connector 24B.
[0068]
At this time, similarly, the connection can be recognized also on the node 21 side.
[0069]
These series of operations are the same in connection between the nodes 22 and 23 performed via the cable 30B and the cable connector 32B.
[0070]
Therefore, also in the node 21 and the node 23, the connection selection in the switch units 26A and 26C can be performed autonomously.
[0071]
The relationship between the positions of the nodes 21 to 23 in the network 20 and the connection detection (ON) and connection non-detection (OFF) of the connection terminals CON1 and CON2 is summarized in Table 1 below.
[0072]
[Table 1]

In Table 1, the connection terminals CON1 and CON2 are both ON, which means that both the left and right connectors of the node are connected. Such a connected state is the node 22 existing in the middle position of the network 20.
[0073]
Further, CON1 is ON and CON2 is OFF when the left connector of the node is connected and the right connector is not connected, which corresponds to the connection state of the node 23 existing at the right end position of the network 20.
[0074]
Furthermore, CON1 is OFF and CON2 is ON when the left connector of the node is not connected and the right connector is connected, which corresponds to the connection state of the node 21 existing at the left end position of the network 20.
[0075]
Note that both of CON1 and CON2 being OFF means that the node is not connected to the network.
[0076]
In general, the recognition of its own position on the network by the node according to the present embodiment is linked to the selection of the connection in the switch unit in the node.
[0077]
For example, as shown in FIGS. 6C and 6D and FIG. 4, by connecting a new node 33 to the right side of the node 23 at the right end position using the cable 30C similar to the cables 30A and 30B. When adding a node to the network 20, the node 23 becomes a node at an intermediate position, and the new node 33 becomes a node at the right end position.
[0078]
Before and after the addition, communication between terminals accommodated in the nodes 21 to 23 is not affected at all.
[0079]
As is clear from FIGS. 6C and 6D, only the right end node 23 changes the packet transmission and processing operations before and after the addition of the node 33. The operations of the left end node 21 and the intermediate node 22 are not changed at all.
[0080]
Even when a new node is added to the left side of the node 21 at the left end position, these points are the same.
[0081]
In other words, each node only needs to respond to changes (node addition, deletion, maintenance, failure occurrence, etc.) that occurred in a location adjacent to itself on the network, and the change detection is performed in software. It is simple both in terms of hardware and can be executed simply by detecting cable connection / disconnection. This also applies to node deletion, change, and failure detection, which will be described later.
[0082]
In addition, according to the present embodiment, there is no transient state corresponding to the conventional FIG. 6B, and the state of FIG. 6C can be directly shifted to the state of FIG. Therefore, the operation for adding a node is simple. This also applies to node changes and deletions, and is a factor that improves maintainability and reliability.
[0083]
For example, when the node 23 is deleted from the network 20 in the state of FIG. 1, the position of the node 22 changes from the intermediate position to the right end position, so that the connection selection state of the switch unit 26B is also the switch unit in the illustrated node 23. It changes to be the same as the connection selection state of 26C.
[0084]
Also in this case, before and after the deletion of the node 23, communication between terminals accommodated in other nodes is not affected at all.
[0085]
This is the same even when the node 21 is deleted from the state of FIG.
[0086]
Further, in the case where the number of nodes of the network 20 is about 8-6, when one node at the intermediate position is deleted, one network 20 transitions to two networks having about 3-4 nodes, for example. It will be divided and operated.
[0087]
In this case, communication between networks that are disconnected due to division may be affected, but communication within the same network is not affected at all before and after the deletion.
[0088]
Note that the change of the node is an operation in which the deletion of the node and the addition of the node described above are combined, and thus the description thereof is omitted.
[0089]
Further, the case where the cable-side connectors 30A and 30B are attached and detached has been described above. However, for example, if the connection of the connection notification transmission path 31A is disconnected due to a failure of the cable 30A or the node 21, The node 22 autonomously recognizes that the node 22 itself has been changed to the left end position, the connection selection function of the switch unit 26B works, and the node 21 is disconnected from the network 20.
[0090]
In such a case, the connection selection function of the switch units 26A to 26C can be positively regarded as a fault detection function and a function for preventing the influence of the fault from spreading on the network to the minimum. This is because, in general, when the connection of the connection notification transmission path is disconnected, there is a high possibility that packet processing and transmission cannot be performed normally.
[0091]
That is, the connection selection function of the switch units 26A to 26C is also used as it is as a very simple fault diagnosis function.
[0092]
Next, in the conventional network 10 having a ring topology both logically and physically, and the network 20 of the present embodiment, the maximum linear distance between any two nodes in the network is determined by the number of nodes. Compare while changing.
[0093]
The maximum value of the transmission distance capability per one node interval determined by the amplification factor of the transmission unit S, output power, cable characteristics, and the like is DL, the total number of nodes in the network is N, and the maximum distance of the network in question is the network 10 Here, it is assumed that M1 and M2 in the network 40.
[0094]
Then, considering M1 and M2 as functions having N as a variable, M1 is written as M1 (N) and M2 is written as M2 (N).
[0095]
When the number of nodes N = 2, M1 (2) = M2 (2).
[0096]
Here, M1 (2) = M2 (2) = DL (constant).
[0097]
When N increases as N = 3, 4, 5,..., M2 (N) increases monotonously by DL, and the relationship M2 (N) = (N−1) × DL is established.
On the other hand, when N increases as N = 3, 4, 5,..., M1 (N) does not increase when N changes from even to odd, but only when N changes from odd to even. Since it increases by DL, M1 (N) alternately takes half of M2 (N) or half of M2 (N) plus DL / 2.
[0098]
As a result, in general, the maximum distance in question in the network 20 of the present embodiment is almost twice that of the conventional network 10.
[0099]
(A-3) Effects of the first embodiment
As described above, according to the present embodiment, when a node is added, changed, deleted, maintained, or a failure occurs, each node only needs to respond to a change that occurs at a location adjacent to itself on the network. Therefore, the scale of each node can be small in terms of software and hardware compared to its function, and the maintainability and reliability of the node and network are high.
[0100]
In addition, when various conditions such as transmission distance, transmission path characteristics, and ambient environment are not so severe, in the daisy chain type network of this embodiment, the linear maximum distance between any two nodes in the network is It is possible to set almost twice as much as before. Therefore, with the same number of nodes, nodes can be installed at farther points than before, and terminals that are geographically far away can be accommodated in the network, enabling efficient use of resources and efficient operation of the network. become.
[0101]
(B) Second embodiment
In the right-direction transmission line PR of the first embodiment, the signal transmitted from the left end node 21 is subjected to waveform shaping and amplification operations once by the receiving unit 27B and the transmitting unit 28B of the intermediate node 22, and then the right end node. 23 is transmitted.
[0102]
On the other hand, in the transmission line TL in the left direction, the signal transmitted from the transmission unit 28C in the right end node 23 is received by the reception unit 27A in the left end node 21 until the signal is received by the cable 30B, the node 22, the cable 30A, etc. Since it simply passes through the transmission line, it monotonously attenuates and deteriorates. Since waveform shaping and amplification at the intermediate node 22 are not performed, the degree of attenuation and deterioration is greater than that of the transmission path PR in the right direction.
[0103]
This attenuation or deterioration becomes a factor that limits the actual node interval to be smaller than the maximum value DL of the transmission distance capability per one node interval described above.
[0104]
That is, depending on the transmission distance, the characteristics of the transmission path, the surrounding environment, etc., the possibility that normal processing cannot be performed by the received right end node 21 increases, and it becomes difficult to expand the node interval to the maximum value DL. There is a possibility that the performance inherent in the network configuration of the first embodiment cannot be fully utilized.
[0105]
This embodiment is an improvement of the first embodiment in order to deal with such a point.
[0106]
(B-1) Configuration and operation of the second embodiment
A network 40 of the second embodiment is shown in FIG.
[0107]
5, parts associated with the same reference numerals as those in FIG. 1 are the same as those in FIG. 1, and thus detailed description thereof is omitted.
[0108]
That is, in FIG. 5, the parts denoted by reference numerals 24 </ b> A to C, 25 </ b> A to C, 26 </ b> A to C, 27 </ b> A to C, 28 </ b> A to C, 29 </ b> A to C, 30 </ b> A and B Same as each part.
[0109]
Similar to the network 20 described above, in the network 40, the spatial distance between the nodes 41 and 43 is about several hundred meters to several kilometers at the maximum on the premise of devices and cables having general specifications. Therefore, on the transmission line TL, the packet is transmitted at a distance of about several hundred m to several km without any processing, and there is a concern that the signal is attenuated or deteriorated during transmission.
[0110]
In FIG. 5, a signal amplifier (AMP) 44 shown in a node 42 at an intermediate position of the network 40 has an input terminal LI of the left connector 24B as a connection destination of the output terminal RO of the right connector 25B. This is a circuit for amplifying a signal and adjusting a waveform when selecting, that is, on the transmission line TL.
[0111]
The signal amplifier 44 may be the same as the amplifier circuit provided in the transmitter 28A and the receiver 27A described in the first embodiment. Of course, a circuit having a different specification and a different circuit configuration from the amplifier circuit such as the transmitter 28A, such as changing the amplification degree, may be used.
[0112]
When there are a plurality of intermediate nodes in the network, the signal amplifier amplifies the signal and shapes the waveform for each intermediate node.
[0113]
In FIG. 5, signal amplifiers are not shown inside the leftmost node 41 and the rightmost node 43, but under normal conditions where there may be node additions, deletions, and changes, all nodes in the network become intermediate nodes. Therefore, it is preferable to provide signal amplifiers at the nodes at the right end position and the left end position.
[0114]
However, the high necessity of signal amplifiers depends greatly on various conditions such as transmission distance, transmission path characteristics, and surrounding environment (unnecessary radiation in the network, noise factors such as external noise). Depending on these conditions, an operation mode in which a node having a signal amplifier and a node not having a signal amplifier are mixed in the same network is also realistic.
[0115]
When the node 42 becomes the right end node or the left end node due to node deletion or failure occurrence, the connection selection of the switch unit 26B is changed, and the signal amplifier 44 is not used.
[0116]
(B-2) Effects of the second embodiment
As described above, according to the present embodiment, exactly the same effects as those of the first embodiment can be obtained.
[0117]
In addition to this, in the present embodiment, signal shaping due to transmission is performed by performing waveform shaping and amplification even in a transmission path (TL) in a direction where logical processing is not performed by the function of the signal amplifier in the intermediate node. Even when conditions such as transmission distance, transmission path characteristics, and surrounding environment are severe for the network, communication quality can be maintained and further improvement in reliability can be expected.
[0118]
Furthermore, according to the present embodiment, it is possible to suppress the effects of signal attenuation and degradation on the transmission line TL, so that the transmission distance capability per one node interval determined by the amplification factor of the transmission unit S, output power, cable characteristics, and the like. The distance between the nodes can be extended to the vicinity of the maximum value DL, and the degree of freedom in network design increases.
[0119]
In addition, due to the function of the signal amplifier of the intermediate node, even when the above-mentioned conditions are severe, the node interval can be made closer to the maximum value DL more reliably than in the first embodiment, and any arbitrary network in the network can be operated. The maximum linear distance between two nodes can be almost doubled compared to the conventional one.
[0120]
For this reason, the performance inherent to the network configuration of the daisy chain format of this embodiment (and the first embodiment) can be fully extracted, and nodes can be installed at a longer distance than before with the same number of nodes. Since terminals located at geographically distant points can be accommodated in the network, the possibility of effective use of resources and efficient operation of the network increases.
[0121]
(C) Other embodiments
In the above description, the cable connecting the nodes is a coaxial cable that transmits an electrical signal, but an optical fiber cable may be used. When a multimode optical fiber is used, the spatial distance between the end nodes 21 and 23 is, for example, about 200 m at the maximum, assuming a device or cable of general specifications. If the single mode is used under the same premise, this distance is about 1 km at the maximum.
[0122]
Further, the configuration for detecting the connection between the nodes is not limited to the configuration using the pull-up resistor or the ground terminal, and other circuit configurations may be used.
[0123]
Furthermore, in the above description, for example, the left transmission line TL and the right transmission line PR exist in the cable 30A, which is a single cable, but one cable is used per direction. For example, the cable 30A may be replaced with two cables, and three or more cables may be used as necessary.
[0124]
In the above description, the cable is connected to the right connector CNR at the node 21 at the left end position, and the cable is connected to the left connector CNL at the node 23 at the right end position, but the left and right connectors CNL and CNR are connected. The distinction is convenient. That is, the left and right connectors CNL and CNR need only be recognized by each node system inside each node, and there is no point in distinguishing them when viewed from outside the node.
[0125]
Therefore, for example, even if the cable 30B is connected to the right connector 25C of the node 23 at the right end position in FIG. 1, there is no contradiction in the operation of the node 23 and the network 20.
[0126]
In both the first and second embodiments, the logical ring type path is single, but it may be a double or more multiple loop.
[0127]
Furthermore, the above-described network scale, that is, the number of nodes, the number of terminals accommodated in the node, the maximum spatial distance between the right end node and the left end node, and the like are examples that are considered preferable in the application of the present invention. However, the scope of application of the present invention is not limited to these numerical values.
[0128]
In the above embodiment, the unit signal transmitted over the network is a fixed-length ATM cell, but a variable-length packet may be used.
[0129]
That is, the present invention has a ring topology at least logically, and can be widely applied to a network that transmits unit signals, a node that can be used in such a network, and a node connection method. it can.
[0130]
【The invention's effect】
As described above, according to the present invention, when operating a network in which the logical topology is a ring type and the physical topology is a serial daisy chain, each adjacent node is connected by a cable connection to a connector. The logical and physical topology of the network is maintained by changing the transmission path of the unit signal in the own node according to the connection state of the cable, so that each node is adjacent to itself on the network. You only need to respond to changes that occur in the location. Nodes other than the node adjacent to the change location on the daisy chain need only maintain the operation state before the change, and do not need to detect that the change has occurred.
[0131]
Since this change may include node addition, change, deletion, maintenance, failure occurrence, etc., the size of each node should be configured smaller in software and hardware than its function. Node and network maintenance and reliability are high.
[0132]
In addition, according to the daisy chain type network of the present invention, the maximum linear distance between any two nodes in the network is almost twice that of the prior art, and with the same number of nodes, the distance is longer than before. Since the terminal existing at the point can be accommodated in the network, effective use of resources and efficient operation of the network become possible.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a network and nodes according to a first embodiment.
FIG. 2 is a block diagram showing a configuration of a conventional network and nodes.
FIG. 3 is a block diagram showing a configuration for a node according to the first embodiment to detect a cable connection state;
FIG. 4 is a schematic diagram illustrating a state in which a node is added in the first embodiment.
FIG. 5 is a block diagram showing a configuration of a network and nodes according to a second embodiment.
FIG. 6 is a schematic diagram showing an operation change when adding a node in the first embodiment and the conventional network;
[Explanation of symbols]
10, 20, 40 ... Network, 11-13, 21-23, 33, 41-43 ... Node, 24A-24C ... Left connector, 25A-25C ... Right connector, 26A-26C ... Switch part, 27A-27C ... Reception , 28A-28C ... transmission unit, 29A-29C ... packet processing unit, 30A-30C ... cable, 31A, 31B ... connection notification transmission line, CON1, CON2 ... connect terminal, TL ... transmission line for communication in the left direction, PR ... Communication transmission line in the right direction.

Claims (2)

In a method for connecting nodes of a network having at least logically a ring topology and transmitting unit signals,
The network, each node that is at least physically make up of a daisy chain form so as to form a serial topology is
A transmission line that transmits a unit signal between nodes and a cable that forms a connection notification transmission line that conveys the connection between the nodes, and a ground terminal and a pull-up resistor that are connected to the cable as the connection notification transmission line A pair of left and right connectors having a connection terminal and a voltage detector ;
The right and left pair of connectors connecting the set and alternatively selecting switch unit,
The voltage detection unit recognizes the connection between the node and the connection destination node connected through the cable by detecting the voltage of the pull-up resistor connection terminal;
The node selection method , wherein the switch unit selects the cable-side connector to be connected to the pull-up resistor connection terminal whose voltage is detected by the voltage detection unit . The node connection method according to claim 1, wherein:
Before Symbol each node is at least,
The switch part is
Of the pair of left and right connectors, the connection destination of the input terminal and the output terminal of one of the left and right connectors is determined according to the connection state of the cable to the left and right connectors. from among the connector output terminal or the input terminal of, alternatively selected,
A logical processing unit that receives a unit signal transmitted in the network from the receiving unit, performs at least logical processing, and then sends the unit signal to the transmitting unit, so that logically as described above. Is a ring type, but physically forms a series topology,
Further, when selecting a connection received from the output terminal of one of the left and right connectors and supplied to the input terminal of the other left and right connector, the switch unit amplifies the unit signal using an amplifying unit. Connection method.

Images