JP2004260493A

JP2004260493A - Loudspeaker call system for multiple dwelling house

Info

Publication number: JP2004260493A
Application number: JP2003048281A
Authority: JP
Inventors: Minoru Fukushima; 実福島; Satoshi Sugimoto; 敏杉本; Hiroaki Takeyama; 博昭竹山; Yasuhisa Ihira; 靖久井平; 恵一 ▲吉▼田; Keiichi Yoshida; Akihiro Kikuchi; 彰洋菊池; Akira Terasawa; 章寺澤
Original assignee: Matsushita Electric Works Ltd
Current assignee: Panasonic Electric Works Co Ltd
Priority date: 2003-02-25
Filing date: 2003-02-25
Publication date: 2004-09-16
Anticipated expiration: 2023-02-25
Also published as: JP4079008B2

Abstract

PROBLEM TO BE SOLVED: To reduce a sense of call interruption just after the start of a call while preventing unpleasant echo and howling and to shorten a period of time when the call is started until the call reaches a two-way simultaneous vall. SOLUTION: A nonvolatile memory 150 for storing initial values Lt(0)<SB>1</SB>to Lt(0)<SB>4</SB>optimum to every kind of opposite parties is provided in a master unit 1A and an auxiliary subsidiary master unit 1B. The initial values Lt(0)<SB>3</SB>, Lt(0)<SB>4</SB>for other call terminals (the auxiliary master unit 1B or the master unit 1A) for carrying out an extension call and a handset type outside house call terminal (a warning monitor panel 80) and the initial values Lt(0)<SB>1</SB>, Lt(0)<SB>2</SB>for a loudspeaker type outside house call terminal (an intercom slave unit 50 and a lobby interphone 60) can respectively separately be set. Thus, while preventing unpleasant echo or howling, a sense of call interruption just after the start of the call can be reduced and the period of time when the call is started until the call reaches the two-way simultaneous call enabled state can be shortened. COPYRIGHT: (C)2004,JPO&NCIPI

Description

【０００１】
【発明の属する技術分野】
本発明は、マンション等の集合住宅に用いられる集合住宅用拡声通話システムに関するものである。
【０００２】
【従来の技術】
近年、マンションなどの集合住宅では、集合住宅の共同玄関に設置されるロビーインターホンと、管理人室や防災監視室などに設置される監視盤と、各住戸内に設置される親機と、同じく各住戸内に設置される副親機と、各住戸の戸外（玄関先）に設置されるドアホン子器とを備え、各住戸の親機又は副親機（これらを総称して「宅内通話端末」と呼ぶ）とロビーインターホンやドアホン子器あるいは監視盤（これらを総称して「宅外通話端末」と呼ぶ）との間の通話、並びに親機と副親機との間の通話（内線通話）を可能とした集合住宅用拡声通話システムが利用されている。
【０００３】
このような集合住宅用拡声通話システムに用いられる親機や副親機として、本出願人は図１４に示すような拡声通話装置を既に提案している（特許文献１参照）。以下、この従来の拡声通話装置について説明する。
【０００４】
上記従来の拡声通話装置は、マイクロホン１００、スピーカ１０１、２線−４線変換回路１０３、マイクロホンアンプＧ１、回線（２線の伝送路）への送話信号を増幅する回線出力アンプＧ２、回線からの受話信号を増幅する回線入力アンプＧ３、スピーカアンプＧ４、送話音量調整用増幅器Ｇ５、受話音量調整用増幅器Ｇ６、並びに第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂと音声スイッチ１２０からなり、ハウリングやエコーの抑制を含む拡声通話のための処理を行う通話処理手段とで構成される。
【０００５】
第１のエコーキャンセラ１１０Ａは適応フィルタ１１１Ａと減算器１１２Ａからなる従来周知の構成を有し、スピーカ１０１−マイクロホン１００間の音響結合により形成される帰還経路（音響エコー経路）Ｈ_ＡＣのインパルス応答を適応フィルタ１１１Ａにより適応的に同定し、参照信号（スピーカアンプＧ４への入力信号）から推定したエコー成分（音響エコー）を減算器１１２ＡによりマイクロホンアンプＧ１の出力信号から減算することでエコー成分を抑圧するものである。また、第２のエコーキャンセラ１１０Ｂも適応フィルタ１１１Ｂと減算器１１２Ｂからなる従来周知の構成を有し、２線−４線変換回路１０３と伝送路との間のインピーダンスの不整合による反射および相手側の拡声通話端末（例えば、ドアホン子器）におけるスピーカ−マイクロホン間の音響結合とにより形成される帰還経路（回線エコー経路）Ｈ_ＬＩＮのインパルス応答を適応フィルタ１１１Ｂにより適応的に同定し、参照信号（回線出力アンプＧ２への入力信号、すなわち送話信号）から推定したエコー成分（回線エコー）を減算器１１２Ｂにより受話信号から減算することでエコー成分を抑圧するものである。
【０００６】
また、第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂに挟まれた送話信号及び受話信号の信号経路上に設けられる送話音量調整用増幅器Ｇ５及び受話音量調整用増幅器Ｇ６の間に音声スイッチ１２０が設けてある。この音声スイッチ１２０は、送話側の信号経路に損失を挿入する送話側損失挿入手段１２１と、受話側の信号経路に損失を挿入する受話側損失挿入手段１２２と、送話側及び受話側の各損失挿入手段１２１，１２２から挿入する損失量を制御する挿入損失量制御手段１２３とを具備する。挿入損失量制御手段１２３は、受話側損失挿入手段１２２の出力点Ｒｏｕｔから音響エコー経路Ｈ_ＡＣを介して送話側損失挿入手段１２１の入力点Ｔｉｎへ帰還する経路（以下、「音響側帰還経路」という）の音響側帰還利得αを推定するとともに、送話側損失挿入手段１２１の出力点Ｔｏｕｔから回線エコー経路Ｈ_ＬＩＮを介して受話側損失挿入手段１２２の入力点Ｒｉｎへ帰還する経路（以下、「回線側帰還経路」という）の回線側帰還利得βを推定し、音響側及び回線側の各帰還利得α，βの推定値α’，β’に基づいて閉ループに挿入すべき損失量の総和（送話側損失挿入手段１２１の挿入損失量と受話側損失挿入手段１２２の挿入損失量の和）を算出する総損失量算出部１２４と、送話信号及び受話信号を監視して通話状態を推定し、この推定結果と総損失量算出部１２４の算出値に応じて送話側損失挿入手段１２１及び受話側損失挿入手段１２２の各挿入損失量の配分を決定する挿入損失量分配処理部１２５とからなる。なお、第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂと音声スイッチ１２０からなる通話処理手段は、例えばＤＳＰ（ＤｉｇｉｔａｌＳｉｇｎａｌＰｒｏｃｅｓｓｏｒ）のハードウェアに専用のソフトウェアを搭載することで実現される。
【０００７】
総損失量算出部１２４では、整流平滑器や低域通過フィルタ等を用いて送話側損失挿入手段１２１の入力信号の短時間における時間平均パワーを推定し、同じく整流平滑器や低域通過フィルタ等を用いて受話側損失挿入手段１２２の出力信号の短時間における時間平均パワーを推定し、音響側帰還経路Ｈ_ＡＣにて想定される最大遅延時間において受話側損失挿入手段１２２の出力信号の時間平均パワーの推定値の最小値を求め、この最小値で送話側損失挿入手段１２１の入力信号の時間平均パワーの推定値を除算した値を音響側帰還利得αの推定値α’とするとともに、整流平滑器や低域通過フィルタ等を用いて受話側損失挿入手段１２２の入力信号の短時間における時間平均パワーを推定し、同じく整流平滑器や低域通過フィルタ等を用いて送話側損失挿入手段１２１の出力信号の短時間における時間平均パワーを推定し、回線側帰還経路Ｈ_ＬＩＮにて想定される最大遅延時間において送話側損失挿入手段１２１の出力信号の時間平均パワーの推定値の最小値を求め、この最小値で受話側損失挿入手段１２２の入力信号の時間平均パワーの推定値を除算した値を回線側帰還利得βの推定値β’とする。そして、総損失量算出部１２４は音響側帰還利得α及び回線側帰還利得βの各推定値α’，β’から所望の利得余裕ＭＧを得るために必要な総損失量Ｌｔを算出し、その値Ｌｔを挿入損失量分配処理部１２５に出力する。
【０００８】
挿入損失量分配処理部１２５では、送話側損失挿入手段１２１の入出力信号及び受話側損失挿入手段１２２の入出力信号を監視し、これらの信号のパワーレベルの大小関係並びに音声信号の有無などの情報から通話状態（受話状態、送話状態等）を判定するとともに、判定された通話状態に応じた割合で総損失量Ｌｔを送話側損失挿入手段１２１と受話側損失挿入手段１２２に分配するように各損失挿入手段１２１，１２２の挿入損失量を調整する。
【０００９】
ところで、この従来例における総損失量算出部１２４は、上述のように各帰還利得α，βの推定値α’，β’に基づいて閉ループに挿入すべき損失量の総和を算出して適応更新する更新モード、並びに総損失量を所定の初期値に固定する固定モードの２つの動作モードを有し、相手側通話端末との通話開始から第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂが充分に収束するまでの期間には固定モードで動作するとともに第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂが充分に収束した後の期間には更新モードで動作する。すなわち、総損失量算出部１２４では音響側帰還利得α及び回線側帰還利得βの推定値α’，β’がともに通話開始から所定時間（数百ミリ秒）以上継続して所定の閾値ε（例えば、通話開始時における各推定値α’，β’に対して１０ｄＢ〜１５ｄＢ小さい値）を下回った時点で第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂが充分に収束したものとみなし、上記時点以前には総損失量を初期値に固定する固定モードで動作し、上記時点以降には各推定値α’，β’に基づいて総損失量を適応更新する更新モードに動作モードを切り換える。なお、固定モードにおける総損失量の初期値は更新モードにおいて随時更新される総損失量よりも充分に大きな値に設定される。
【００１０】
而して、通話開始直後の第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂが充分に収束していない状態においては、固定モードで動作する総損失量算出部１２４によって充分に大きな値に設定される初期値の総損失量が閉ループに挿入されるため、不快なエコー（音響エコー並びに回線エコー）やハウリングの発生を抑制して安定した半二重通話を実現することができる。また、通話開始から時間が経過して第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂが充分に収束した状態においては、総損失量算出部１２４の動作モードが固定モードから更新モードに切り換わって閉ループに挿入する総損失量が初期値よりも充分に低い値に減少するため、双方向の同時通話が実現できるものである。
【００１１】
次に、更新モードにおける総損失量算出部１２４の具体的な動作を図１５のフローチャートを参照して説明する。
【００１２】
総損失量算出部１２４は、固定モードから更新モードに移行した時点から所定のサンプリング周期で音響側帰還利得α並びに回線側帰還利得βの推定処理を実行してその推定値α’（ｎ），β’（ｎ）を算出し（ステップ１）、これら２つの推定値α’（ｎ），β’（ｎ）の積と利得余裕ＭＧとから、閉ループの利得余裕をＭＧ［ｄＢ］に保つために必要とされる総損失量所望値Ｌｒ（ｎ）を下式により算出する（ステップ２）。
【００１３】
Ｌｒ（ｎ）＝２０ｌｏｇ｜α’（ｎ）・β’（ｎ）｜＋ＭＧ［ｄＢ］
なお、α’（ｎ），β’（ｎ），Ｌｒ（ｎ）はそれぞれ更新モード移行時点からｎ回目のサンプリングによって算出された帰還利得の推定値並びに総損失量所望値を示す。さらに、総損失量算出部１２４は上式から算出したｎ回目の総損失量所望値Ｌｒ（ｎ）と、前回（ｎ−１回目）の総損失量Ｌｔ（ｎ−１）、すなわち前回の処理で決定されて実際に挿入された総損失量に対して今回算出した総損失量所望値Ｌｒ（ｎ）が大きい場合、前回の総損失量Ｌｔ（ｎ−１）に微少な増加量Δｉ［ｄＢ］を加算した値を今回の総損失量Ｌｔ（ｎ）＝Ｌｔ（ｎ−１）＋Δｉとし（ステップ３、ステップ４）、前回の総損失量Ｌｔ（ｎ−１）に対して今回算出した総損失量所望値Ｌｒ（ｎ）が小さい場合、前回の総損失量Ｌｔ（ｎ−１）から微少な減少量Δｄ［ｄＢ］を減算した値を今回の総損失量Ｌｔ（ｎ）＝Ｌｔ（ｎ−１）−Δｄとする（ステップ５、ステップ６）。
【００１４】
このように総損失量算出部１２４による総損失量の増減をΔｉ又はΔｄの微少な値に抑えることにより、相手側通話端末との通話開始直後のように第１及び第２のエコーキャンセラ１１０Ａ，１１０Ｂが収束に向かって活発に係数を更新しているために音響側帰還利得α及び回線側帰還利得βの変化が激しい状態においても、聴感上の違和感をなくすことができる。
【００１５】
【特許文献１】
特開２００２−３５９５８０号公報（段落００２７−段落００３６、第１図及び第２図）
【００１６】
【発明が解決しようとする課題】
ところで、上記帰還利得（回線側帰還利得β）は相手側通話端末の種類によって異なっており、通常、屋外に設置されるドアホン子器や、通話路の長さが他の通話端末に比較してかなり長くなるロビーインターホンに対する帰還利得に比べて、通話処理手段を具備している親機又は副親機や、ハンドセット型の宅外通話端末である監視盤に対する帰還利得の方がかなり小さくなっている。
【００１７】
しかしながら上記従来の拡声通話装置では、通話開始時の固定モードにおける総損失量の初期値が何れの通話端末に対しても一律に同じ値に設定されるため、ドアホン子器やロビーインターホンに対してはエコーやハウリングの抑制に必要十分な値であっても親機又は副親機や監視盤に対しては必要以上に大きな値となってしまう。このため、宅内通話端末同士の内線通話や宅内通話端末とハンドセット型の宅外通話端末（監視盤）との通話において、通話開始直後に相手の話声が途切れて聞こえる感覚（これを「切断感」と呼ぶ）が強くなったり、固定モードから更新モードへ移行した後に総損失量が初期値よりも十分に小さい適正値に設定されて双方向の同時通話が可能となるまでに必要以上に長い時間がかかってしまう虞があった。
【００１８】
本発明は上記事情に鑑みて為されたものであり、その目的は、不快なエコーやハウリングを防止しつつ通話開始直後の切断感が軽減できるとともに通話開始から双方向の同時通話が可能な状態に至るまでの時間が短縮できる集合住宅用拡声通話システムを提供することにある。
【００１９】
【課題を解決するための手段】
請求項１の発明は、上記目的を達成するために、集合住宅における各住戸の外に設置される複数の宅外通話端末と、各住戸内に設置されて前記宅外通話端末との間で通話を行う複数の宅内通話端末とを備え、前記宅外通話端末にはマイクロホンとスピーカを具備して拡声通話を行う拡声型の宅外通話端末と、ハンドセットを具備したハンドセット型の宅外通話端末とが含まれ、前記宅内通話端末には、マイクロホン、スピーカ並びにハウリングやエコーの抑制を含む拡声通話のための処理を行う通話処理手段を具備し前記宅外通話端末との間に形成される通話路を介して拡声通話を行う親機と、マイクロホン、スピーカ並びにハウリングやエコーの抑制を含む拡声通話のための処理を行う通話処理手段を具備し前記親機を介して前記宅外通話端末と拡声通話を行うとともに当該親機との間で内線通話を行う１乃至複数の副親機とが含まれる集合住宅用拡声通話システムにおいて、前記宅内通話端末の通話処理手段は、マイクロホンとスピーカの音響結合によって生じる音響エコーを抑圧する第１のエコーキャンセラと、相手側の通話端末における音響結合又は信号の回り込みによって生じる回線エコーを抑圧する第２のエコーキャンセラと、第１及び第２のエコーキャンセラの間に設けられ、前記音響エコー経路並びに回線エコー経路により形成される閉ループの一巡利得を低減してハウリングを抑制する音声スイッチとを有し、第１及び第２のエコーキャンセラは、前記エコー経路の特性を適応的に同定する適応フィルタを具備し、前記音声スイッチは、送話側の信号経路に損失を挿入する送話側損失挿入手段と、受話側の信号経路に損失を挿入する受話側損失挿入手段と、送話側及び受話側の前記各損失挿入手段から挿入する損失量を制御する挿入損失量制御手段とを具備し、該挿入損失量制御手段は、前記受話側損失挿入手段の出力点から前記音響エコー経路を介して前記送話側損失挿入手段の入力点へ帰還する経路の音響側帰還利得を推定するとともに、前記送話側損失挿入手段の出力点から前記回線エコー経路を介して前記受話側損失挿入手段の入力点へ帰還する経路の回線側帰還利得を推定し、音響側及び回線側の各帰還利得の推定値に基づいて前記閉ループに挿入すべき損失量の総和を算出する総損失量算出部と、送話信号及び受話信号を監視して通話状態を推定し、この推定結果と前記総損失量算出部の算出値に応じて前記送話側損失挿入手段及び受話側挿入損失手段の各挿入損失量の配分を決定する挿入損失量分配処理部とからなり、前記総損失量算出部は、前記各帰還利得の推定値に基づいて前記閉ループに挿入すべき損失量の総和を算出して適応更新する更新モード、並びに総損失量を所定の初期値に固定する固定モードの２つの動作モードを有し、相手側通話端末との通話開始から第１及び第２のエコーキャンセラが充分に収束するまでの期間には固定モードで動作するとともに第１及び第２のエコーキャンセラが充分に収束した後の期間には更新モードで動作し、更新モードにおいては音響側及び回線側の各帰還利得の推定値と利得余裕値とから総損失量所望値を求め、総損失量所望値が更新前の総損失量よりも大きければ総損失量を所定の増加量だけ増加した値に更新し、総損失量所望値が更新前の総損失量よりも小さければ総損失量を所定の減少量だけ減少させた値に更新するとともに、総損失量所望値が更新前の総損失量と等しければ総損失量を更新しないことで総損失量を算出しないものであり、相手側通話端末に対応した前記初期値を個別に記憶する不揮発性の記憶手段と、相手側通話端末との通話開始前に前記記憶手段に記憶した初期値を前記総損失量算出部に設定する初期値設定手段とを前記宅内通話端末に具備し、内線通話を行う他の宅内通話端末に対応した前記初期値並びにハンドセット型の宅外通話端末に対応した前記初期値を拡声型の宅外通話端末に対応した前記初期値よりも小さい値に設定したことを特徴とする。
【００２０】
請求項２の発明は、請求項１の発明において、前記総損失量算出部が更新モードで動作しているときにとりうる前記総損失量の下限値を、前記宅内通話端末の種別と宅外通話端末の種別に応じて個別に設定する設定手段を前記宅内通話端末に具備したことを特徴とする。
【００２１】
請求項３の発明は、請求項１又は２の発明において、相手側通話端末が宅内通話端末である場合の前記増加量並びに減少量を相手側通話端末が宅外通話端末である場合の前記増加量並びに減少量よりも大きくしたことを特徴とする。
【００２２】
請求項４の発明は、請求項１又は２又は３の発明において、前記副親機は宅内に設置された他の副親機との間で内線通話を行うものであって、通話路のインピーダンスを整合するマッチング手段と、前記親機と副親機又は前記副親機間に形成される通話路の見かけ上のインピーダンスを宅内通話端末と少なくとも何れか一つの宅外通話端末の間に形成される通話路の見かけ上のインピーダンスに略等しくするインピーダンス調整手段とを前記宅内通話端末に具備したことを特徴とする。
【００２３】
【発明の実施の形態】
（実施形態１）
以下、本発明を住宅情報盤システムに適用した実施形態により詳細に説明する。
【００２４】
本実施形態の住宅情報盤システムは、図２に示すように集合住宅の共同玄関に設置されるロビーインターホン６０と、集合住宅の管理室等に設置される警報監視盤８０と、各住戸内に設置される住宅情報盤親機（以下、「親機」と略す）１Ａ並びに住宅情報盤副親機（以下、「副親機」と略す）１Ｂと、各住戸の戸外に設置されるドアホン子器５０とを備える。また、管理人室等には警報監視盤８０とともに非常電源９０及び放送設備９１が設置され、それぞれ非常電源線Ｌ２及び音声出力線Ｌ３を介して各住戸の親機１Ａに接続される。そして、警報監視盤８０と各住戸の親機１Ａが、通話音声や警報音等の音声信号とロビーインターホン６０が具備するテレビカメラ６８で撮像した映像信号を多重伝送する通信線Ｌａ、呼出や警報等の制御信号を多重伝送する制御線Ｌｂによって接続される。ここで、通信線Ｌａ及び制御線Ｌｂを合わせて情報幹線Ｌ１と呼ぶことにする。なお、本実施形態では映像信号を音声信号と多重伝送する構成を例示するが、映像信号を独立した映像伝送線で伝送する構成としても良い。
【００２５】
図４は本実施形態におけるロビーインターホン６０並びに警報監視盤８０の構成を示すブロック図である。ロビーインターホン６０は、ＣＰＵを主構成要素とする信号処理部６１と、テンキーや呼出釦等を具備する操作部６２と、表示ランプや液晶ディスプレイを具備する表示部６３と、制御線Ｌｂを介して警報監視盤８０との間で制御信号を送受信する伝送送受信部６４と、マイクロホン６５並びにスピーカ６６と、マイクロホン６５並びにスピーカ６６を用いて親機１Ａや警報監視盤８０との間で拡声通話を行うための通話制御部６７と、ＣＣＤのような撮像素子を有するテレビカメラ６８と、テレビカメラ６８の出力を信号処理して映像信号を出力する映像信号処理部６９と、通話制御部６７から出力される音声信号と映像信号処理部６９から出力される映像信号を多重化して通信線Ｌａに送出するとともに通信線Ｌａにより送られてくる音声信号を通話制御部６７に出力する入出力処理部７０とを備えている。
【００２６】
一方、警報監視盤８０は、ＣＰＵを主構成要素とする信号処理部８１と、テンキーや呼出釦等を具備する操作部８２と、表示ランプや液晶ディスプレイを具備する表示部８３と、制御線Ｌｂを介して各親機１Ａ、副親機１Ｂ、ロビーインターホン６０との間で制御信号を送受信する伝送送受信部８４と、送受話器（ハンドセット）８５と、送受話器８５を用いて親機１Ａ、副親機１Ｂ、ロビーインターホン６０との間で同時通話を行うための通話制御部８７とを備えている。
【００２７】
図３は本実施形態における親機１Ａ並びにドアホン子器５０の構成を示すブロック図であり、通話線Ｌｃにより親機１Ａとドアホン子器５０とが接続されている。ドアホン子器５０は、マイクロホン５１並びにスピーカ５２と、２線−４線変換部５３と、ＣＣＤのような撮像素子を有するテレビカメラ５４と、テレビカメラ５４の出力を信号処理して映像信号を出力する映像信号処理部５５と、２線−４線変換部５３から出力される音声信号と映像信号処理部５５から出力される映像信号を多重化して通話線Ｌｃに送出するとともに通話線Ｌｃにより送られてくる音声信号を２線−４線変換部５３に出力する入出力処理部５６と、呼出釦５７ａ（図２参照）が操作されたときに通話線Ｌｃの電圧レベルを変化させることで呼出信号を送出する呼出信号送出部５７とを備えている。
【００２８】
親機１Ａは、ＣＰＵを主構成要素とする主制御部２と、応答釦３ａ等を具備する操作部３と、ＥＥＰＲＯＭ等の不揮発性メモリからなる記憶部４と、制御線Ｌｂを介して警報監視盤８０との間で制御信号を送受信する伝送送受信部５と、火災感知器Ｓ１、ガス／ＣＯ複合センサＳ２、防犯外部セットスイッチＳＷ、防犯スイッチＳＷ１、コールスイッチＳＷ２、非常スイッチＳＷ３等の各種安全装置が接続され、火災感知器Ｓ１の発報信号やガス／ＣＯ複合センサＳ２からのガス漏れ信号あるいはコールスイッチＳＷ２の操作信号等を監視する信号監視部６と、通話線Ｌｃの電圧レベル変化を監視してドアホン子器５０からの呼出信号を検出する呼出検出部７と、商用交流電源ＡＣ又は非常電源９０から電源供給を受けて親機１Ａ並びに副親機１Ｂの動作電源を作成する電源部８と、電源部８への電源供給元を商用交流電源ＡＣと非常電源９０に切り換える電源切換部９とを備える。
【００２９】
また親機１Ａは、マイクロホン１０及びスピーカ１１と、ＣＲＴや液晶ディスプレイなどからなるモニタ部１３と、ドアホン子器５０又はロビーインターホン６０からの映像信号を信号処理してモニタ部１３に映像を表示させるとともに主制御部２から与えられる表示制御信号に応じて所定のシンボルを表示（スーパーインポーズ）させる映像表示部１４と、ドアホン子器５０又はロビーインターホン６０から多重化されて伝送される音声信号と映像信号を分離する信号分離部１６と、２線−４線変換部１５と、ドアホン子器５０、警報監視盤８０並びに副親機１Ｂとの間で拡声同時通話を実現するための処理を行うとともにロビーインターホン６０との間で拡声同時通話を実現するための処理を行う通話処理部１７と、呼出音やバックトーンあるいは警報メッセージや警報音等を生成する音声生成部１８と、マイクロホン１０の出力信号（送話信号）を増幅するマイクロホンアンプ１９と、マイクロホンアンプ１９で増幅されたアナログの送話信号をデジタル信号に変換する第１Ａ／Ｄ変換部２０と、通話処理部１７から出力されるデジタルの送話信号をアナログ信号に変換する第１Ｄ／Ａ変換部２１と、第１Ｄ／Ａ変換部２１で変換されたアナログの送話信号を増幅する送話信号増幅器２２と、信号分離部１６で分離された音声信号（受話信号）を増幅する受話信号増幅器２３と、受話信号増幅器２３で増幅されたアナログの受話信号をデジタル信号に変換する第２Ａ／Ｄ変換部２４と、通話処理部１７から出力されるデジタルの受話信号をアナログ信号に変換する第２Ｄ／Ａ変換部２５と、第２Ｄ／Ａ変換部２５で変換されたアナログの受話信号を増幅する第１スピーカアンプ２６と、音声生成部１８で生成された呼出音等のデジタルの音声信号をアナログ信号に変換する第３Ｄ／Ａ変換部２７と、第３Ｄ／Ａ変換部２７で変換されたアナログの音声信号を増幅する第２スピーカアンプ２８と、第１スピーカアンプ２６で増幅された受話信号と第２スピーカアンプ２８で増幅された呼出音等の音声信号を多重化する第１アナログ信号多重部２９と、スピーカ１１の入力端子Ｔｓ１を第１アナログ信号多重部２９の出力端子Ｔｓ２と音声出力線Ｌ３が接続された接続端子Ｔｓ３とに切り換える一斉放送切換部３０と、音声生成部１８で生成された呼出音等のデジタルの音声信号をアナログ信号に変換する第４Ｄ／Ａ変換部３１と、第４Ｄ／Ａ変換部３１で変換されたアナログの音声信号を増幅する音声信号増幅器３２と、送話信号増幅器２２で増幅された送話信号と音声信号増幅器３２で増幅された呼出音等の音声信号を多重化して２線−４線変換部１５に出力する第２アナログ信号多重部３３と、信号分離部１６の入出力端子Ｔａ１をドアホン子器５０からの通話線Ｌｃが接続された端子Ｔａ２と通信線Ｌａが接続された端子Ｔａ３に切り換える通話切換部３４とを備える。なお、本実施形態では通話処理部１７並びに音声生成部１８をデジタル・シグナル・プロセッサ（ＤＳＰ）２００で構成し、ＤＳＰ２００のハードウェアを専用のソフトウェアで制御することによって後述する通話処理部１７及び音声生成部１８の機能を実現している。また、映像表示部１４の出力端には映像信号を出力するためのコネクタＣＮが接続されており、上記コネクタＣＮに接続することでテレビ受像機やパーソナルコンピュータのモニタ装置の画面にモニタ部１３と同じ映像を映し出すことができる。
【００３０】
一方、副親機１Ｂは、図５に示すように親機１Ａとほぼ同一の構成要素を備えているので、同一の構成要素には同一の符号を付して説明を省略する。
【００３１】
親機１Ａ（副親機１Ｂも共通）の通話処理部１７は、図６に示すように従来例の拡声通話装置と共通の構成を有するものであって、マイクロホン１０とスピーカ１１の音響結合によって生じる音響エコーを抑圧する音響エコーキャンセラ３５と、ドアホン子器５０、ロビーインターホン６０又は警報監視盤８０における音響結合又は通信線Ｌａを介した信号の回り込みによって生じる回線エコーを抑圧する回線エコーキャンセラ３６と、音響エコーキャンセラ３５及び回線エコーキャンセラ３６の間に設けられ、音響エコー経路並びに回線エコー経路により形成される閉ループ（エコーパス）の一巡利得を低減してハウリングを抑制する音声スイッチ３７と、音響エコーキャンセラ３５及び回線エコーキャンセラ３６に挟まれた送話信号並びに受話信号の信号経路上に各々設けられる送話側増幅器３８及び受話側増幅器３９とを有する。
【００３２】
音響エコーキャンセラ３５は、従来例における第１のエコーキャンセラと同様に適応フィルタ３５ａと減算器３５ｂからなり、スピーカ１１−マイクロホン１０間の音響結合により形成される帰還経路（音響エコー経路）のインパルス応答を適応フィルタ３５ａにより適応的に同定し、参照信号（音声スイッチ３７から出力される受話信号）から推定したエコー成分（音響エコー）を減算器３５ｂにより第１Ａ／Ｄ変換部２０の出力信号（送話信号）から減算することでエコー成分を相殺して抑圧するものである。また、回線エコーキャンセラ３６も従来例における第２のエコーキャンセラと共通の構成を有しており、適応フィルタ３６ａと減算器３６ｂからなり、２線−４線変換部１５と通信線Ｌａ又は通話線Ｌｃとの間のインピーダンスの不整合による反射およびドアホン子器５０等におけるスピーカ−マイクロホン間の音響結合とにより形成される帰還経路（回線エコー経路）のインパルス応答を適応フィルタ３６ａにより適応的に同定し、参照信号（第１Ｄ／Ａ変換部２１へ出力される送話信号）から推定したエコー成分（回線エコー）を減算器３６ｂにより受話信号から減算することでエコー成分を相殺して抑圧するものである。
【００３３】
音声スイッチ３７は、送話側の信号経路に損失を挿入する送話側損失挿入手段としての送話信号減衰器３７ａと、受話側の信号経路に損失を挿入する受話側損失挿入手段としての受話信号減衰器３７ｂと、送話側及び受話側の各減衰器３７ａ，３７ｂから送話側及び受話側の信号経路へそれぞれ挿入する損失量（各減衰器３７ａ，３７ｂの利得）を制御する挿入損失量制御部３７ｃとを具備する。さらに挿入損失量制御部３７ｃは、受話信号減衰器３７ｂの出力点から音響エコー経路を介して送話信号減衰器３７ａの入力点へ帰還する経路（以下、「音響側帰還経路」という）の音響側帰還利得を推定するとともに、送話信号減衰器３７ａの出力点から回線エコー経路を介して受話信号減衰器３７ｂの入力点へ帰還する経路（以下、「回線側帰還経路」という）の回線側帰還利得を推定し、音響側及び回線側の各帰還利得の推定値α’，β’に基づいて閉ループに挿入すべき損失量の総和（送話信号減衰器３７ａの挿入損失量と受話信号減衰器３７ｂの挿入損失量の和）を算出する総損失量算出部３７ｄと、送話信号及び受話信号を監視して通話状態を推定し、この推定結果と総損失量算出部３７ｄの算出値に応じて送話信号減衰器３７ａ及び受話信号減衰器３７ｂの各挿入損失量の配分、すなわち利得を決定する挿入損失量分配処理部３７ｅとを具備する。
【００３４】
総損失量算出部３７ｄでは、整流平滑器や低域通過フィルタ等を用いて送話信号減衰器３７ａの入力信号の短時間における時間平均パワーを推定し、同じく整流平滑器や低域通過フィルタ等を用いて受話信号減衰器３７ｂの出力信号の短時間における時間平均パワーを推定し、音響側帰還経路にて想定される最大遅延時間において受話信号減衰器３７ｂの出力信号の時間平均パワーの推定値の最小値を求め、この最小値で送話信号減衰器３７ａの入力信号の時間平均パワーの推定値を除算した値を音響側帰還利得の推定値α’とするとともに、整流平滑器や低域通過フィルタ等を用いて受話信号減衰器３７ｂの入力信号の短時間における時間平均パワーを推定し、同じく整流平滑器や低域通過フィルタ等を用いて送話信号減衰器３７ａの出力信号の短時間における時間平均パワーを推定し、回線側帰還経路にて想定される最大遅延時間において送話信号減衰器３７ａの出力信号の時間平均パワーの推定値の最小値を求め、この最小値で受話信号減衰器３７ｂの入力信号の時間平均パワーの推定値を除算した値を回線側帰還利得の推定値β’とする。そして、総損失量算出部３７ｄは音響側帰還利得及び回線側帰還利得の各推定値α’，β’から所望の利得余裕を得るために必要な総損失量を算出し、その値を挿入損失量分配処理部３７ｅに出力する。
【００３５】
挿入損失量分配処理部３７ｅでは、送話信号減衰器３７ａの入出力信号及び受話信号減衰器３７ｂの入出力信号を監視し、これらの信号のパワーレベルの大小関係並びに音声信号の有無などの情報から通話状態（受話状態、送話状態等）を判定するとともに、判定された通話状態に応じた割合で総損失量を送話信号減衰器３７ａと受話信号減衰器３７ｂに分配するように各減衰器３７ａ，３９の挿入損失量（利得）を調整する。
【００３６】
また通話処理部１７は、送話信号が単一周波数のトーン信号か否かを検出する送話信号シングルトーン検出器１４０、受話信号が単一周波数のトーン信号か否かを検出する受話信号シングルトーン検出器１４１、これら２つのシングルトーン検出器１４０，１４１の検出結果に基づいてハウリングの有無を検出するハウリング検出部１４２を具備する。
【００３７】
２つのシングルトーン検出器１４０，１４１は、例えば送話信号あるいは受話信号の高速フーリエ変換を利用したり、複数の帯域分割フィルタを用いて送話信号あるいは受話信号のパワーを比較することでトーン信号か否かを検出するものであるが、何れの検出方式についても従来周知の技術を用いて実現可能であるから詳しい説明は省略する。
【００３８】
ハウリング検出部１４２では、シングルトーン検出器１４０，１４１がともに単一周波数のトーン信号を検出する状態が所定時間（例えば、２００〜３００ｍｓ程度）だけ継続した場合にハウリングが生じているものとして音響エコーキャンセラ３５及び回線エコーキャンセラ３６にハウリング検出信号を出力し、これ以外の場合にはハウリング検出信号を出力しない。
【００３９】
次に、図７のフローチャートを参照して本実施形態の動作説明を行う。通話が開始されると、音声スイッチ３７と２つのエコーキャンセラ３５，３６において初期化処理が行われ（ステップ１）、音声スイッチ３７では、初期化処理によって総損失量算出部３７ｄの動作モードが固定モードとなる（ステップ２）。総損失量算出部３７ｄではハウリング検出部１４２の出力状態を監視しており（ステップ３）、固定モードで動作中にハウリング検出部１４２からハウリング検出信号が出力されると上記初期化処理に戻り、音声スイッチ３７と２つのエコーキャンセラ３５，３６において初期化処理が行われ（ステップ１）、総損失量算出部３７ｄも全ての変数を初期化した後に再度固定モードで動作する（ステップ２）。
【００４０】
一方、ハウリング検出部１４２からハウリング検出信号が出力されない場合、総損失量算出部３７ｄにおいては、２つのエコーキャンセラ３５，３６が収束条件（例えば、帰還利得α，βの推定値α’，β’が所定時間継続して所定の閾値εを下回ることなど）を満足したか否かを判定する（ステップ４）。そして、少なくとも何れか一方のエコーキャンセラ３５又は３６が収束条件を満足していなければ、更新モードに移行せずに固定モードの動作を継続し（ステップ２）、２つのエコーキャンセラ３５，３６が何れも収束条件を満足すれば更新モードに移行する（ステップ５）。
【００４１】
そして、更新モードで動作中にハウリング検出部１４２からハウリング検出信号が出力されると上記初期化処理に戻り、音声スイッチ３７と２つのエコーキャンセラ３５，３６において初期化処理が行われ（ステップ１）、総損失量算出部３７ｄも全ての変数を初期化した後に再度固定モードで動作する（ステップ２）。
【００４２】
このように音声スイッチ３７においては、固定モード又は更新モードの何れの動作モードにおいても、通話中にハウリングが検出された際には速やかに２つのエコーキャンセラ３５，３６の適応フィルタ３５ａ，３６ａの係数を初期化するとともに、総損失量を初期値に戻して２つのエコーキャンセラ３５，３６が収束するまでは総損失量算出部３７ｄが固定モードで動作するため、通話系に形成される閉ループの一巡利得を抑制してハウリングに対する充分なマージンを得ることができ、通話中に万が一ハウリングが生じたとしても、直ちに（数百ｍｓ以内）ハウリングを抑制し、安定した通話を再開することができる。
【００４３】
次に本発明の要旨について説明する。更新モードにおける総損失量算出部３７ｄの動作は従来例と共通であるが（図１５のフローチャート参照）、本実施形態においては通話を開始する際に相手側の通話端末の種類、すなわち宅内通話端末（副親機１Ｂ又は親機１Ａ）並びにハンドセット型の宅外通話端末（警報監視盤８０）と、拡声型の宅外通話端末（ドアホン子器５０及びロビーインターホン６０）とで総損失量の初期値を異なる値、具体的には副親機１Ｂ（又は親機１Ａ）及び警報監視盤８０に対する初期値をドアホン子器５０及びロビーインターホン６０に対する初期値よりも小さい値（例えば、デシベル換算で２分の１以下）に設定している。
【００４４】
ここで、親機１Ａ（又は副親機１Ｂ）と拡声型の宅外通話端末（ドアホン子器５０又はロビーインターホン６０）との通話時における初期値に対して、宅内通話端末同士（親機１Ａと副親機１Ｂ又は副親機１Ｂ同士）の内線通話若しくは親機１Ａ（又は副親機１Ｂ）とハンドセット型の宅外通話端末（警報監視盤８０）との通話時における初期値を小さく設定してもエコーやハウリングが十分に防止できる理由を簡単に説明する。
【００４５】
図８は親機１Ａと拡声型の宅外通話端末（ドアホン子器５０又はロビーインターホン６０）との間に通話路が形成されている状態を示した概略構成図である。宅外通話端末はＤＳＰからなる通話処理部を具備しないため、親機１Ａにおける送話信号利得（マイクロホン１０から２線−４線変換部１５に入力される前までの利得）Ｔ１Ａは、宅外通話端末における送話信号利得（親機１Ａの送話信号が入力されてからスピーカ５２に達するまでの利得）Ｔ２よりも非常に大きく（Ｔ２≪Ｔ１Ａ）、同じく親機１Ａにおける受話信号利得（２線−４線変換部１５に出力されてからスピーカ１１に達するまでの利得）Ｒ１Ａは、宅外通話端末における受話信号利得（マイクロホン５１より出力された送話信号が親機１Ａの２線−４線変換部１５に入力される前までの利得）Ｒ２よりも非常に大きい（Ｒ２≪Ｒ１Ａ）。また、親機１Ａの２線−４線変換部１５におけるインピーダンス不整合による反射の帰還利得β１は、宅外通話端末のマイクロホンとスピーカの音響結合による帰還利得β２よりも小さい（β１＜β２）。
【００４６】
したがって、２線−４線変換部１５でのインピーダンス不整合による反射による閉ループＬＰ１の一巡利得Ｇ_ＬＰ１（＝α×Ｔ１Ａ×β１×Ｒ１Ａ）は、宅外通話端末の音響結合による閉ループＬＰ２の一巡利得Ｇ_ＬＰ２（＝α×Ｔ１Ａ×β２×Ｒ１Ａ）よりも小さくなる（Ｇ_ＬＰ１＜Ｇ_ＬＰ２）。
【００４７】
一方、図９は親機１Ａと宅内通話端末（副親機１Ｂ）との間に通話路が形成されている状態を示した概略構成図である。副親機１Ｂは親機１Ａと同様に音声スイッチ３７やエコーキャンセラ３５，３６を具備した通話処理部１７（ＤＳＰ２００）を備えているため、親機１Ａにおける送話信号利得Ｔ１Ａ’は、副親機１Ｂにおける受話信号利得Ｒ２’と略同一であり、同じく親機１Ａにおける受話信号利得Ｒ１Ａ’は、副親機１Ｂにおける送話信号利得Ｔ２’と略同一である。また、親機１Ａの２線−４線変換部１５におけるインピーダンス不整合による反射の帰還利得β１’は、副親機１Ｂのマイクロホン１０とスピーカ１１の音響結合による帰還利得β２’よりも大きい（β２’＜β１’）。
【００４８】
したがって、２線−４線変換部１５でのインピーダンス不整合による反射による閉ループＬＰ１’の一巡利得Ｇ_ＬＰ１’（＝α’×Ｔ１Ａ’×β１’×Ｒ１Ａ’）は、副親機１Ｂの音響結合による閉ループＬＰ２’の一巡利得Ｇ_ＬＰ２’（＝α’×Ｔ１Ａ’×β２’×Ｒ１Ａ’）よりも大きくなる（Ｇ_ＬＰ２’＜Ｇ_ＬＰ _１’）。ここで、親機１Ａの音響結合による帰還利得α，α’は相手の通話端末に関係なく一定であり（α＝α’）、親機１Ａにおける送話信号利得Ｔ１Ａ，Ｔ１Ａ’並びに受話信号利得Ｒ１Ａ，Ｒ１Ａ’は通話相手が副親機１Ｂの場合の方が小さくなる（Ｔ１Ａ’＜Ｔ１Ａ、Ｒ１Ａ’＜Ｒ１Ａ）。
【００４９】
よって、通話相手が副親機１Ｂの場合における閉ループＬＰ１’の一巡利得Ｇ_ＬＰ１’は、通話相手が拡声型の宅外通話端末の場合における閉ループＬＰ２の一巡利得Ｇ_ＬＰ２よりも小さい値（経験的にデシベル換算でおよそ２分の１）となるから、親機１Ａ（又は副親機１Ｂ）と拡声型の宅外通話端末（ドアホン子器５０又はロビーインターホン６０）との通話時における初期値に対して、宅内通話端末同士（親機１Ａと副親機１Ｂ又は副親機１Ｂ同士）の内線通話時における初期値を小さく設定してもエコーやハウリングが十分に防止できることになる。なお、相手の通話端末がハンドセット型の宅外通話端末（警報監視盤８０）である場合には送受話器８５の音響結合による帰還利得β２’は親機１Ａの帰還利得β１’よりも小さくなるから、上述の副親機１Ｂの場合と同様にハンドセット型の宅外通話端末（警報監視盤８０）との通話時における初期値を小さく設定してもエコーやハウリングが十分に防止できることになる。
【００５０】
そこで本実施形態では、図１に示すように通話相手の種類、すなわちドアホン子器５０、ロビーインターホン６０、警報監視盤８０並びに宅内通話端末（副親機１Ｂ又は親機１Ａ）毎に最適な初期値Ｌｔ（０）_１〜Ｌｔ（０）_４を格納するためにフラッシュメモリのように書換可能な不揮発性メモリ１５０を親機１Ａ（副親機１Ｂについても共通）に備えている。
【００５１】
一方、通話処理部１７を実現するＤＳＰ２００には、通信用インタフェースを介して主制御部２からコマンドを受信するとともに該コマンドを解釈して必要な処理を行うコマンド処理部２０１と、相手の通話端末に適した総損失量の初期値Ｌｔ（０）_１〜Ｌｔ（０）_４を不揮発性メモリ１５０から読み込んで通話処理部１７を実現するソフトウェアモジュールに対して設定するパラメータ設定部２０２とが設けてある。但し、コマンド処理部２０１並びにパラメータ設定部２０２もＤＳＰ２００のハードウェアをソフトウェアで制御することによって実現される
通話を開始する際には、まず主制御部２からＤＳＰ２００に対して通話開始要求コマンドを通信用インタフェースを介して送信し、この通話開始要求コマンドを受け取ったコマンド処理部２０１がその内容を解釈し、通話処理部１７において用いるパラメータ（総損失量の初期値Ｌｔ（０）_１〜Ｌｔ（０）_４）の格納場所を示すアドレスデータをパラメータ設定部２０２に与える。パラメータ設定部２０２では受け取ったアドレスデータに基づいて相手の通話端末に対応する初期値Ｌｔ（０）_１，…のデータを不揮発性メモリ１５０から読み込み、通話処理部１７の総損失量算出部３７ｄを実現するソフトウェアモジュールに対して初期設定する。これら初期設定が完了した後、パラメータ設定部２０２はパラメータの初期化が完了したことを示すフラグをセットし、このフラグのセットを受けてコマンド処理部２０１が応答コマンドを主制御部２に対して送信し、フラグを受け取った主制御部２が通話処理部１７を起動して通話を開始させるのである。
【００５２】
本実施形態は上述のように構成されるものであって、内線通話を行う他の宅内通話端末（副親機１Ｂ又は親機１Ａ）やハンドセット型の宅外通話端末（警報監視盤８０）に対する初期値Ｌｔ（０）_３，Ｌｔ（０）_４と、拡声型の宅外通話端末（ドアホン子器５０及びロビーインターホン６０）に対する初期値Ｌｔ（０）_１，Ｌｔ（０）_２をそれぞれ不快なエコーやハウリングの防止に必要十分な値に個別に設定することができ、これにより不快なエコーやハウリングを防止しつつ通話開始直後の切断感が軽減できるとともに通話開始から双方向の同時通話が可能な状態に至るまでの時間が短縮できる。
【００５３】
（実施形態２）
ところで総損失量算出部３７ｄの演算誤差などにより音声スイッチ３７の総損失量が必要以上に小さい値に設定されてしまうとハウリングが生じる虞がある。そこで本実施形態では、総損失量算出部３７ｄで算出する総損失量に対して下限値Ｌｔ_ＵＬを設けるとともに、その下限値Ｌｔ_ＵＬを相手の通話端末毎に個別に設定できるようにしている。
【００５４】
図１０は本実施形態における宅内通話端末（親機１Ａ及び副親機１Ｂ）の一部省略した構成を示している。但し、実施形態１と共通の構成要素には同一の符号を付して説明を省略する。
【００５５】
本実施形態においては、通話相手の種類、すなわちドアホン子器５０、ロビーインターホン６０、警報監視盤８０並びに宅内通話端末（副親機１Ｂ又は親機１Ａ）毎に最適な初期値Ｌｔ（０）_１〜Ｌｔ（０）_４並びに下限値Ｌｔ_ＵＬ１〜Ｌｔ_ＵＬ４を不揮発性メモリ１５０に格納しており、通話開始時に主制御部２から与えられるコマンドに応じてパラメータ設定部２０２が不揮発性メモリ１５０から初期値Ｌｔ（０）_１，…と下限値Ｌｔ_ＵＬ１，…を読み込んで通話処理部１７の総損失量算出部３７ｄを実現するソフトウェアモジュールに対して初期設定する。
【００５６】
総損失量算出部３７ｄでは、図１５のフローチャートに示した更新モードの処理では、ステップ６において前回の総損失量Ｌｔ（ｎ−１）が所定の減少量Δｄに下限値Ｌｔ_ＵＬ１，…を加えた値よりも大きければ前回の総損失量Ｌｔ（ｎ−１）から減少量Δｄを減算した値を今回の総損失量Ｌｔ（ｎ）とし、総損失量Ｌｔ（ｎ−１）が減少量Δｄに下限値Ｌｔ_ＵＬ１，…を加えた値以下であれば下限値Ｌｔ_ＵＬ１，…を今回の総損失量Ｌｔ（ｎ）とする。
【００５７】
このように本実施形態では、総損失量算出部３７ｄで算出する総損失量Ｌｔ（ｎ）に相手の通話端末に応じた下限値Ｌｔ_ＵＬ１，…を設定しているから、総損失量算出部３７ｄの演算誤差などにより音声スイッチ３７の総損失量が必要以上に小さい値に設定されることがなくなってハウリングが防止できる。
【００５８】
（実施形態３）
本実施形態は、宅内通話端末（親機１Ａ及び副親機１Ｂ）における音声スイッチ３７の総損失量算出部３７ｄが更新モードで動作する場合に、相手側通話端末が宅内通話端末である場合の増加量Δｉ１並びに減少量Δｄ１を相手側通話端末が宅外通話端末である場合の増加量Δｉ２並びに減少量Δｄ２よりも大きくした点に特徴がある。但し、これ以外のシステム構成並びに親機１Ａや副親機１Ｂ等の構成及び動作は実施形態１と共通であるから図示並びに説明は省略する。
【００５９】
次に、更新モードにおける総損失量算出部３７ｄの具体的な動作を図１１のフローチャートを参照して説明する。但し、図１４に示した従来例の動作と共通の点については説明を省略する。
【００６０】
固定モードから更新モードに移行した時点から所定のサンプリング周期で音響側帰還利得α並びに回線側帰還利得βの推定処理を実行してその推定値α’（ｎ），β’（ｎ）を算出し（ステップ１）、これら２つの推定値α’（ｎ），β’（ｎ）の積と利得余裕ＭＧとから、閉ループの利得余裕をＭＧ［ｄＢ］に保つために必要とされる総損失量所望値Ｌｒ（ｎ）を算出し（ステップ２）、さらに総損失量算出部３７ｄは算出したｎ回目の総損失量所望値Ｌｒ（ｎ）と、前回（ｎ−１回目）の総損失量Ｌｔ（ｎ−１）を比較する（ステップ３）。今回算出した総損失量所望値Ｌｒ（ｎ）の方が大きい場合、内線通話か否か（つまり、相手が宅内通話端末か否か）を判断し（ステップ４）、内線通話であれば前回の総損失量Ｌｔ（ｎ−１）に宅内通話端末用の増加量Δｉ１［ｄＢ］を加算した値を今回の総損失量Ｌｔ（ｎ）＝Ｌｔ（ｎ−１）＋Δｉ１とし（ステップ５）、内線通話でなければ宅外通話端末用の増加量Δｉ２［ｄＢ］を加算した値を今回の総損失量Ｌｔ（ｎ）＝Ｌｔ（ｎ−１）＋Δｉ２とする（ステップ６）。ここで、宅内通話端末用の増加量Δｉ１は例えば宅外通話端末用の増加量Δｉ２の約２倍に設定される。
【００６１】
そして、前回の総損失量Ｌｔ（ｎ−１）と今回算出した総損失量所望値Ｌｒ（ｎ）を比較し（ステップ７）、今回の総損失量所望値Ｌｒ（ｎ）の方が小さければ内線通話か否かを判断し（ステップ８）、内線通話であれば、前回の総損失量Ｌｔ（ｎ−１）から宅内通話端末用の減少量Δｄ１［ｄＢ］を減算した値を今回の総損失量Ｌｔ（ｎ）＝Ｌｔ（ｎ−１）−Δｄとし（ステップ９）、内線通話でなければ、前回の総損失量Ｌｔ（ｎ−１）から宅外通話端末用の減少量Δｄ２［ｄＢ］を減算した値を今回の総損失量Ｌｔ（ｎ）＝Ｌｔ（ｎ−１）−Δｄ２とする（ステップ１０）。ここで、宅内通話端末用の減少量Δｄ１は例えば宅外通話端末用の減少量Δｄ２の約２倍に設定される。なお、ステップ３において今回算出した総損失量所望値Ｌｒ（ｎ）が前回の総損失量Ｌｔ（ｎ−１）以下であればステップ７にとんで前回の総損失量Ｌｔ（ｎ−１）と今回算出した総損失量所望値Ｌｒ（ｎ）を比較する。また、ステップ７において今回算出した総損失量所望値Ｌｒ（ｎ）が前回の総損失量Ｌｔ（ｎ−１）以上であれば処理を終了する。
【００６２】
而して、内線通話は音声スイッチ３７を具備した宅内通話端末（親機１Ａ及び副親機１Ｂ）間で行われるから、音声スイッチ３７を具備しない拡声型の宅外通話端末（ドアホン子器５０及びロビーインターホン６０）との通話時に比較して総損失量Ｌｔ（ｎ）の更新速度が遅くなって通話者に違和感を与えてしまう可能性がある。これに対して本実施形態では、宅内通話端末（親機１Ａ及び副親機１Ｂ）の総損失量算出部３７ｄが更新モードにおいて上述のような処理を行うことにより、内線通話時と宅外通話端末との通話時において総損失量算出部３７ｄによる総損失量Ｌｔ（ｎ）の更新速度をほぼ等しくすることができ、通話者に違和感を与えずに双方向の同時通話が行える。
【００６３】
（実施形態４）
図１２は本実施形態における宅内通話端末（親機１Ａ並びに２台の副親機１Ｂ_１，１Ｂ_２）の要部の概略構成を示すブロック図である。但し、実施形態１における親機１Ａ及び副親機１Ｂと共通の構成要素には同一の符号を付して説明あるいは説明及び図示を省略する。本実施形態は、親機１Ａと副親機１Ｂ_１，１Ｂ_２又は副親機１Ｂ_１，１Ｂ_２間に形成される通話路の見かけ上のインピーダンスと、宅内通話端末（本実施形態の場合は親機１Ａ及び副親機１Ｂ_１，１Ｂ_２）と少なくとも何れか一つの宅外通話端末（ロビーインターホン６０又は警報監視盤８０）の間に形成される通話路の見かけ上のインピーダンスとを略等しくするインピーダンス調整手段を副親機１Ｂ_１，１Ｂ_２に具備した点に特徴がある。
【００６４】
すなわち、マンション等の集合住宅においては、通常、情報幹線Ｌ１（通信線Ｌａ及び制御線Ｌｂ）の配線長が数百メートルから数キロメートルに達し、宅内通話端末（親機１Ａと副親機１Ｂ_１，１Ｂ_２あるいは副親機１Ｂ_１と副親機１Ｂ_２）の間の通話線Ｌｄの配線長（数十メートル）に比較して遙かに長くなってしまうため、親機１Ａから副親機１Ｂ_１，１Ｂ_２をみたときのインピーダンスと親機１Ａから警報監視盤８０（並びにロビーインターホン６０）をみたときのインピーダンスの差が大きくなり、警報監視盤８０（又はロビーインターホン６０）との通話時の音量が低下してしまう虞があるが、本実施形態はこのような音量低下を防止できるものである。
【００６５】
親機１Ａは、内線通話、ドアホン通話（ドアホン子器５０との通話）及び警報監視盤８０（並びにロビーインターホン６０）との通話において通話路と通話回路（通話処理部１７等）のインピーダンスを整合させるためのマッチング回路１６０を備えている。また、マッチング回路１６０と通話切換部３４の間には通話路を形成するための第１のスイッチ１６１及び第２のスイッチ１６２が直列に挿入されており、２つのスイッチ１６１，１６２の接続点Ｔｃに副親機１Ｂ_１，１Ｂ_２との通話路を形成する通話線Ｌｅ_１，Ｌｅ_２が接続されている。なお、これら２つのスイッチ１６１，１６２は主制御部２によって開閉される。また通話線Ｌｅ_１，Ｌｅ_２は不平衡のペア線からなるが、平行なペア線であっても構わない。
【００６６】
一方、副親機１Ｂ_１，１Ｂ_２は、親機１Ａと同様のマッチング回路１６０と、通話線Ｌｅ_１，Ｌｅ_２とマッチング回路１６０の間の信号経路を開閉する第３のスイッチ１６３_１（副親機１Ｂ_２では第４のスイッチ１６３_２）と、マッチング回路１６０と第３のスイッチ１６３_１（及び第４のスイッチ１６３_２）の間に挿入されて上記信号経路を２通りに切り換える第１の切換スイッチ１６４_１（副親機１Ｂ_２では第２の切換スイッチ１６４_２）と、インピーダンス調整手段たるコンデンサ１６５とを備えている。コンデンサ１６５は内線通話の通話路Ｌｅ_１，Ｌｅ_２に通信線Ｌａと等価なインピーダンスを挿入するものであって、通信線Ｌａの総配線長に相当する容量成分と等価な容量値に設定される。
【００６７】
副親機１Ｂ_１においては、第１の切換スイッチ１６４_１の共通接点Ｔｚ_１が第３のスイッチ１６３_１と接続され、２つの切換接点Ｔｘ_１，Ｔｙ_１がそれぞれマッチング回路１６０に並列に接続されており、片方の切換接点Ｔｙ_１とグランドの間にコンデンサ１６５が接続されている。また副親機１Ｂ_２においても、第２の切換スイッチ１６４_２の共通接点Ｔｚ_２が第４のスイッチ１６３_２と接続され、２つの切換接点Ｔｘ_２，Ｔｙ_２がそれぞれマッチング回路１６０に並列に接続されており、片方の切換接点Ｔｙ_２とグランドの間にコンデンサ１６５が接続されている。なお、第３及び第４のスイッチ１６３_１，１６３_２、第１及び第２の切換スイッチ１６４_１，１６４_２はそれぞれ副親機１Ｂ_１，１Ｂ_２の主制御部２によって開閉及び切り換えられる。
【００６８】
ここで、親機１Ａ及び副親機１Ｂ_１，１Ｂ_２の各主制御部２で第１〜第４のスイッチ１６１，…、第１及び第２の切換スイッチ１６４_１，…のそれぞれの状態と宅内通話端末間又は宅内通話端末と宅外通話端末の間で形成される通話路との関係を図１３に示す。例えば、親機１Ａとロビーインターホン６０（又は警報監視盤８０）との間で通話する場合、親機１Ａでは第１及び第２のスイッチ１６１，１６２をオンするとともに通話切換部３４を切換接点Ｔａ２に切り換え、副親機１Ｂ_１，１Ｂ_２では第３及び第４のスイッチ１６３_１，１６３_２をオフすることで通信線Ｌａを介した通話路が形成される。このときは親機１Ａのマッチング回路１６０によってインピーダンス整合が行われる。
【００６９】
また親機１Ａと副親機１Ｂ_１の間で内線通話を行う場合、親機１Ａでは第１のスイッチ１６１をオンするとともに第２のスイッチ１６２をオフし、副親機１Ｂ_１では第３のスイッチ１６３_１をオンするとともに第１の切換スイッチ１６４_１を切換接点Ｔｙ_１に切り換え、副親機１Ｂ_２では第４のスイッチ１６３_２をオフすることにより通話線Ｌｅ_１を介して内線通話の通話路が形成される。このとき通話路にはインピーダンス調整手段たるコンデンサ１６５が並列に接続されるため、副親機１Ｂ_１から親機１Ａをみたときのインピーダンスと副親機１Ｂ_１からロビーインターホン６０（又は警報監視盤８０）をみたときのインピーダンスがほぼ等しくなり、同じく親機１Ａから副親機１Ｂ_１をみたときのインピーダンスと親機１Ａからロビーインターホン６０（又は警報監視盤８０）をみたときのインピーダンスがほぼ等しくなるため、親機１Ａ及び副親機１Ｂ_１のマッチング回路１６０のマッチング定数を変えること無くインピーダンスの整合が行える。
【００７０】
あるいは副親機１Ｂ_１，１Ｂ_２の間で内線通話を行う場合、親機１Ａでは第１及び第２のスイッチ１６１，１６２をオフし、副親機１Ｂ_１では第３のスイッチ１６３_１をオンするとともに第１の切換スイッチ１６４_１を切換接点Ｔｙ_１に切り換え、副親機１Ｂ_２では第４のスイッチ１６３_２をオンするとともに第２の切換スイッチ１６４_２を切換接点Ｔｙ_２に切り換えることにより通話線Ｌｅ_１，Ｌｅ_２を介して内線通話の通話路が形成される。このときには一方の通話線Ｌｅ_１にのみインピーダンス調整手段たるコンデンサ１６５が並列に接続されるため、一方の副親機１Ｂ_１（又は１Ｂ_２）から他方の副親機１Ｂ_２（又は１Ｂ_１）をみたときのインピーダンスと副親機１Ｂ_１からロビーインターホン６０（又は警報監視盤８０）をみたときのインピーダンスがほぼ等しくなり、副親機１Ｂ_１，１Ｂ_２のマッチング回路１６０のマッチング定数を変えること無くインピーダンスの整合が行える。
【００７１】
このように本実施形態では、副親機１Ｂ_１，１Ｂ_２にインピーダンス調整手段たるコンデンサ１６５を具備したので、相手の通話端末の種類に応じてインピーダンス整合のためのマッチング定数を変える必要が無く、マッチング回路１６０の回路構成が簡素化できる。
【００７２】
【発明の効果】
請求項１の発明は、集合住宅における各住戸の外に設置される複数の宅外通話端末と、各住戸内に設置されて前記宅外通話端末との間で通話を行う複数の宅内通話端末とを備え、前記宅外通話端末にはマイクロホンとスピーカを具備して拡声通話を行う拡声型の宅外通話端末と、ハンドセットを具備したハンドセット型の宅外通話端末とが含まれ、前記宅内通話端末には、マイクロホン、スピーカ並びにハウリングやエコーの抑制を含む拡声通話のための処理を行う通話処理手段を具備し前記宅外通話端末との間に形成される通話路を介して拡声通話を行う親機と、マイクロホン、スピーカ並びにハウリングやエコーの抑制を含む拡声通話のための処理を行う通話処理手段を具備し前記親機を介して前記宅外通話端末と拡声通話を行うとともに当該親機との間で内線通話を行う１乃至複数の副親機とが含まれる集合住宅用拡声通話システムにおいて、前記宅内通話端末の通話処理手段は、マイクロホンとスピーカの音響結合によって生じる音響エコーを抑圧する第１のエコーキャンセラと、相手側の通話端末における音響結合又は信号の回り込みによって生じる回線エコーを抑圧する第２のエコーキャンセラと、第１及び第２のエコーキャンセラの間に設けられ、前記音響エコー経路並びに回線エコー経路により形成される閉ループの一巡利得を低減してハウリングを抑制する音声スイッチとを有し、第１及び第２のエコーキャンセラは、前記エコー経路の特性を適応的に同定する適応フィルタを具備し、前記音声スイッチは、送話側の信号経路に損失を挿入する送話側損失挿入手段と、受話側の信号経路に損失を挿入する受話側損失挿入手段と、送話側及び受話側の前記各損失挿入手段から挿入する損失量を制御する挿入損失量制御手段とを具備し、該挿入損失量制御手段は、前記受話側損失挿入手段の出力点から前記音響エコー経路を介して前記送話側損失挿入手段の入力点へ帰還する経路の音響側帰還利得を推定するとともに、前記送話側損失挿入手段の出力点から前記回線エコー経路を介して前記受話側損失挿入手段の入力点へ帰還する経路の回線側帰還利得を推定し、音響側及び回線側の各帰還利得の推定値に基づいて前記閉ループに挿入すべき損失量の総和を算出する総損失量算出部と、送話信号及び受話信号を監視して通話状態を推定し、この推定結果と前記総損失量算出部の算出値に応じて前記送話側損失挿入手段及び受話側挿入損失手段の各挿入損失量の配分を決定する挿入損失量分配処理部とからなり、前記総損失量算出部は、前記各帰還利得の推定値に基づいて前記閉ループに挿入すべき損失量の総和を算出して適応更新する更新モード、並びに総損失量を所定の初期値に固定する固定モードの２つの動作モードを有し、相手側通話端末との通話開始から第１及び第２のエコーキャンセラが充分に収束するまでの期間には固定モードで動作するとともに第１及び第２のエコーキャンセラが充分に収束した後の期間には更新モードで動作し、更新モードにおいては音響側及び回線側の各帰還利得の推定値と利得余裕値とから総損失量所望値を求め、総損失量所望値が更新前の総損失量よりも大きければ総損失量を所定の増加量だけ増加した値に更新し、総損失量所望値が更新前の総損失量よりも小さければ総損失量を所定の減少量だけ減少させた値に更新するとともに、総損失量所望値が更新前の総損失量と等しければ総損失量を更新しないことで総損失量を算出しないものであり、相手側通話端末に対応した前記初期値を個別に記憶する不揮発性の記憶手段と、相手側通話端末との通話開始前に前記記憶手段に記憶した初期値を前記総損失量算出部に設定する初期値設定手段とを前記宅内通話端末に具備し、内線通話を行う他の宅内通話端末に対応した前記初期値並びにハンドセット型の宅外通話端末に対応した前記初期値を拡声型の宅外通話端末に対応した前記初期値よりも小さい値に設定したことを特徴とし、内線通話を行う他の宅内通話端末やハンドセット型の宅外通話端末に対する初期値と、拡声型の宅外通話端末に対する初期値をそれぞれ不快なエコーやハウリングの防止に必要十分な値に個別に設定することにより、不快なエコーやハウリングを防止しつつ通話開始直後の切断感が軽減できるとともに通話開始から双方向の同時通話が可能な状態に至るまでの時間が短縮できる。
【００７３】
請求項２の発明は、請求項１の発明において、前記総損失量算出部が更新モードで動作しているときにとりうる前記総損失量の下限値を、前記宅内通話端末の種別と宅外通話端末の種別に応じて個別に設定する設定手段を前記宅内通話端末に具備したことを特徴とし、総損失量算出部の演算誤差などにより総損失量が必要以上に小さい値に設定されることがなくなってハウリングが防止できる。
【００７４】
請求項３の発明は、請求項１又は２の発明において、相手側通話端末が宅内通話端末である場合の前記増加量並びに減少量を相手側通話端末が宅外通話端末である場合の前記増加量並びに減少量よりも大きくしたことを特徴とし、内線通話時と宅外通話端末との通話時において総損失量算出部による総損失量の更新速度をほぼ等しくすることができ、通話者に違和感を与えずに双方向の同時通話が行える。
【００７５】
請求項４の発明は、請求項１又は２又は３の発明において、前記副親機は宅内に設置された他の副親機との間で内線通話を行うものであって、通話路のインピーダンスを整合するマッチング手段と、前記親機と副親機又は前記副親機間に形成される通話路の見かけ上のインピーダンスを宅内通話端末と少なくとも何れか一つの宅外通話端末の間に形成される通話路の見かけ上のインピーダンスに略等しくするインピーダンス調整手段とを前記宅内通話端末に具備したことを特徴とし、相手の通話端末の種類に応じてインピーダンス整合のためのマッチング定数を変える必要が無く、マッチング手段の構成が簡素化できる。
【図面の簡単な説明】
【図１】本実施形態１における親機及び副親機の要部を示すブロック図である。
【図２】同上のシステム構成図である。
【図３】同上における親機並びにドアホン子器の構成を示すブロック図である。
【図４】同上におけるロビーインターホン並びに警報監視盤の構成を示すブロック図である。
【図５】同上における副親機の構成を示すブロック図である。
【図６】同上における親機及び副親機の通話処理部の構成を示すブロック図である。
【図７】同上における通話処理部の動作を説明するためのフローチャートである。
【図８】同上の動作説明図である。
【図９】同上の動作説明図である。
【図１０】本実施形態２における親機及び副親機の要部を示すブロック図である。
【図１１】実施形態３における総損失量算出部の動作説明用のフローチャートである。
【図１２】実施形態４における親機及び副親機の要部を示すブロック図である。
【図１３】同上の動作説明図である。
【図１４】親機及び副親機の従来例に相当する拡声通話装置のブロック図である。
【図１５】同上の拡声通話装置における総損失量算出部の動作説明用のフローチャートである。
【符号の説明】
１Ａ住宅情報盤親機
１Ｂ住宅情報盤副親機
２主制御部
１７通話処理部
１５０不揮発性メモリ
２００ＤＳＰ
２０１コマンド処理部
２０２パラメータ設定部[0001]
TECHNICAL FIELD OF THE INVENTION
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice communication system for an apartment house used for an apartment house or the like.
[0002]
[Prior art]
In recent years, in apartments and other condominiums, lobby intercoms installed at the common entrance of apartments, monitoring panels installed in the manager's room and disaster prevention monitoring room, and master units installed in each dwelling unit have also been Equipped with a sub-master unit installed in each dwelling unit and a door phone slave unit installed outside (entrance) of each dwelling unit. ") And a call between a lobby intercom, a door phone handset or a monitoring panel (collectively referred to as" out-of-home call terminals "), and a call between a base unit and a sub-base unit (extension call). ) Is being used.
[0003]
The present applicant has already proposed a loudspeaker apparatus as shown in FIG. 14 as a base unit or a sub-base unit used in such a loudspeaker system for an apartment house (see Patent Document 1). Hereinafter, this conventional loudspeaker apparatus will be described.
[0004]
The conventional loudspeaker system includes a microphone 100, a speaker 101, a two-wire to four-wire conversion circuit 103, a microphone amplifier G1, a line output amplifier G2 for amplifying a transmission signal to a line (two-wire transmission line), and a line. A line input amplifier G3, a speaker amplifier G4, a transmission volume adjustment amplifier G5, a reception volume adjustment amplifier G6, and first and second echo cancellers 110A and 110B and a voice switch 120 for amplifying the reception signal of the receiver. And call processing means for performing processing for a loudspeaker call including echo suppression.
[0005]
The first echo canceller 110A has a conventionally well-known configuration including an adaptive filter 111A and a subtractor 112A, and a feedback path (acoustic echo path) H formed by acoustic coupling between the speaker 101 and the microphone 100._ACIs adaptively identified by the adaptive filter 111A, and the echo component (acoustic echo) estimated from the reference signal (input signal to the speaker amplifier G4) is subtracted from the output signal of the microphone amplifier G1 by the subtractor 112A. It suppresses the echo component. The second echo canceller 110B also has a conventionally well-known configuration including an adaptive filter 111B and a subtractor 112B. Return path (line echo path) H formed by the acoustic coupling between the speaker and the microphone in the loudspeaker terminal (for example, a door phone slave unit)_LINIs adaptively identified by the adaptive filter 111B, and the echo component (line echo) estimated from the reference signal (the input signal to the line output amplifier G2, that is, the transmission signal) is subtracted from the reception signal by the subtractor 112B. This suppresses the echo component.
[0006]
An audio switch 120 is provided between a transmission volume adjustment amplifier G5 and a reception volume adjustment amplifier G6 provided on a signal path of a transmission signal and a reception signal sandwiched between the first and second echo cancellers 110A and 110B. Is provided. The voice switch 120 includes a transmission-side loss insertion unit 121 that inserts a loss in a transmission-side signal path, a reception-side loss insertion unit 122 that inserts a loss in a reception-side signal path, a transmission-side and reception-side And an insertion loss control means 123 for controlling the amount of loss inserted from each of the loss insertion means 121 and 122. The insertion loss amount control means 123 receives the acoustic echo path H from the output point Rout of the receiving-side loss insertion means 122._ACThe acoustic feedback gain α of a path (hereinafter, referred to as “acoustic feedback path”) that returns to the input point Tin of the transmitting-side loss inserting means 121 via the output side of the transmitting-side loss inserting means 121 is estimated. Line echo path H from Tout_LIN, A line-side feedback gain β of a path (hereinafter, referred to as a “line-side feedback path”) that returns to the input point Rin of the receiving-side loss insertion unit 122 is estimated, and the acoustic-side and line-side feedback gains α, β To calculate the total sum of the loss amounts to be inserted into the closed loop (the sum of the insertion loss amount of the transmitting-side loss inserting unit 121 and the insertion loss amount of the receiving-side loss inserting unit 122) based on the estimated values α ′ and β ′ of The loss amount calculation unit 124 monitors the transmission signal and the reception signal to estimate the communication state, and according to the estimation result and the value calculated by the total loss amount calculation unit 124, the transmission-side loss insertion unit 121 and the reception-side loss And an insertion loss distribution processor 125 that determines the distribution of each insertion loss of the insertion means 122. The call processing means including the first and second echo cancellers 110A and 110B and the voice switch 120 is realized by, for example, mounting dedicated software on hardware of a DSP (Digital Signal Processor).
[0007]
The total loss amount calculation unit 124 estimates a short-time average power of the input signal of the transmitting-side loss insertion unit 121 using a rectifier smoother, a low-pass filter, or the like. The time average power of the output signal of the receiving-side loss insertion means 122 in a short time is estimated using_ACThe minimum value of the estimated value of the time-average power of the output signal of the receiving-side loss insertion means 122 is obtained at the maximum delay time assumed in the following. The value obtained by dividing the estimated value is used as an estimated value α ′ of the acoustic feedback gain α, and the time-average power of the input signal of the receiving-side loss insertion unit 122 in a short time is measured using a rectifier smoother or a low-pass filter. The time-average power of the output signal of the transmitting-side loss insertion means 121 is also estimated in a short time by using a rectifier / smoothing device or a low-pass filter, and the like._LINThe minimum value of the estimated value of the time-average power of the output signal of the transmitting-side loss insertion means 121 is determined at the maximum delay time assumed in the above. The value obtained by dividing the estimated value is used as an estimated value β ′ of the line-side feedback gain β. Then, the total loss amount calculation unit 124 calculates the total loss amount Lt necessary to obtain a desired gain margin MG from the estimated values α ′ and β ′ of the acoustic feedback gain α and the line feedback gain β. The value Lt is output to the insertion loss distribution processing unit 125.
[0008]
The insertion loss distribution processor 125 monitors the input / output signals of the transmitting-side loss inserting means 121 and the input / output signals of the receiving-side loss inserting means 122, and determines the magnitude relationship between the power levels of these signals and the presence or absence of a voice signal. And the total loss Lt is distributed to the transmitting-side loss inserting means 121 and the receiving-side loss inserting means 122 at a rate corresponding to the determined calling state. The insertion loss amount of each of the loss insertion means 121 and 122 is adjusted so as to perform the operation.
[0009]
By the way, the total loss amount calculating section 124 in the conventional example calculates the total sum of the loss amounts to be inserted into the closed loop based on the estimated values α ′ and β ′ of the feedback gains α and β as described above, and performs adaptive updating. Update mode, and a fixed mode in which the total loss amount is fixed to a predetermined initial value, and the first and second echo cancellers 110A and 110B are sufficiently operated from the start of communication with the other party's communication terminal. It operates in the fixed mode during the period until the convergence, and operates in the update mode during the period after the first and second echo cancellers 110A and 110B sufficiently converge. That is, in the total loss amount calculation unit 124, the estimated values α ′ and β ′ of the acoustic feedback gain α and the line feedback gain β both continue for a predetermined time (several hundred milliseconds) or more from the start of the call and have a predetermined threshold ε ( For example, when the estimated values α ′ and β ′ at the start of a call fall below 10 dB to 15 dB), the first and second echo cancellers 110 A and 110 B are considered to have sufficiently converged. Previously, the operation was performed in a fixed mode in which the total loss amount was fixed at an initial value, and after the time point, the operation mode was switched to an update mode in which the total loss amount was adaptively updated based on the estimated values α ′ and β ′. Note that the initial value of the total loss amount in the fixed mode is set to a value sufficiently larger than the total loss amount updated as needed in the update mode.
[0010]
Thus, in a state where the first and second echo cancellers 110A and 110B have not sufficiently converged immediately after the start of the call, the total loss calculator 124 operating in the fixed mode sets the value to a sufficiently large value. Since the total loss amount of the initial value is inserted in the closed loop, unpleasant echo (acoustic echo and line echo) and howling can be suppressed, and a stable half-duplex call can be realized. Further, when the first and second echo cancellers 110A and 110B have sufficiently converged after a lapse of time from the start of the call, the operation mode of the total loss calculator 124 switches from the fixed mode to the update mode, and the closed loop is closed. Is reduced to a value sufficiently lower than the initial value, so that two-way simultaneous communication can be realized.
[0011]
Next, a specific operation of the total loss amount calculation unit 124 in the update mode will be described with reference to a flowchart of FIG.
[0012]
The total loss amount calculation unit 124 performs estimation processing of the acoustic feedback gain α and the line feedback gain β at a predetermined sampling cycle from the point in time when the mode is shifted from the fixed mode to the update mode, and estimates the estimated values α ′ (n), β ′ (n) is calculated (step 1), and the product of these two estimated values α ′ (n) and β ′ (n) and the gain margin MG are used to maintain the gain margin of the closed loop at MG [dB]. Is calculated by the following formula (step 2).
[0013]
Lr (n) = 20 log | α ′ (n) · β ′ (n) | + MG [dB]
Note that α ′ (n), β ′ (n), and Lr (n) indicate the estimated value of the feedback gain and the desired value of the total loss amount, respectively, calculated by the n-th sampling from the transition to the update mode. Further, the total loss amount calculation unit 124 calculates the n-th desired total loss amount value Lr (n) calculated from the above equation and the previous (n−1-th) total loss amount Lt (n−1), that is, the previous processing. When the total loss amount desired value Lr (n) calculated this time is larger than the total loss amount actually determined and inserted in the above, the previous total loss amount Lt (n−1) is slightly increased by Δi [dB]. ] Is added to the current total loss amount Lt (n) = Lt (n−1) + Δi (steps 3 and 4), and the total amount calculated this time with respect to the previous total loss amount Lt (n−1) is calculated. When the desired loss amount Lr (n) is small, the value obtained by subtracting the slight decrease amount Δd [dB] from the previous total loss amount Lt (n−1) is the current total loss amount Lt (n) = Lt (n). -1) -Δd (steps 5 and 6).
[0014]
As described above, by suppressing the increase / decrease of the total loss amount by the total loss amount calculation unit 124 to a small value of Δi or Δd, the first and second echo cancellers 110A, 110A, 110B actively updates the coefficient toward convergence, so that even in a state where the acoustic feedback gain α and the line feedback gain β change drastically, it is possible to eliminate a sense of incongruity in hearing.
[0015]
[Patent Document 1]
JP-A-2002-359580 (paragraphs 0027 to 0036, FIGS. 1 and 2)
[0016]
[Problems to be solved by the invention]
By the way, the above-mentioned feedback gain (line-side feedback gain β) differs depending on the type of the other party's communication terminal, and usually, the doorphone handset installed outdoors or the length of the communication path is longer than those of other communication terminals. Compared to the return gain for the lobby intercom, which is considerably long, the return gain for the parent device or the sub-parent device having the call processing means or the monitoring panel, which is a handset type out-of-home telephone terminal, is much smaller. .
[0017]
However, in the above-mentioned conventional loudspeaker, the initial value of the total loss in the fixed mode at the start of the call is uniformly set to the same value for any of the call terminals. Even if the value is sufficient and sufficient to suppress echo and howling, the value is unnecessarily large for the master unit, the auxiliary master unit, and the monitoring panel. For this reason, in the case of an extension call between home call terminals or a call between a home call terminal and a handset type outside call terminal (monitoring panel), the sense that the other party's voice is interrupted immediately after the start of the call (this is referred to as “disconnection feeling”) ") Becomes stronger, or the total loss is set to an appropriate value that is sufficiently smaller than the initial value after the transition from the fixed mode to the update mode, and it is longer than necessary until two-way simultaneous communication becomes possible. There was a risk that it would take time.
[0018]
The present invention has been made in view of the above circumstances, and its purpose is to reduce uncomfortable feeling immediately after the start of a call while preventing unpleasant echo and howling and to enable simultaneous two-way communication from the start of the call. It is an object of the present invention to provide a loudspeaker communication system for a housing complex, which can shorten the time required to reach the service.
[0019]
[Means for Solving the Problems]
In order to achieve the above object, the invention according to claim 1 includes a plurality of out-of-home call terminals installed outside each dwelling unit in an apartment house, and a plurality of out-of-home call terminals installed in each dwelling unit. A plurality of in-home call terminals for making a call, wherein the out-of-home call terminal includes a microphone and a speaker to perform a loudspeaker call; and a handset-type out-of-home call terminal including a handset. Wherein the in-home call terminal includes a microphone, a speaker, and call processing means for performing processing for a loudspeaker call including suppression of howling and echo, and a call formed with the out-of-home call terminal. A base unit for making a voice call via a road; and a microphone, a speaker, and call processing means for performing processing for the voice call including suppression of howling and echo, and the outside-home call via the base unit. In a loudspeaker system for multiple dwelling units including one or a plurality of sub-base units for making a loudspeaker call with the base unit and making an extension call with the base unit, the call processing means of the in-home call terminal includes a microphone and a speaker. A first echo canceller for suppressing an acoustic echo caused by acoustic coupling of the first and second echo cancellers, a second echo canceller for suppressing a line echo caused by acoustic coupling or signal wraparound in a call terminal of the other party, and first and second echoes A voice switch provided between the cancellers to reduce a loop gain of a closed loop formed by the acoustic echo path and the line echo path to suppress howling; the first and second echo cancellers An adaptive filter for adaptively identifying the characteristics of the path, wherein the voice switch includes a lossy signal path in the transmitting side. A transmitting-side loss inserting unit to be inserted, a receiving-side loss inserting unit that inserts a loss in a signal path of the receiving side, and an insertion loss amount that controls a loss amount inserted from each of the loss inserting units on the transmitting side and the receiving side. Control means, the insertion loss amount control means comprising: an acoustic feedback of a path returning from an output point of the receiving side loss inserting means to an input point of the transmitting side loss inserting means via the acoustic echo path. Estimating the gain and estimating the line-side feedback gain of the path returning from the output point of the transmitting-side loss insertion means to the input point of the receiving-side loss insertion means via the line echo path, A total loss amount calculator for calculating a total loss amount to be inserted into the closed loop based on the estimated value of each feedback gain on the side, and a transmission state is estimated by monitoring a transmission signal and a reception signal. And the calculation of the total loss amount calculation unit An insertion loss amount distribution processing unit that determines the distribution of each insertion loss amount of the transmission side loss insertion unit and the reception side insertion loss unit according to the output value, and wherein the total loss amount calculation unit Has two operation modes: an update mode in which the sum of the loss amounts to be inserted into the closed loop is calculated based on the estimated value and an adaptive update, and a fixed mode in which the total loss amount is fixed to a predetermined initial value. It operates in the fixed mode during the period from the start of communication with the local communication terminal until the first and second echo cancellers sufficiently converge, and during the period after the first and second echo cancellers sufficiently converge. It operates in the update mode.In the update mode, the total loss desired value is obtained from the estimated values of the feedback gains on the sound side and the line side and the gain margin value, and the total loss desired value is smaller than the total loss amount before the update. The larger the total loss If the total loss amount is smaller than the total loss amount before the update, the total loss amount is updated to a value reduced by a predetermined decrease amount, and the total loss amount is updated to a value increased by a predetermined amount. If the value is equal to the total loss amount before the update, the total loss amount is not calculated by not updating the total loss amount, and a non-volatile storage unit that individually stores the initial value corresponding to the other party's call terminal. An initial value setting means for setting an initial value stored in the storage means to the total loss amount calculating unit before the start of a call with the other party's call terminal; It is characterized in that the initial value corresponding to the call terminal and the initial value corresponding to the handset type out-of-home call terminal are set to values smaller than the initial value corresponding to the loudspeaker type out-of-home call terminal.
[0020]
According to a second aspect of the present invention, in the first aspect of the present invention, the lower limit value of the total loss amount that can be taken when the total loss amount calculating section is operating in the update mode is determined by the type of the home communication terminal and the out-of-home call. The in-home communication terminal is provided with setting means for individually setting according to the type of the terminal.
[0021]
According to a third aspect of the present invention, in the first or second aspect of the invention, the increase amount and the decrease amount when the other party's call terminal is an in-home call terminal are determined by the increase when the other party's call terminal is an out-of-home call terminal. It is characterized in that it is larger than the amount and the reduction amount.
[0022]
According to a fourth aspect of the present invention, in the first or second or third aspect of the invention, the sub master unit performs an extension call with another sub master unit installed in the house, and the impedance of the communication path is increased. Matching means for matching an apparent impedance of a communication path formed between the master unit and the sub master unit or the sub master unit between the in-home communication terminal and at least one out-of-home communication terminal. And an impedance adjusting means for making the impedance substantially equal to the apparent impedance of the communication path.
[0023]
BEST MODE FOR CARRYING OUT THE INVENTION
(Embodiment 1)
Hereinafter, an embodiment in which the present invention is applied to a house information panel system will be described in detail.
[0024]
As shown in FIG. 2, the house information panel system according to the present embodiment includes a lobby intercom 60 installed at a common entrance of an apartment house, an alarm monitoring panel 80 installed at a management room of the apartment house, and the like. A home information panel main unit (hereinafter abbreviated as “main unit”) 1A and a home information panel sub-main unit (hereinafter abbreviated as “sub-main unit”) 1B to be installed, and a door phone child installed outside each dwelling unit And a vessel 50. In addition, an emergency power supply 90 and a broadcast facility 91 are installed in an administrator room and the like together with an alarm monitoring panel 80, and are connected to the master unit 1A of each dwelling unit via an emergency power supply line L2 and an audio output line L3, respectively. Then, the alarm monitor panel 80 and the master unit 1A of each dwelling unit multiplex the audio signal such as the call audio or the alarm sound and the video signal captured by the television camera 68 provided in the lobby intercom 60, a communication line La for calling and alarm. And the like are connected by a control line Lb for multiplex transmission of control signals. Here, the communication line La and the control line Lb are collectively referred to as an information trunk line L1. In the present embodiment, a configuration in which a video signal is multiplexed with an audio signal is illustrated, but a configuration in which a video signal is transmitted through an independent video transmission line may be employed.
[0025]
FIG. 4 is a block diagram showing a configuration of the lobby intercom 60 and the alarm monitoring panel 80 in the present embodiment. The lobby intercom 60 includes a signal processing unit 61 having a CPU as a main component, an operation unit 62 including a numeric keypad and a call button, a display unit 63 including a display lamp and a liquid crystal display, and a control line Lb. A transmission / reception unit 64 for transmitting / receiving a control signal to / from the alarm monitoring panel 80, a microphone 65 and a speaker 66, and a loudspeaker call is performed between the master unit 1A and the alarm monitoring panel 80 using the microphone 65 and the speaker 66. Control unit 67, a television camera 68 having an image pickup device such as a CCD, a video signal processing unit 69 that processes an output of the television camera 68 to output a video signal, and a signal output from the call control unit 67. The audio signal and the video signal output from the video signal processing unit 69 are multiplexed and transmitted to the communication line La, and the sound transmitted through the communication line La is multiplexed. And an output processing unit 70 for outputting a signal to the call control unit 67.
[0026]
On the other hand, the alarm monitoring panel 80 includes a signal processing unit 81 having a CPU as a main component, an operation unit 82 having a numeric keypad and a call button, a display unit 83 having a display lamp and a liquid crystal display, and a control line Lb. , A transmission / reception unit 84 for transmitting / receiving a control signal to / from each of the base unit 1A, the sub-base unit 1B, and the lobby intercom 60, a handset 85, and a base unit 1A using the handset 85. A call control unit 87 for making a simultaneous call between the master unit 1B and the lobby intercom 60 is provided.
[0027]
FIG. 3 is a block diagram showing a configuration of the base unit 1A and the doorphone slave unit 50 in the present embodiment, and the base unit 1A and the doorphone slave unit 50 are connected by a communication line Lc. The doorphone slave unit 50 includes a microphone 51 and a speaker 52, a two-wire to four-wire conversion unit 53, a television camera 54 having an image sensor such as a CCD, and a signal processing of the output of the television camera 54 to output a video signal. The video signal processing unit 55 which multiplexes the audio signal output from the two-wire to four-wire conversion unit 53 and the video signal output from the video signal processing unit 55, transmits the multiplexed signal to the communication line Lc, and transmits the multiplexed signal via the communication line Lc. The input / output processing unit 56 that outputs the received voice signal to the 2-line to 4-line conversion unit 53, and the calling by changing the voltage level of the communication line Lc when the call button 57a (see FIG. 2) is operated. A call signal transmitting section 57 for transmitting a signal.
[0028]
Base unit 1A includes a main control unit 2 having a CPU as a main component, an operation unit 3 including a response button 3a and the like, a storage unit 4 including a nonvolatile memory such as an EEPROM, and an alarm via a control line Lb. A transmission / reception unit 5 for transmitting / receiving a control signal to / from the monitoring panel 80, and various types such as a fire detector S1, a combined gas / CO sensor S2, a security external set switch SW, a security switch SW1, a call switch SW2, and an emergency switch SW3. A signal monitoring unit 6 to which a safety device is connected to monitor an alarm signal from the fire detector S1, a gas leak signal from the combined gas / CO sensor S2 or an operation signal of the call switch SW2, and a voltage level change of the communication line Lc. Detecting unit 7 for monitoring a call signal from the intercom slave unit 50 and receiving a power supply from the commercial AC power supply AC or the emergency power supply 90 to receive the main unit 1A and the sub-master unit. It includes a power source part 8 to create operating power 1B, a power supply switching unit 9 for switching the power supply source for the power supply unit 8 to the AC voltage AC and emergency power 90.
[0029]
Master device 1A also processes a video signal from microphone 10 and speaker 11, a monitor unit 13 such as a CRT or a liquid crystal display, and a door phone slave unit 50 or lobby intercom 60 to display an image on monitor unit 13. And a video display unit 14 for displaying (superimposing) a predetermined symbol in accordance with a display control signal given from the main control unit 2, an audio signal multiplexed and transmitted from the door phone slave unit 50 or the lobby interphone 60, and A process for realizing simultaneous voice communication between the signal separation unit 16 for separating the video signal, the 2-wire to 4-wire conversion unit 15, the intercom slave unit 50, the alarm monitor panel 80, and the sub master unit 1B is performed. And a call processing unit 17 that performs processing for realizing simultaneous voice communication with the lobby intercom 60, Alternatively, a voice generation unit 18 that generates an alarm message or an alarm sound, a microphone amplifier 19 that amplifies an output signal (transmission signal) of the microphone 10, and an analog transmission signal amplified by the microphone amplifier 19 is converted into a digital signal. A first A / D converter 20 for conversion, a first D / A converter 21 for converting a digital transmission signal output from the call processor 17 into an analog signal, and a first D / A converter 21 A transmission signal amplifier 22 for amplifying an analog transmission signal, a reception signal amplifier 23 for amplifying a voice signal (reception signal) separated by the signal separation unit 16, and an analog reception signal amplified by the reception signal amplifier 23 To a digital signal, and a second D / A to convert a digital reception signal output from the speech processing unit 17 into an analog signal. Conversion unit 25, a first speaker amplifier 26 that amplifies the analog reception signal converted by the second D / A conversion unit 25, and converts a digital voice signal such as a ringing tone generated by the voice generation unit 18 into an analog signal. A third D / A converter 27 for conversion; a second speaker amplifier 28 for amplifying the analog audio signal converted by the third D / A converter 27; a reception signal amplified by the first speaker amplifier 26; A first analog signal multiplexing unit 29 for multiplexing a sound signal such as a ringing tone amplified by the speaker amplifier 28, an input terminal Ts1 of the speaker 11 is connected to an output terminal Ts2 of the first analog signal multiplexing unit 29, and a sound output line L3. A broadcast switching unit 30 for switching to the connected connection terminal Ts3, and a fourth D / A conversion for converting a digital audio signal such as a ring tone generated by the audio generation unit 18 into an analog signal. Unit 31, an audio signal amplifier 32 for amplifying the analog audio signal converted by the fourth D / A conversion unit 31, a transmission signal amplified by the transmission signal amplifier 22, and a call amplified by the audio signal amplifier 32. A second analog signal multiplexing unit 33 that multiplexes audio signals such as sounds and outputs the multiplexed signal to the two-wire to four-wire conversion unit 15 is connected to the input / output terminal Ta1 of the signal separation unit 16 by the communication line Lc from the door phone handset 50. And a communication switching unit 34 for switching to a terminal Ta2 connected to the communication terminal La and a terminal Ta3 connected to the communication line La. In the present embodiment, the call processing unit 17 and the voice generation unit 18 are configured by a digital signal processor (DSP) 200, and the hardware of the DSP 200 is controlled by dedicated software, so that the call processing unit 17 and the voice The function of the generation unit 18 is realized. Further, a connector CN for outputting a video signal is connected to an output terminal of the video display unit 14, and by connecting to the connector CN, the monitor unit 13 and the monitor unit 13 are displayed on a screen of a monitor device of a television receiver or a personal computer. The same video can be projected.
[0030]
On the other hand, as shown in FIG. 5, sub-base unit 1B has substantially the same components as base unit 1A, and therefore the same reference numerals are given to the same components and description thereof will be omitted.
[0031]
The call processing unit 17 of the master unit 1A (and the sub master unit 1B is also common) has the same configuration as that of the conventional loudspeaker, as shown in FIG. An acoustic echo canceller 35 for suppressing the generated acoustic echo, and a line echo canceller 36 for suppressing the line echo generated by the acoustic coupling in the door phone handset 50, the lobby intercom 60 or the alarm monitoring panel 80 or the signal wraparound via the communication line La. A voice switch 37 provided between the acoustic echo canceller 35 and the line echo canceller 36 to reduce a loop gain of a closed loop (echo path) formed by the acoustic echo path and the line echo path to suppress howling; Of the transmission signal between the line 35 and the line echo canceller 36 And a transmission-side amplifier 38 and the receiving side amplifier 39 respectively provided on the signal path of the reception signal.
[0032]
The acoustic echo canceller 35 includes an adaptive filter 35a and a subtractor 35b, similarly to the first echo canceller in the conventional example, and has an impulse response of a feedback path (acoustic echo path) formed by acoustic coupling between the speaker 11 and the microphone 10. Is adaptively identified by the adaptive filter 35a, and the echo component (acoustic echo) estimated from the reference signal (the reception signal output from the voice switch 37) is output from the first A / D converter 20 by the subtractor 35b. (E.g., a speech signal) to cancel and suppress the echo component. The line echo canceller 36 also has a common configuration with the second echo canceller in the conventional example, and includes an adaptive filter 36a and a subtractor 36b, and the two-to-four-wire converter 15 and the communication line La or the communication line La. The adaptive filter 36a adaptively identifies the impulse response of the feedback path (line echo path) formed by the reflection due to the impedance mismatch with Lc and the acoustic coupling between the speaker and the microphone in the doorphone handset 50 or the like. The subtractor 36b subtracts the echo component (line echo) estimated from the reference signal (the transmission signal output to the first D / A converter 21) from the reception signal to cancel out the echo component and suppress the echo component. is there.
[0033]
The voice switch 37 includes a transmission signal attenuator 37a as a transmission-side loss insertion unit that inserts a loss into a transmission-side signal path, and a reception side as a reception-side loss insertion unit that inserts a loss into a reception-side signal path. Insertion loss for controlling the signal attenuator 37b and the amount of loss (gain of each attenuator 37a, 37b) inserted from each of the attenuators 37a, 37b on the transmitting side and the receiving side to the signal path on the transmitting side and the receiving side, respectively. And an amount control unit 37c. Further, the insertion loss control section 37c controls the sound of a path (hereinafter, referred to as an “acoustic side return path”) that returns from the output point of the received signal attenuator 37b to the input point of the transmitted signal attenuator 37a via the acoustic echo path. The side feedback gain is estimated, and the line side of the path (hereinafter, referred to as “line side feedback path”) that returns from the output point of the transmission signal attenuator 37a to the input point of the reception signal attenuator 37b via the line echo path. The feedback gain is estimated, and the sum of the loss amounts to be inserted into the closed loop (the insertion loss amount of the transmission signal attenuator 37a and the reception signal attenuation) based on the estimated values α ′ and β ′ of the feedback gains on the acoustic and line sides. Total loss calculating section 37d for calculating the sum of the insertion loss of the transmitter 37b), and monitoring the transmission signal and the receiving signal to estimate the call state. The estimation result and the value calculated by the total loss calculating section 37d are calculated. According to the transmission signal attenuator 37a And a distribution unit 37e for allocating each insertion loss of the reception signal attenuator 37b, that is, an insertion loss distribution processing unit 37e for determining a gain.
[0034]
The total loss calculator 37d estimates a short-time average power of the input signal of the transmission signal attenuator 37a using a rectifier smoother, a low-pass filter, or the like, and similarly uses a rectifier smoother, a low-pass filter, or the like. Is used to estimate the time average power of the output signal of the reception signal attenuator 37b in a short time, and the estimated value of the time average power of the output signal of the reception signal attenuator 37b at the maximum delay time assumed in the acoustic feedback path. Is obtained, and a value obtained by dividing the estimated value of the time average power of the input signal of the transmission signal attenuator 37a by the minimum value is used as an estimated value α 'of the acoustic feedback gain, and a rectifier smoother and a low-frequency The time average power of the input signal of the reception signal attenuator 37b in a short time is estimated using a pass filter or the like, and the output of the transmission signal attenuator 37a is similarly used using a rectifier smoother or a low-pass filter. The average value of the time average power of the output signal of the transmission signal attenuator 37a at the maximum delay time assumed in the line-side feedback path is obtained, and the minimum value is calculated. Then, the value obtained by dividing the estimated value of the time average power of the input signal of the received signal attenuator 37b is used as the estimated value β 'of the line side feedback gain. The total loss calculator 37d calculates the total loss required to obtain a desired gain margin from the estimated values α ′ and β ′ of the acoustic feedback gain and the line feedback gain, and calculates the insertion loss. Output to the quantity distribution processing unit 37e.
[0035]
The insertion loss distribution processor 37e monitors the input / output signal of the transmission signal attenuator 37a and the input / output signal of the reception signal attenuator 37b, and obtains information such as the magnitude relationship between the power levels of these signals and the presence or absence of a voice signal. , A call state (received state, transmitted state, etc.) is determined, and each attenuation is distributed so as to distribute the total loss amount to the transmitted signal attenuator 37a and the received signal attenuator 37b at a rate corresponding to the determined call state. The insertion loss (gain) of the devices 37a and 39 is adjusted.
[0036]
The call processing unit 17 includes a transmission signal single tone detector 140 for detecting whether the transmission signal is a single frequency tone signal, a reception signal single tone detector 140 for detecting whether the reception signal is a single frequency tone signal. The apparatus includes a tone detector 141 and a howling detection unit 142 for detecting the presence or absence of howling based on the detection results of the two single tone detectors 140 and 141.
[0037]
The two single tone detectors 140 and 141 use, for example, a fast Fourier transform of a transmission signal or a reception signal, or compare the power of the transmission signal or the reception signal using a plurality of band division filters to generate a tone signal. Whether or not the detection method is used can be realized by using a conventionally well-known technique, and a detailed description thereof will be omitted.
[0038]
In the howling detecting section 142, when the state in which the single tone detectors 140 and 141 both detect a tone signal of a single frequency continues for a predetermined time (for example, about 200 to 300 ms), it is determined that acoustic feedback has occurred. A howling detection signal is output to the canceller 35 and the line echo canceller 36, and otherwise, no howling detection signal is output.
[0039]
Next, the operation of this embodiment will be described with reference to the flowchart of FIG. When a call is started, initialization processing is performed in the voice switch 37 and the two echo cancellers 35 and 36 (step 1). In the voice switch 37, the operation mode of the total loss amount calculation unit 37d is fixed by the initialization processing. Mode (step 2). The total loss calculating section 37d monitors the output state of the howling detecting section 142 (step 3). When a howling detecting signal is output from the howling detecting section 142 during operation in the fixed mode, the process returns to the initialization processing. Initialization processing is performed in the voice switch 37 and the two echo cancellers 35 and 36 (step 1), and the total loss calculator 37d also operates in the fixed mode again after initializing all variables (step 2).
[0040]
On the other hand, when the howling detection signal is not output from the howling detection unit 142, in the total loss amount calculation unit 37d, the two echo cancellers 35 and 36 determine the convergence condition (for example, the estimated values α ′ and β ′ of the feedback gains α and β). Is lower than a predetermined threshold value ε for a predetermined period of time (step 4). If at least one of the echo cancellers 35 or 36 does not satisfy the convergence condition, the operation in the fixed mode is continued without shifting to the update mode (step 2). Also satisfies the convergence condition, the mode shifts to the update mode (step 5).
[0041]
Then, when a howling detection signal is output from the howling detection unit 142 during operation in the update mode, the process returns to the initialization processing, and the audio switch 37 and the two echo cancellers 35 and 36 perform initialization processing (step 1). The total loss amount calculator 37d also operates in the fixed mode again after initializing all variables (step 2).
[0042]
Thus, in the voice switch 37, in any of the fixed mode and the update mode, when a howling is detected during a call, the coefficients of the adaptive filters 35a and 36a of the two echo cancellers 35 and 36 are promptly changed. Is initialized, the total loss is returned to the initial value, and the total loss calculator 37d operates in the fixed mode until the two echo cancellers 35 and 36 converge. A sufficient margin for howling can be obtained by suppressing the gain, and even if howling occurs during a call, howling can be suppressed immediately (within several hundred ms) and a stable call can be resumed.
[0043]
Next, the gist of the present invention will be described. The operation of the total loss calculator 37d in the update mode is the same as that of the conventional example (see the flowchart of FIG. 15), but in the present embodiment, when starting a call, the type of the call terminal of the other party, that is, the in-home call terminal (Auxiliary master unit 1B or master unit 1A) and a handset type outside-the-home communication terminal (alarm monitoring panel 80) and a loudspeaker-type outside-the-home communication terminal (the doorphone slave unit 50 and the lobby intercom 60), the initial amount of the total loss. The initial value for the secondary master unit 1B (or the master unit 1A) and the alarm monitor panel 80 is set to a value smaller than the initial value for the doorphone slave unit 50 and the lobby intercom 60 (for example, 2 in decibel conversion). (1 or less).
[0044]
Here, the initial value at the time of the telephone call between the base unit 1A (or the sub-base unit 1B) and the loudspeaker type out-of-home telephone terminal (doorphone slave unit 50 or lobby intercom 60) is compared with the home telephone terminals (the parent device 1A). The initial value at the time of an extension call between the base unit 1B and the sub-base unit 1B or between the sub-base units 1B or between the base unit 1A (or the sub-base unit 1B) and the handset type outside-home telephone terminal (alarm monitoring panel 80) is set to a small value. The reason why echo and howling can be sufficiently prevented will be briefly described.
[0045]
FIG. 8 is a schematic configuration diagram showing a state in which a communication path is formed between base unit 1A and a loudspeaker type outside call terminal (doorphone handset 50 or lobby intercom 60). Since the out-of-home call terminal does not include a call processing unit including a DSP, the transmission signal gain (gain before input from the microphone 10 to the two-wire to four-wire conversion unit 15) T1A in the base unit 1A is outside the home. The transmission signal gain (gain from the input of the transmission signal of the base unit 1A to the arrival at the speaker 52) T2 in the telephone terminal is much larger than T2 (T2≪T1A), and the reception signal gain (2 The gain from the output from the line-to-line converter 15 to the arrival at the speaker 11) R1A is the reception signal gain (the transmission signal output from the microphone 51 is the 2-line-4 signal of the base unit 1A) at the outside telephone terminal. The gain before input to the line conversion unit 15) is very large (R2 （R1A). In addition, feedback gain β1 of reflection due to impedance mismatch in 2-wire to 4-wire conversion unit 15 of base unit 1A is smaller than feedback gain β2 due to acoustic coupling between the microphone and the speaker of the outside-home telephone terminal (β1 <β2).
[0046]
Therefore, the loop gain G of the closed loop LP1 due to the reflection due to the impedance mismatch in the two-wire to four-wire conversion unit 15_LP1(= Α × T1A × β1 × R1A) is the loop gain G of the closed loop LP2 due to the acoustic coupling of the external call terminal._LP2(= Α × T1A × β2 × R1A) (G_LP1<G_LP2).
[0047]
On the other hand, FIG. 9 is a schematic configuration diagram showing a state in which a communication path is formed between base unit 1A and the home communication terminal (sub-base unit 1B). Since the sub master unit 1B includes the speech processing unit 17 (DSP 200) including the voice switch 37 and the echo cancellers 35 and 36 similarly to the master unit 1A, the transmission signal gain T1A ′ in the master unit 1A is The received signal gain R2 'of the base unit 1A is substantially the same as the received signal gain R2' of the base unit 1B, and the transmitted signal gain T2 'of the base unit 1B is substantially the same. The feedback gain β1 ′ of reflection due to impedance mismatch in the 2-wire to 4-wire conversion unit 15 of the base unit 1A is larger than the feedback gain β2 ′ due to acoustic coupling between the microphone 10 and the speaker 11 of the sub-base unit 1B (β2 '<Β1').
[0048]
Therefore, the loop gain G of the closed loop LP 1 ′ due to the reflection due to the impedance mismatch in the two-wire to four-wire converter 15._LP1′ (= Α ′ × T1A ′ × β1 ′ × R1A ′) is a loop gain G of the closed loop LP2 ′ due to the acoustic coupling of the sub master unit 1B._LP2(= Α ′ × T1A ′ × β2 ′ × R1A ′) (G_LP2’<G_LP ₁’). Here, feedback gains α and α ′ due to acoustic coupling of master unit 1A are constant (α = α ′) irrespective of the other communication terminal, and transmission signal gains T1A, T1A ′ and reception signal gain in master unit 1A. R1A and R1A 'are smaller when the communication partner is the sub master unit 1B (T1A' <T1A, R1A '<R1A).
[0049]
Therefore, the loop gain G of the closed loop LP1 'when the communication partner is the sub master unit 1B_LP1′ Is the loop gain G of the closed loop LP2 when the call partner is a loudspeaker type outside-the-home call terminal_LP2Since the value is smaller than the actual value (approximately one half in decibels), the base unit 1A (or the sub base unit 1B) and the loudspeaker type outside call terminal (doorphone handset 50 or lobby intercom 60) can be used. Echo and howling are sufficiently prevented even if the initial value at the time of the extension call between the home call terminals (the main unit 1A and the sub-base unit 1B or between the sub-base units 1B) is set smaller than the initial value at the time of the telephone call. You can do it. If the other party's call terminal is a handset type outside-the-home call terminal (alarm monitoring panel 80), the feedback gain β2 ′ due to the acoustic coupling of the handset 85 becomes smaller than the feedback gain β1 ′ of the base unit 1A. As in the case of the sub master unit 1B, echo and howling can be sufficiently prevented even if the initial value at the time of talking with the handset-type outside communication terminal (alarm monitoring panel 80) is set small.
[0050]
Therefore, in the present embodiment, as shown in FIG. 1, the optimal initial setting is made for each type of the other party, that is, for each of the doorphone handset 50, the lobby intercom 60, the alarm monitor 80, and the in-home communication terminal (the sub master unit 1B or the master unit 1A). Value Lt (0)₁~ Lt (0)₄A non-volatile memory 150 such as a flash memory is provided in the master unit 1A (also common to the sub-master unit 1B) to store the same.
[0051]
On the other hand, the DSP 200 that implements the call processing unit 17 includes a command processing unit 201 that receives a command from the main control unit 2 via a communication interface, interprets the command, and performs necessary processing. Initial value Lt (0) of the total loss amount suitable for₁~ Lt (0)₄And a parameter setting unit 202 that reads the data from the nonvolatile memory 150 and sets the software module that implements the call processing unit 17. However, the command processing unit 201 and the parameter setting unit 202 are also realized by controlling the hardware of the DSP 200 by software.
When a call is started, first, a call start request command is transmitted from the main control unit 2 to the DSP 200 via the communication interface, and the command processing unit 201 having received the call start request command interprets the content. , Parameters used in the call processing unit 17 (the initial value Lt (0) of the total loss amount)₁~ Lt (0)₄) Is given to the parameter setting unit 202. The parameter setting unit 202 sets the initial value Lt (0) corresponding to the call terminal of the other party based on the received address data.₁,... Are read from the non-volatile memory 150, and initialized for a software module that implements the total loss amount calculation unit 37d of the call processing unit 17. After these initializations are completed, the parameter setting unit 202 sets a flag indicating that the parameter initialization is completed, and upon receiving this flag setting, the command processing unit 201 sends a response command to the main control unit 2. The main control unit 2 that has transmitted and received the flag activates the call processing unit 17 to start a call.
[0052]
The present embodiment is configured as described above, and is applied to another in-home communication terminal (sub-base unit 1B or base unit 1A) for making an extension call or a handset type out-of-home communication terminal (alarm monitoring panel 80). Initial value Lt (0)₃, Lt (0)₄And the initial value Lt (0) for the loudspeaker-type out-of-home telephone terminal (doorphone slave unit 50 and lobby intercom 60)₁, Lt (0)₂Can be individually set to values that are necessary and sufficient to prevent unpleasant echoes and howling, thereby reducing the feeling of disconnection immediately after starting a call while preventing unpleasant echoes and howling. The time required to reach a state where simultaneous calls can be made can be reduced.
[0053]
(Embodiment 2)
If the total loss of the voice switch 37 is set to a value smaller than necessary due to a calculation error of the total loss calculator 37d, howling may occur. Therefore, in the present embodiment, the lower limit Lt is used for the total loss calculated by the total loss calculator 37d._ULAnd its lower limit Lt_ULCan be set individually for each call terminal of the other party.
[0054]
FIG. 10 shows a configuration in which a part of the in-home communication terminals (base unit 1A and sub-base unit 1B) according to the present embodiment is partially omitted. However, the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted.
[0055]
In the present embodiment, the optimum initial value Lt (0) for each type of call partner, that is, for each of the doorphone handset 50, the lobby intercom 60, the alarm monitor 80, and the in-home call terminal (the sub master unit 1B or the master unit 1A).₁~ Lt (0)₄And the lower limit Lt_UL1~ Lt_UL4Is stored in the non-volatile memory 150, and the parameter setting unit 202 reads the initial value Lt (0) from the non-volatile memory 150 in response to a command given from the main control unit 2 at the start of a call.₁, ... and lower limit Lt_UL1,... Are read and initialized for a software module that realizes the total loss amount calculation unit 37d of the call processing unit 17.
[0056]
In the process of the update mode shown in the flowchart of FIG. 15, the total loss amount calculation unit 37d reduces the previous total loss amount Lt (n-1) to the predetermined decrease amount Δd by the lower limit Lt in step S6._UL1,... Are larger than the previous total loss amount Lt (n-1), the value obtained by subtracting the decrease amount Δd from the previous total loss amount Lt (n-1) is set as the current total loss amount Lt (n), and the total loss amount Lt (n-1) is obtained. Is the lower limit Lt_UL1, ..., the lower limit Lt_UL1,... Are the current total loss amount Lt (n).
[0057]
As described above, in the present embodiment, the total loss amount Lt (n) calculated by the total loss amount calculation unit 37d is set to the lower limit Lt according to the other party's communication terminal._UL1Are set, the total loss amount of the voice switch 37 is not set to an unnecessarily small value due to a calculation error of the total loss amount calculation unit 37d, and howling can be prevented.
[0058]
(Embodiment 3)
In the present embodiment, when the total loss amount calculation unit 37d of the voice switch 37 in the in-home telephone terminal (the main unit 1A and the sub-main unit 1B) operates in the update mode, the other-party telephone terminal is the in-home telephone terminal. The feature is that the increase amount Δi1 and the decrease amount Δd1 are made larger than the increase amount Δi2 and the decrease amount Δd2 when the other party's communication terminal is an out-of-home communication terminal. However, the other system configuration and the configurations and operations of the master unit 1A and the auxiliary master unit 1B and the like are common to the first embodiment, and thus illustration and description are omitted.
[0059]
Next, a specific operation of the total loss amount calculation unit 37d in the update mode will be described with reference to a flowchart of FIG. However, description of points common to the operation of the conventional example shown in FIG. 14 will be omitted.
[0060]
Estimation processing of the acoustic feedback gain α and the line feedback gain β is performed at a predetermined sampling period from the point in time when the mode is shifted from the fixed mode to the update mode, and the estimated values α ′ (n) and β ′ (n) are calculated. (Step 1), from the product of these two estimated values α ′ (n) and β ′ (n) and the gain margin MG, the total loss amount required to keep the closed loop gain margin at MG [dB] The desired value Lr (n) is calculated (step 2), and the total loss calculating unit 37d calculates the n-th desired total loss amount Lr (n) and the previous (n-1) total loss amount Lt. (N-1) is compared (step 3). If the total loss desired value Lr (n) calculated this time is larger, it is determined whether or not the call is an extension call (that is, whether or not the other party is a home call terminal) (step 4). A value obtained by adding the increase amount Δi1 [dB] for the home communication terminal to the total loss amount Lt (n−1) is set as the total loss amount Lt (n) = Lt (n−1) + Δi1 at this time (step 5), and the extension If the call is not a call, the value obtained by adding the increase amount Δi2 [dB] for the outside-home communication terminal is set to the total loss amount Lt (n) = Lt (n−1) + Δi2 at this time (step 6). Here, the increase amount Δi1 for the in-home call terminal is set to, for example, about twice the increase amount Δi2 for the out-of-home call terminal.
[0061]
Then, the previous total loss amount Lt (n-1) is compared with the total loss desired value Lr (n) calculated this time (step 7), and if the current total loss amount desired value Lr (n) is smaller, It is determined whether or not the call is an extension call (step 8). If the call is an extension call, the value obtained by subtracting the reduction amount Δd1 [dB] for the home call terminal from the previous total loss amount Lt (n−1) is used as the total value of the current call. The loss amount Lt (n) = Lt (n−1) −Δd (step 9). If the call is not an extension call, the reduction amount Δd2 [dB] for the outside-home communication terminal from the previous total loss amount Lt (n−1). ] Is subtracted from the current total loss amount Lt (n) = Lt (n−1) −Δd2 (step 10). Here, the reduction amount Δd1 for the in-home communication terminal is set to, for example, about twice the reduction amount Δd2 for the out-of-home communication terminal. If the total loss amount desired value Lr (n) calculated this time in step 3 is equal to or less than the previous total loss amount Lt (n-1), the process proceeds to step 7 and the previous total loss amount Lt (n-1) is set. The total loss desired value Lr (n) calculated this time is compared. If the total loss desired value Lr (n) calculated this time in step 7 is equal to or greater than the previous total loss Lt (n-1), the process ends.
[0062]
Since the extension call is performed between the in-home communication terminal (the master unit 1A and the sub master unit 1B) having the voice switch 37, the loudspeaker type out-of-home communication terminal (the door phone slave unit 50) without the voice switch 37 is provided. In addition, the update speed of the total loss amount Lt (n) may be slower than that during a call with the lobby intercom 60), giving the caller an uncomfortable feeling. On the other hand, in the present embodiment, the total loss calculation unit 37d of the in-home communication terminals (master device 1A and sub-master device 1B) performs the above-described processing in the update mode, so that the extension call and the outside call can be performed. At the time of a call with the terminal, the update speed of the total loss amount Lt (n) by the total loss amount calculation unit 37d can be made substantially equal, and two-way simultaneous communication can be performed without giving the caller an uncomfortable feeling.
[0063]
(Embodiment 4)
FIG. 12 shows a home communication terminal (base unit 1A and two sub-base units 1B) in this embodiment.₁, 1B₂3) is a block diagram illustrating a schematic configuration of a main part of FIG. However, the same components as those of the master unit 1A and the auxiliary master unit 1B in the first embodiment are denoted by the same reference numerals, and the description or the description and illustration thereof are omitted. In this embodiment, the master unit 1A and the auxiliary master unit 1B₁, 1B₂Or secondary master 1B₁, 1B₂The apparent impedance of the communication path formed between the main unit 1A and the sub-unit 1B₁, 1B₂) And at least one of the out-of-home telephone terminals (lobby intercom 60 or alarm monitor 80) to provide an impedance adjusting means for making the apparent impedance of the communication path substantially equal to that of the sub master unit 1B.₁, 1B₂The feature is that it is provided with:
[0064]
That is, in an apartment house such as a condominium, the wiring length of the information trunk line L1 (the communication line La and the control line Lb) usually reaches several hundred meters to several kilometers, and the in-home communication terminals (the main unit 1A and the auxiliary main unit 1B).₁, 1B₂Or sub master unit 1B₁And Vice Master 1B₂) Is much longer than the wiring length (several tens of meters) of the communication line Ld between the parent device 1A and the sub-master device 1B.₁, 1B₂The difference between the impedance when looking at the alarm monitoring board 80 and the impedance when looking at the alarm monitoring board 80 (and the lobby intercom 60) from the master unit 1A becomes large, and the volume when talking with the alarm monitoring board 80 (or the lobby intercom 60) decreases. However, the present embodiment can prevent such a decrease in volume.
[0065]
Base unit 1A matches the impedance of the call path and the call circuit (such as call processing unit 17) in the extension call, the door phone call (call with door intercom unit 50), and the call with alarm monitor panel 80 (and lobby interphone 60). And a matching circuit 160 for performing the adjustment. A first switch 161 and a second switch 162 for forming a communication path are inserted in series between the matching circuit 160 and the communication switching unit 34, and a connection point Tc between the two switches 161 and 162 is provided. Sub-master 1B₁, 1B₂Communication line Le forming a communication path with₁, Le₂Is connected. Note that these two switches 161 and 162 are opened and closed by the main control unit 2. Call line Le₁, Le₂Consists of unbalanced pair wires, but may be parallel pair wires.
[0066]
On the other hand, sub master unit 1B₁, 1B₂Is a matching circuit 160 similar to that of base unit 1A, and communication line Le.₁, Le₂Switch 163 that opens and closes a signal path between the circuit and the matching circuit 160₁(Secondary master unit 1B₂Then, the fourth switch 163₂), The matching circuit 160 and the third switch 163₁(And the fourth switch 163₂), The first changeover switch 164 for switching the signal path in two ways.₁(Secondary master unit 1B₂Then, the second changeover switch 164₂), And a capacitor 165 as an impedance adjusting means. The condenser 165 is used for a communication path Le for an extension call.₁, Le₂The impedance equivalent to that of the communication line La is inserted into the communication line La, and is set to a capacitance value equivalent to a capacitance component corresponding to the total wiring length of the communication line La.
[0067]
Secondary master unit 1B₁, The first changeover switch 164₁Common contact Tz₁Is the third switch 163₁And two switching contacts Tx₁, Ty₁Are connected in parallel to the matching circuit 160, and one of the switching contacts Ty₁The capacitor 165 is connected between the ground and the ground. In addition, secondary master unit 1B₂, The second changeover switch 164₂Common contact Tz₂Is the fourth switch 163₂And two switching contacts Tx₂, Ty₂Are connected in parallel to the matching circuit 160, and one of the switching contacts Ty₂The capacitor 165 is connected between the ground and the ground. Note that the third and fourth switches 163₁, 163₂, First and second changeover switches 164₁, 164₂Is the secondary master unit 1B₁, 1B₂Is opened / closed and switched by the main control unit 2.
[0068]
Here, master unit 1A and sub master unit 1B₁, 1B₂, The first and second changeover switches 164 in each of the main control units 2₁,... And the communication paths formed between the in-home communication terminals or between the in-home communication terminals and the out-of-home communication terminals are shown in FIG. For example, when a call is made between the base unit 1A and the lobby intercom 60 (or the alarm monitoring panel 80), the base unit 1A turns on the first and second switches 161 and 162 and sets the call switching unit 34 to the switching contact Ta2. Switch to sub master unit 1B₁, 1B₂Now, the third and fourth switches 163₁, 163₂Is turned off, a communication path via the communication line La is formed. At this time, impedance matching is performed by the matching circuit 160 of the master unit 1A.
[0069]
In addition, master unit 1A and sub master unit 1B₁When an extension call is made between the parent device 1A, the main device 1A turns on the first switch 161 and turns off the second switch 162, and₁Then, the third switch 163₁Is turned on and the first switch 164 is turned on.₁Is the switching contact Ty₁Switch to sub master unit 1B₂Then, the fourth switch 163₂By turning off the telephone line Le₁, A communication path for an extension call is formed. At this time, since the capacitor 165 as the impedance adjusting means is connected in parallel to the communication path, the sub master unit 1B₁And sub-master 1B when looking at master 1A from₁When the lobby intercom 60 (or the alarm monitoring panel 80) is viewed from above, the impedance becomes substantially equal, and similarly, the master unit 1A to the subsidiary master unit 1B₁And the impedance when viewing the lobby intercom 60 (or the alarm monitoring panel 80) from the master unit 1A becomes substantially equal, so that the master unit 1A and the auxiliary master unit 1B₁Impedance matching can be performed without changing the matching constant of the matching circuit 160.
[0070]
Or sub master unit 1B₁, 1B₂In the case where an extension call is performed between the master unit 1A, the master unit 1A turns off the first and second switches 161 and 162, and the sub master unit 1B₁Then, the third switch 163₁Is turned on and the first switch 164 is turned on.₁Is the switching contact Ty₁Switch to sub master unit 1B₂Then, the fourth switch 163₂And the second changeover switch 164₂Is the switching contact Ty₂Switch to Talk line Le₁, Le₂, A communication path for an extension call is formed. At this time, one communication line Le₁Is connected in parallel with a capacitor 165 as an impedance adjusting means, so that one sub-base unit 1B₁(Or 1B₂) To the other sub master unit 1B₂(Or 1B₁) And the sub master unit 1B₁When the lobby intercom 60 (or the alarm monitoring panel 80) is viewed from above, the impedance becomes substantially equal, and the secondary master unit 1B₁, 1B₂Impedance matching can be performed without changing the matching constant of the matching circuit 160.
[0071]
As described above, in the present embodiment, the sub master unit 1B₁, 1B₂Since a capacitor 165 is provided as an impedance adjusting means, there is no need to change a matching constant for impedance matching according to the type of the other party's communication terminal, and the circuit configuration of the matching circuit 160 can be simplified.
[0072]
【The invention's effect】
The invention of claim 1 provides a plurality of out-of-home call terminals installed outside each dwelling unit in a multiple dwelling unit, and a plurality of in-home call terminals installed in each dwelling unit and communicating with the out-of-home call terminal. Wherein the out-of-home call terminal includes a loudspeaker type out-of-home call terminal having a microphone and a speaker for performing a loudspeaker call, and a handset type out-of-home call terminal having a handset. The terminal includes a microphone, a speaker, and a call processing means for performing processing for a loudspeaking call including suppression of howling and echo, and performs a loudspeaking call via a communication path formed between the terminal and the outside call terminal. A main unit, a microphone, a speaker, and a call processing unit for performing processing for a loudspeaker call including suppression of howling and echo, and a loudspeaker call with the out-of-home call terminal via the base unit. In a loudspeaker system for multiple dwelling units including one or a plurality of sub-base units for making an extension call with the base unit, the call processing means of the home communication terminal may include an acoustic echo generated by acoustic coupling between a microphone and a speaker. A first echo canceller for suppressing line echo, a second echo canceller for suppressing line echo caused by acoustic coupling or signal wraparound in a call terminal of the other party, and a first echo canceller provided between the first and second echo cancellers. A voice switch for reducing a loop gain of a closed loop formed by the acoustic echo path and the line echo path to suppress howling, wherein the first and second echo cancellers adaptively adjust the characteristics of the echo path. An adaptive filter for identifying, wherein the voice switch includes a transmitter-side loss insertion means for inserting a loss into a transmitter signal path. And a receiver-side loss insertion unit that inserts a loss in the signal path of the receiver, and an insertion loss amount controller that controls the amount of loss inserted from each of the loss insertion units on the transmitter side and the receiver side. The insertion loss control means estimates the acoustic feedback gain of a path that returns from the output point of the receiving-side loss inserting means to the input point of the transmitting-side loss inserting means via the acoustic echo path, and Estimate the line-side feedback gain of the path that returns from the output point of the talker-side loss insertion means to the input point of the receiver-side loss insertion means via the line echo path, and estimates the respective feedback gains on the acoustic and line sides. And a total loss amount calculating unit that calculates the total sum of the loss amounts to be inserted into the closed loop based on the transmission signal and the reception signal, and estimates a call state. The estimation result and the total loss amount calculation unit The transmitting side loss according to the calculated value An insertion loss amount distribution processing unit that determines the distribution of each insertion loss amount of the insertion unit and the receiving-side insertion loss unit, wherein the total loss amount calculation unit inserts into the closed loop based on the estimated value of each feedback gain. It has two operation modes: an update mode in which the sum of loss amounts to be calculated is calculated and adaptively updated, and a fixed mode in which the total loss amount is fixed to a predetermined initial value. And operates in the fixed mode until the second echo canceller sufficiently converges, and operates in the update mode during the period after the first and second echo cancellers sufficiently converge. The desired value of the total loss is obtained from the estimated values of the feedback gains on the sound side and the line side and the gain margin, and if the desired value of the total loss is larger than the total loss before the update, the total loss is increased by a predetermined amount. Just increased value Update, if the desired total loss amount is smaller than the total loss amount before the update, the total loss amount is updated to a value reduced by a predetermined decrease amount, and the desired total loss amount is the total loss amount before the update. If they are the same, the total loss amount is not updated by not updating the total loss amount. Non-volatile storage means for individually storing the initial value corresponding to the other party's call terminal, and a call between the other party's call terminal An initial value setting means for setting the initial value previously stored in the storage means to the total loss amount calculation unit is provided in the home call terminal, and the initial value corresponding to another home call terminal performing an extension call and Characterized in that the initial value corresponding to the handset type out-of-home call terminal is set to a value smaller than the initial value corresponding to the loudspeaker type out-of-home call terminal, and other in-home call terminals or handsets that perform extension calls Out-of-home call end By setting the initial value for, and the initial value for the loudspeaker-type out-of-home call terminal individually to values necessary and sufficient to prevent unpleasant echo and howling, immediately after the start of the call while preventing unpleasant echo and howling The feeling of disconnection can be reduced, and the time from the start of a call to the state where simultaneous two-way calls can be made can be reduced.
[0073]
According to a second aspect of the present invention, in the first aspect of the present invention, the lower limit value of the total loss amount that can be taken when the total loss amount calculating section is operating in the update mode is determined by the type of the home communication terminal and the out-of-home call. The in-home communication terminal is provided with setting means for individually setting according to the type of terminal, and the total loss amount may be set to an unnecessarily small value due to a calculation error of the total loss amount calculation unit or the like. It disappears and howling can be prevented.
[0074]
According to a third aspect of the present invention, in the first or second aspect of the invention, the increase amount and the decrease amount when the other party's call terminal is an in-home call terminal are determined by the increase when the other party's call terminal is an out-of-home call terminal. It is characterized in that it is larger than the volume and the reduction amount, and the update speed of the total loss amount by the total loss amount calculation unit can be made substantially equal at the time of the extension call and at the time of the call with the outside call terminal, making the caller feel uncomfortable You can make a two-way simultaneous call without giving.
[0075]
According to a fourth aspect of the present invention, in the first or second or third aspect of the invention, the sub master unit performs an extension call with another sub master unit installed in the house, and the impedance of the communication path is increased. Matching means for matching an apparent impedance of a communication path formed between the master unit and the sub master unit or the sub master unit between the in-home communication terminal and at least one out-of-home communication terminal. And an impedance adjusting means for making the impedance substantially equal to the apparent impedance of the communication path in the home communication terminal, eliminating the need to change the matching constant for impedance matching according to the type of the communication terminal of the other party. The structure of the matching means can be simplified.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a main part of a master unit and a sub master unit according to the first embodiment.
FIG. 2 is a system configuration diagram of the above system.
FIG. 3 is a block diagram showing a configuration of a master unit and a door phone slave unit according to the first embodiment;
FIG. 4 is a block diagram showing a configuration of a lobby intercom and an alarm monitoring panel in the above.
FIG. 5 is a block diagram showing a configuration of a sub master unit in the embodiment.
FIG. 6 is a block diagram showing a configuration of a call processing unit of a parent device and a sub-parent device in the above.
FIG. 7 is a flowchart for explaining the operation of the call processing unit in the embodiment.
FIG. 8 is an operation explanatory view of the above.
FIG. 9 is an operation explanatory view of the above.
FIG. 10 is a block diagram illustrating main parts of a master unit and a sub master unit according to the second embodiment.
FIG. 11 is a flowchart for explaining the operation of a total loss amount calculation unit according to the third embodiment.
FIG. 12 is a block diagram showing a main part of a master unit and a sub master unit in the fourth embodiment.
FIG. 13 is an operation explanatory diagram of the above.
FIG. 14 is a block diagram of a loudspeaker apparatus corresponding to a conventional example of a base unit and a sub-base unit.
FIG. 15 is a flowchart for explaining the operation of a total loss amount calculation unit in the voice communication device of the above.
[Explanation of symbols]
1A Housing information panel parent machine
1B Housing information panel vice parent machine
2 Main control unit
17 Call processing section
150 Non-volatile memory
200 DSP
201 Command processing unit
202 Parameter setting section

Claims

A plurality of out-of-home call terminals installed outside each dwelling unit in the apartment house, and a plurality of in-home call terminals installed in each dwelling unit and making a call with the out-of-home call terminal; The call terminals include a loudspeaker type outside call terminal having a microphone and a speaker for making a loudspeaker call, and a handset type outside call terminal having a handset. A base unit that includes a call processing unit that performs processing for a loudspeaker call including suppression of howling and echo, and performs a loudspeaker call via a call path formed between the remote terminal and a microphone and a speaker And a call processing means for performing processing for a loudspeaker call including suppression of howling and echo, and performs a loudspeaker call with the out-of-home call terminal via the base unit and between the base unit and the main unit. In the loudspeaker system for multiple dwelling units including one or a plurality of sub master units for making a line call, the call processing means of the in-home call terminal includes a first echo for suppressing an acoustic echo caused by an acoustic coupling between the microphone and the speaker. A canceller, a second echo canceller for suppressing a line echo caused by acoustic coupling or signal wraparound at a call terminal of the other party, and a sound echo path and a line echo provided between the first and second echo cancellers. A voice switch for reducing a loop gain of a closed loop formed by the path to suppress howling, and wherein the first and second echo cancellers include an adaptive filter for adaptively identifying characteristics of the echo path. The voice switch comprises: a transmission-side loss insertion means for inserting a loss into a transmission-side signal path; and a reception-side signal. A receiver-side loss insertion unit that inserts a loss into a path, and an insertion loss amount control unit that controls a loss amount inserted from each of the loss insertion units on the transmission side and the reception side, wherein the insertion loss amount control unit includes: Estimating the acoustic feedback gain of the path that returns from the output point of the receiving-side loss insertion means to the input point of the transmission-side loss insertion means via the acoustic echo path, Estimate the line-side feedback gain of the path that returns from the output point to the input point of the receiver-side loss insertion means via the line echo path, and based on the estimated values of the acoustic-side and line-side feedback gains, enter the closed loop. A total loss amount calculation unit for calculating the sum of the loss amounts to be inserted, and a call state estimated by monitoring the transmission signal and the reception signal, and according to the estimation result and the calculated value of the total loss amount calculation unit, Transmitter loss insertion means and receiver An insertion loss amount distribution processing unit that determines the distribution of each insertion loss amount of the insertion loss means, wherein the total loss amount calculation unit calculates a loss amount to be inserted into the closed loop based on the estimated value of each feedback gain. It has two operation modes, an update mode for calculating the sum and adaptively updating the sum, and a fixed mode for fixing the total loss amount to a predetermined initial value, and the first and second echoes from the start of the call with the other party's call terminal. During the period until the canceller sufficiently converges, it operates in the fixed mode, and during the period after the first and second echo cancellers sufficiently converge, it operates in the update mode. The desired value of the total loss is obtained from the estimated value of each feedback gain and the gain margin value, and if the desired value of the total loss is larger than the total loss before the update, the total loss is increased by a predetermined amount. Update and total loss If the desired value is smaller than the total loss amount before the update, the total loss amount is updated to a value reduced by a predetermined decrease amount, and if the total loss amount is equal to the total loss amount before the update, the total loss amount is updated. By not updating, the total loss amount is not calculated, and the non-volatile storage means for individually storing the initial value corresponding to the other party's communication terminal, and the storage means before the start of communication with the other party's communication terminal An initial value setting means for setting the stored initial value in the total loss amount calculation unit in the in-home telephone terminal, the initial value corresponding to another in-home telephone terminal performing an extension call, and a handset type out-of-home telephone call A loudspeaker communication system for a collective housing, wherein the initial value corresponding to the terminal is set to a value smaller than the initial value corresponding to the loudspeaker type outside call terminal.

The setting means for individually setting the lower limit value of the total loss amount that can be taken when the total loss amount calculating unit is operating in the update mode, according to the type of the in-home call terminal and the type of the out-of-home call terminal. 2. The voice communication system for a residential complex according to claim 1, wherein the system is provided in a home communication terminal.

2. The communication system according to claim 1, wherein the increase amount and the decrease amount when the other party's call terminal is an in-home call terminal are larger than the increase amount and the decrease amount when the other party's call terminal is an out-of-home call terminal. Or the voice call system for multi-family housing according to 2.

The sub-base unit is for making an extension call with another sub-base unit installed in the house, and includes a matching unit that matches the impedance of a communication path, and the main unit and the sub-base unit or the sub-base unit. Said impedance adjusting means for making the apparent impedance of the communication path formed between the devices substantially equal to the apparent impedance of the communication path formed between the in-home communication terminal and at least one of the out-of-home communication terminals. 4. The voice communication system for an apartment house according to claim 1, wherein the voice communication system is provided in an in-home communication terminal.