JP3881078B2

JP3881078B2 - Frequency estimation method, frequency estimation device, Doppler sonar and tidal meter

Info

Publication number: JP3881078B2
Application number: JP5866497A
Authority: JP
Inventors: 哲奥西
Original assignee: Furuno Electric Co Ltd
Current assignee: Furuno Electric Co Ltd
Priority date: 1997-03-13
Filing date: 1997-03-13
Publication date: 2007-02-14
Anticipated expiration: 2017-03-13
Also published as: JPH10253758A

Description

【０００１】
【発明の属する技術分野】
この発明は、信号周波数を離散フーリエ変換（ＤＦＴ）で求められたスペクトル分布に基づいて推定する周波数推定方法に関し、特に、信号のＳ／Ｎ比が小さいときに正確な周波数を推定できる周波数推定方法および周波数推定装置に関する。また、この発明は、反射波のＳ／Ｎ比が小さいときでも、正確な船速や潮流速度を測定することのできるドップラソナーおよび潮流計に関する。
【０００２】
【従来の技術】
ドップラソナーや潮流計などのドップラ効果を利用する機器では、受信した信号の周波数を、ＤＦＴ（実際の演算は高速フーリエ変換（ＦＦＴ）で行う。）を利用して推定する手法が適用されている。たとえば、田部井誠、上田光宏両氏による「ＦＦＴを用いた高精度周波数決定法」（電子通信学会情報誌、１９８７年５月号、７９８頁〜８０５頁）にＤＦＴを用いる高精度な周波数推定法（以下、田部井・上田法という）が提案されている。この田部井・上田法は、入力された時系列データにハニング窓を乗じたのちＤＦＴを行って得られる離散的な振幅スペクトルのうち、最大値をとるスペクトルと、これに隣り合う周波数での振幅スペクトルとの比に基づき、補間を行って真のピーク周波数ｆｐｅａｋを推定しようとするものであり（図６参照）、信号のＳ／Ｎ比がある程度以上大きければこの方法によって精度のよい周波数推定を行うことができる。
【０００３】
【発明が解決しようとする課題】
しかし、上記のような周波数推定法では、信号のＳ／Ｎ比が小さい場合には必ずしも真値に近い推定値を得ることができない。たとえば、白色雑音を含む正弦波信号をサンプリングして得られる系列（あるいは、この系列と適当な窓関数の系列との積）をｇ（ｎ）（ｎ＝０，１，２，…，Ｎ−１）とすると、ｇ（ｎ）のＤＦＴは次式で定義される。
【０００４】
【数１】

【０００５】
ここで、ｇ（ｎ）は振幅スペクトル｜Ｇ（ｋ）｜の特徴によって、次の（Ａ），（Ｂ）に分類される。
【０００６】
（Ａ）は｜Ｇ（ｋ）｜の最大値に信号成分が主として寄与する系列（以下、Ａ系列という）である。この系列の｜Ｇ（ｋ）｜の例を図５（Ａ）に示す。｜Ｇ（ｋ）｜が最大値をとるときの周波数番号ｋをｋｍａｘとすると、Ａ系列のｋｍａｘは信号の周波数ｆに対応した値をとり、両者の関係は次式で表される。
ｆ≒ｋｍａｘ（ｆｓ／Ｎ）
ここで、ｆｓはサンプリング周波数、Ｎはサンプリング点数である。
【０００７】
一方、（Ｂ）は｜Ｇ（ｋ）｜の最大値に信号成分が寄与しない系列（以下、Ｂ系列という）である。この系列の｜Ｇ（ｋ）｜の例を図５（Ｂ）に示す。Ｂ系列のｋｍａｘは信号の周波数ｆとは無関係な値をとる。すなわち、信号のスペクトルがノイズのスペクトルのなかに埋もれてしまい、ノイズ成分がピークをとってしまう。
【０００８】
Ｓ／Ｎ比が小さいほどＢ系列の生じる確率は高くなるため、従来のＤＦＴによる周波数推定法は、Ｓ／Ｎ比が小さいときには信頼性に欠ける問題点があった。
【０００９】
この発明は、ＤＦＴを用いた周波数推定法において、Ｓ／Ｎ比が小さい場合でも信頼できる周波数推定値を得ることができる周波数推定方法および周波数推定装置を提供することを目的とし、また、この発明は、精度のよい計測を可能としたドップラソナーおよび潮流計を提供することを目的とする。
【００１０】
【課題を解決するための手段】
この出願の請求項１の発明は、信号成分以外のノイズ成分が重畳された信号を入力し、該信号をサンプリングし、アナログ・ディジタル変換するとともに離散フーリエ変換してスペクトルを求め、このスペクトルに基づいて前記信号成分の周波数推定値を求める方法において、
前記信号の複数のサンプリング系列毎に前記スペクトルに基づいて周波数推定値を求め記憶するとともに、前記複数のサンプリング系列について前記スペクトルの振幅最大となる周波数番号の度数を計数し、
前記複数のサンプリング系列について求めた周波数推定値のうち前記度数が最大値をとる周波数番号に対応する周波数を含む一定の周波数範囲に含まれる周波数推定値だけを選別し、これら選別した周波数推定値の平均値を前記信号の周波数推定値として確定することを特徴とする。
【００１１】
この出願の請求項２の発明は、信号成分以外のノイズ成分が重畳された信号を入力し、該信号をサンプリングし、アナログ・ディジタル変換するとともに離散フーリエ変換してスペクトルを求め、このスペクトルに基づいて前記信号成分の周波数推定値を求める装置において、
前記信号の複数のサンプリング系列毎に前記スペクトルに基づいて周波数推定値を求め記憶する記憶手段と、
前記複数のサンプリング系列について前記スペクトルの振幅最大となる周波数番号の度数を計数し、前記複数のサンプリング系列について求めた周波数推定値のうち前記度数が最大値をとる周波数番号に対応する周波数を含む一定の周波数範囲に含まれる周波数推定値だけを選別し、これら選別した周波数推定値の平均値を前記信号の周波数推定値として確定する周波数確定手段と、
を備えたことを特徴とする。
【００１２】
この出願の請求項３の発明は、送波器から送波した超音波信号の対象物からの反射波を受波器で受波し、該反射波のドップラシフト周波数に基づいて前記対象物に対する船舶の相対速度を測定するドップラソナーにおいて、
前記反射波をディジタル信号に変換するアナログ・ディジタル変換手段と、
該ディジタル信号を離散フーリエ変換してスペクトルを求めるスペクトル分解手段と、
該スペクトルに基づいて前記反射波のドップラシフト周波数を推定する周波数推定手段と、
前記スペクトルの振幅最大となる周波数番号である最大振幅番号を検出する最大振幅番号検出手段と、
複数回の送波によって得られる複数個の反射波について、前記最大振幅番号の度数を求めるとともに、該度数が最大値をとる周波数番号に対応する周波数を含む一定の周波数範囲に含まれる周波数推定値だけを選別し、これら選別された周波数推定値の平均値を確定周波数推定値とする周波数確定手段と、
を備えたことを特徴とする。
【００１３】
この出願の請求項４の発明は、送波器から送波した超音波信号の水底および水塊からの反射波を受波器で受波し、各反射波のドップラシフト周波数に基づいて対地速度および対水速度を求め、これらを減算することによって潮流速度を求める潮流計において、
前記水底からの反射波をディジタル信号に変換する水底反射波アナログ・ディジタル変換手段と、
該ディジタル信号を離散フーリエ変換して水底反射波のスペクトルを求める水底反射波スペクトル分解手段と、
該水底反射波のスペクトルに基づいて前記水底反射波の周波数を推定する水底反射波周波数推定手段と、
前記水底反射波のスペクトルの振幅最大となる周波数番号である水底反射波最大振幅番号を検出する水底反射波最大振幅番号検出手段と、
複数回の送波によって得られる複数個の水底反射波について、前記水底反射波最大振幅番号の度数を求めるとともに、該度数が最大値をとる周波数番号に対応する周波数を含む一定の周波数範囲に含まれる水底反射波周波数推定値だけを選別し、これら選別された水底反射波周波数推定値の平均値を確定水底反射波周波数推定値とする水底反射波周波数確定手段と、
前記水塊からの反射波をディジタル信号に変換する水塊反射波アナログ・ディジタル変換手段と、
該ディジタル信号を離散フーリエ変換して水塊反射波のスペクトルを求める水塊反射波スペクトル分解手段と、
該水塊反射波のスペクトルに基づいて前記水塊反射波の周波数を推定する水塊反射波周波数推定手段と、
前記水塊反射波のスペクトルの振幅最大となる周波数番号である水塊反射波最大振幅番号を検出する水塊反射波最大振幅番号検出手段と、
複数回の送波によって得られる複数個の水塊反射波について、前記水塊反射波最大振幅番号の度数が最大値をとる周波数番号に対応する周波数を含む一定の周波数範囲に含まれる水塊反射波周波数推定値だけを選別し、これら選別された水塊反射波周波数推定値の平均値を確定水塊反射波周波数推定値とする水塊反射波周波数確定手段と、
を備えたことを特徴とする。
【００１４】
この発明では、周波数を推定する信号をＡ／Ｄ変換→ＤＦＴ変換し、その周波数を推定する。そしてこの推定周波数、および、振幅スペクトル｜Ｇ（ｋ）｜が最大値をとるときの周波数番号ｋｍａｘを記憶する。これを上記周波数を推定する信号の複数のサンプリング系列について繰り返し実行する。ここで、各サンプリング系列は、複数のサンプリング値からなる時系列データである。この繰り返し処理によるデータの蓄積ののち、ｋｍａｘのヒストグラムをとり、その最頻値（モード）ｋｍａｘ−ｍｏｄｅを割り出す。最頻値とは、最も出現回数の多かったｋｍａｘの値である。この周波数番号ｋｍａｘ−ｍｏｄｅに対応する周波数、

を求め、前記複数求められた周波数推定値のうち、この周波数ｆｏを含む一定の周波数範囲に含まれるものを選別してその平均値を求める。これを最終的な周波数推定値として採用する。
【００１５】
このように、周波数ｆｏに近い値をとる周波数推定値を選別・平均化して最終的な周波数推定値を求めるようにしたことにより、信号のＳ／Ｎ比が小さくても、信号のスペクトルレベルがノイズよりも優勢なときの推定値のみを選別し、それ以外のときの推定値を無視するため、正確な周波数推定値を得ることができる。
【００１６】
【発明の実施の形態】
図面を参照してこの発明について説明する。図１はこの発明が適用される潮流計を搭載した船舶を示す図、図２は同潮流計のブロック図である。船舶１の船底には３つの超音波振動子である送受波器１１が設けられており、３方向に超音波ビーム２を発するとともに、該超音波ビーム２の海底および水塊からの反射波を受信する。３つの送受波器１１ａ，１１ｂ，１１ｃにはそれぞれトラップ１２ａ，１２ｂ，１２ｃを介して送信駆動部１９が接続されている。送信駆動部は送受波器１１に対して電気的なパルス信号を繰り返し供給する。また、各トラップ１２ａ〜１２ｃには受信系統として周波数変換器１３ａ〜１３ｃ，増幅器１４ａ〜１４ｃおよびＡ／Ｄ変換器１５ａ〜１５ｃが接続されている。前記トラップ１２は送信駆動部１９から入力される高レベルの信号を受信系統に回り込まないようにし、且つ、送受波器１１が受信した低レベルの反射信号を受信系統に伝達する回路である。トラップ１２に接続されている周波数変換器１３は受信信号の周波数を低い値に変換する。周波数を下げることにより、後段のＡ／Ｄ変換器１５のサンプリングレートを下げることができる。なお、周波数を変換してもドップラシフトは保存される。周波数変換された信号は増幅器１４で増幅されたのちＡ／Ｄ変換器１５でディジタルデータ（量子化された離散時間信号）に変換される。このディジタルデータは演算部１６に入力される。
【００１７】
また、前記周波数変換器１３から出力された信号は海底検出部２０に入力される。海底検出部２０は各送受波器１１ａ〜１１ｃが受信した反射波形の時間差に基づいて海底深度を割り出す。
【００１８】
演算部１６は送受波器１１が受信した信号（周波数変換されディジタルデータ化された信号）のうち、海底深度に対応する部分の波形と、１または複数の潮流測定深度に対応する部分の波形を抽出し、この抽出された波形をＤＦＴすることによってその周波数を推定する。この各測定深度毎の反射波の周波数推定は各送受波器１１ａ〜１１ｃの受信信号毎に別々に行われる。そして、この各測定深度の推定周波数に基づいて、船舶の海底に対する速度ベクトルである対地速度および船舶の所定深度の水塊に対する速度ベクトルである対水速度を求める。そしてこの対地速度から対水速度を減算することによって、前記水塊の海底に対する速度ベクトルである潮流速度を算出する。算出された対地速度および潮流速度は表示部１７に入力される。表示部１７はこの対地速度，潮流速度を所定の表示形式（たとえば、矢印ベクトル表示や数値・方位表示）に変換し、画面上のパターンに展開してモニタ１８に出力する。モニタ１８がこのパターンを表示することにより、ユーザが対地速度，潮流速度を認識することができる。
【００１９】
図３〜図５を参照して、演算部１６における反射波受信信号の周波数推定処理動作を説明する。図３は上記周波数推定処理動作を示すフローチャート、図４は前記演算部１６に設定される処理結果蓄積テーブルを示す図、図５はＤＦＴによって求められる振幅スペクトルおよびｋｍａｘのヒストグラムを示す図である。
【００２０】
図３の処理動作は、超音波パルスを繰り返し送受信して、各受信信号のサンプリング系列を処理する動作であり、上記各送受波器１１ａ〜１１ｃ毎に、かつ、海底深度および１または複数の測定深度に対応する部分の波形毎に処理動作が繰り返し実行される。
まず、データ収集回数すなわち送受信した超音波パルスの送受信回数を示すカウンタｍに１をセットする（ｓ１０）。潮流測定対象となる測定深度あるいは海底深度に対応する波形区間のサンプリングデータをＮサンプル（ｘ（０）〜ｘ（Ｎ−１））読み込み（ｓ１１）、このサンプリングデータ列に対して、下式に示すようにハニング窓などの適当な窓関数ｗ（ｎ）を乗じる（ｓ１２）。
【００２１】
ｇ（ｎ）＝ｗ（ｎ）・ｘ（ｎ）（ｎ＝０，１，…，Ｎ−１）
このデータ列｛ｇ（ｎ）｝を〔数１〕式にしたがってＤＦＴ変換する（ｓ１３）。このＤＦＴ変換にはＦＦＴアルゴリズムを用いればよい。
【００２２】
このＤＦＴ変換されたＧ（ｋ）の値を用いて周波数推定値ｆ（ｍ）を算出する（ｓ１４）。この周波数推定法にはいくつかの手法があるが、たとえば、上述した田部井・上田法を用いればよい。この周波数推定値は演算部１６に設定されている処理結果蓄積テーブル（図４参照）に記録される。そして振幅スペクトル｜Ｇ（ｋ）｜が最大値をとるときの周波数番号ｋを求め、ｋｍａｘ（ｍ）とする（ｓ１５）。このｋｍａｘ（ｍ）も前記処理結果蓄積テーブルに記録される。このとき受信信号のＳ／Ｎ比が大きければ、スペクトルは図５（Ａ）のようになり、ｋｍａｘは信号周波数に対応する値をとるが、Ｓ／Ｎ比が小さければ、スペクトルは図５（Ｂ）のようになり、ｋｍａｘは信号周波数と無関係な値をとることがある。以上のデータ収集処理をＭ個の受信信号について繰り返し行い（ｓ１６→ｓ１１）、そののちｓ１８の集計動作に進む。
【００２３】
ｓ１８では、処理結果蓄積テーブルに記憶されているＭ個のｋｍａｘ（ｍ）（ｍ＝１〜Ｍ）のヒストグラムをとり（図５（Ｃ）参照）、最頻値ｋｍａｘ−ｍｏｄｅを割り出す。そして、この最頻値に対応するｆｏ＝ｋｍａｘ−ｍｏｄｅ・（ｆｓ／Ｎ）を算出し、上記ｓ１４の動作で算出されたＭ個の周波数推定値ｆ（ｍ）（ｍ＝１〜Ｍ）のうち、上記周波数ｆｏを含む一定の周波数範囲に含まれる値だけを選別する（ｓ１９）。そして、この選別された周波数推定値の平均値を算出し、この平均値を受信信号の周波数として出力する（ｓ２０）。
【００２４】
現実に送受波器１１から送信される超音波ビームはある程度の広がり角をもつため、たとえば、海底に対する入射角は一定ではなく、海底での反射波が受けるドップラシフトには幅がある。このビームの広がり角に対応する周波数範囲内の推定値を選別することにより、有効な推定値を漏れなく利用できる。
【００２５】
以上のように、この発明の方式を用いることにより、受信信号のスペクトルがノイズのスペクトルに見え隠れする程度にＳ／Ｎ比が小さいときでも、複数個の推定値から有効なものを選別することによって、正確な周波数を推定できる。そして、これを潮流計やドップラソナーに適用することにより、正確なドップラシフト周波数を割り出すことができ、精度のよい速度検出をすることができる。
【００２６】
なお、上記実施形態では、１つの超音波振動子を送波器および受波器として用いているが、送波器と受波器を別々に設けてもよい。
【００２７】
なお、上記実施形態は、船舶に搭載される潮流計やドップラソナーに適用した例を説明したが、この発明は、この分野に限定されることなく、周波数推定の処理を含む種々の計測分野に適用できるものである。
【００２８】
【発明の効果】
以上のようにこの発明によれば、ノイズの影響によって生じうる、真値とは大きく異なる推定値を無視できるため、信号のＳ／Ｎ比が小さくても正確な周波数推定値を得ることができる。
【図面の簡単な説明】
【図１】この発明が適用される潮流計が搭載された船舶を示す図
【図２】同潮流計のブロック図
【図３】同潮流計の演算部の動作を示すフローチャート
【図４】同演算部に設定される処理結果蓄積テーブルを示す図
【図５】同演算部が算出するスペクトルおよびヒストグラムの例を示す図
【図６】ＤＦＴに基づく一般的な周波数推定手法を説明する図
【符号の説明】
１１ａ〜１１ｃ…送受波器、１６…演算部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a frequency estimation method for estimating a signal frequency based on a spectrum distribution obtained by discrete Fourier transform (DFT), and more particularly, a frequency estimation method capable of estimating an accurate frequency when a signal S / N ratio is small. And a frequency estimation apparatus. The present invention also relates to a Doppler sonar and a tide meter that can accurately measure ship speed and tidal velocity even when the S / N ratio of reflected waves is small.
[0002]
[Prior art]
In a device using the Doppler effect such as a Doppler sonar or a tide meter, a method of estimating the frequency of the received signal using DFT (actual calculation is performed by fast Fourier transform (FFT)) is applied. . For example, “High-precision frequency determination method using FFT” by Makoto Tabei and Mitsuhiro Ueda (Electronic Communication Society Information Magazine, May 1987, pages 798 to 805), a high-precision frequency estimation method using DFT ( Hereinafter, the Tabei and Ueda methods) have been proposed. This Tabei and Ueda method uses the spectrum that takes the maximum value among the discrete amplitude spectra obtained by multiplying the input time-series data by the Hanning window and then DFT, and the amplitude spectrum at the adjacent frequency. Is used to estimate the true peak frequency fpeak (see FIG. 6), and if the signal S / N ratio is larger than a certain level, accurate frequency estimation is performed by this method. be able to.
[0003]
[Problems to be solved by the invention]
However, the frequency estimation method as described above cannot always obtain an estimated value close to the true value when the S / N ratio of the signal is small. For example, g (n) (n = 0, 1, 2,..., N−) is a sequence obtained by sampling a sine wave signal including white noise (or a product of this sequence and an appropriate window function sequence). 1), the DFT of g (n) is defined by the following equation.
[0004]
[Expression 1]

[0005]
Here, g (n) is classified into the following (A) and (B) according to the characteristics of the amplitude spectrum | G (k) |.
[0006]
(A) is a sequence in which a signal component mainly contributes to the maximum value of | G (k) | (hereinafter referred to as A sequence). An example of | G (k) | of this series is shown in FIG. If the frequency number k at which | G (k) | takes the maximum value is kmax, kmax of the A series takes a value corresponding to the frequency f of the signal, and the relationship between them is expressed by the following equation.
f≈kmax (fs / N)
Here, fs is a sampling frequency, and N is the number of sampling points.
[0007]
On the other hand, (B) is a sequence in which a signal component does not contribute to the maximum value of | G (k) | (hereinafter referred to as B sequence). An example of | G (k) | of this series is shown in FIG. B series kmax takes a value unrelated to the frequency f of the signal. That is, the signal spectrum is buried in the noise spectrum, and the noise component takes a peak.
[0008]
The smaller the S / N ratio, the higher the probability that a B sequence will occur. Therefore, the conventional DFT frequency estimation method has a problem of lack of reliability when the S / N ratio is small.
[0009]
It is an object of the present invention to provide a frequency estimation method and a frequency estimation apparatus capable of obtaining a reliable frequency estimation value even when the S / N ratio is small in a frequency estimation method using DFT. An object of the present invention is to provide a Doppler sonar and a tide meter that enable accurate measurement.
[0010]
[Means for Solving the Problems]
The invention of claim 1 of this application, receives the signal noise components other than the signal component is superimposed, by sampling the signal to obtain the spectrum by discrete Fourier transform as well as analog-to-digital converter, to the spectrum In a method for obtaining a frequency estimate of the signal component based on:
Stores determined frequency estimate based on the spectral plurality of each sampling-series of said signal, counting the frequency of the amplitude becomes maximum frequency number of the spectrum of said plurality of sampling sequence,
Of the frequency estimation values obtained for the plurality of sampling series, only the frequency estimation values included in a certain frequency range including the frequency corresponding to the frequency number at which the frequency has the maximum value are selected, and the frequency estimation values thus selected are selected. An average value is determined as a frequency estimation value of the signal .
[0011]
The invention of claim 2 of this application inputs a signal on which a noise component other than the signal component is superimposed, samples the signal, performs analog-digital conversion and obtains a spectrum by discrete Fourier transform, and based on this spectrum In the apparatus for obtaining the frequency estimation value of the signal component
Storage means for obtaining and storing a frequency estimation value based on the spectrum for each of a plurality of sampling series of the signal;
The frequency of the frequency number having the maximum amplitude of the spectrum is counted for the plurality of sampling sequences, and the frequency number corresponding to the frequency number at which the frequency has the maximum value among the frequency estimation values obtained for the plurality of sampling sequences is constant. Frequency determination means for selecting only the frequency estimation values included in the frequency range, and determining the average value of the selected frequency estimation values as the frequency estimation value of the signal;
It is provided with.
[0012]
In the invention of claim 3 of this application, the reflected wave from the object of the ultrasonic signal transmitted from the transmitter is received by the wave receiver , and the object is detected based on the Doppler shift frequency of the reflected wave. In Doppler sonar that measures the relative speed of a ship,
Analog-to-digital conversion means for converting the reflected wave into a digital signal;
Spectral decomposition means for obtaining a spectrum by performing a discrete Fourier transform on the digital signal;
Frequency estimation means for estimating a Doppler shift frequency of the reflected wave based on the spectrum;
And the maximum amplitude number detecting means for detecting a maximum amplitude number is the frequency number to which the amplitude maximum of the spectral,
For multiple plurality of reflected wave thus obtained transmitting the portions to determine the frequency of pre-Symbol maximum amplitude number, frequency included in a predetermined frequency range including a frequency該度number corresponding to the frequency number having the maximum value A frequency determining means for selecting only the estimated values and using the average value of the selected frequency estimated values as a determined frequency estimated value;
It is provided with.
[0013]
In the invention of claim 4 of this application, the reflected wave from the bottom and the water mass of the ultrasonic signal transmitted from the transmitter is received by the receiver , and the ground speed is based on the Doppler shift frequency of each reflected wave. And in the tide meter to find the tidal velocity by finding the water velocity and subtracting these,
Water bottom reflected wave analog-digital conversion means for converting the reflected wave from the water bottom into a digital signal;
Water bottom reflected wave spectrum decomposing means for obtaining a spectrum of the water bottom reflected wave by performing a discrete Fourier transform on the digital signal;
And water bottom reflection wave frequency estimating means for estimating the frequency of said water bottom reflection wave based on the spectrum of the water bottom reflection waves,
Water bottom reflected wave maximum amplitude number detecting means for detecting a water bottom reflected wave maximum amplitude number which is a frequency number which is the maximum amplitude of the spectrum of the water bottom reflected wave;
For multiple plurality of water bottom reflection wave obtained by transmitting the portions to determine the frequency of pre-Symbol underwater reflected wave maximum amplitude numbers, in a certain frequency range including a frequency該度number corresponding to the frequency number having the maximum value Only the water bottom reflected wave frequency estimation value included is selected, and the bottom reflected wave frequency determining means using the average value of the selected water bottom reflected wave frequency estimated values as the determined water bottom reflected wave frequency estimated value,
A water mass reflected wave analog-digital conversion means for converting a reflected wave from the water mass into a digital signal;
Water mass reflected wave spectrum decomposing means for obtaining a spectrum of the water mass reflected wave by discrete Fourier transform of the digital signal;
A water body reflected wave frequency estimating means for estimating the frequency of the water mass reflected wave based on the spectrum of the water mass reflected waves,
Water mass reflected wave maximum amplitude number detecting means for detecting a water mass reflected wave maximum amplitude number which is a frequency number which is the maximum amplitude of the spectrum of the water mass reflected wave ;
For multiple plurality of water mass reflected wave obtained by transmitting the water mass which the frequency of pre-Symbol water masses reflected wave maximum amplitude number is included in a predetermined frequency range including a frequency corresponding to the frequency number having the maximum value Only the reflected wave frequency estimated value is selected, and the water mass reflected wave frequency determining means using the average value of the selected water reflected wave frequency estimated values as the determined water reflected wave frequency estimated value,
It is provided with.
[0014]
In the present invention, the frequency estimation signal is subjected to A / D conversion → DFT conversion, and the frequency is estimated. Then, the estimated frequency and the frequency number kmax when the amplitude spectrum | G (k) | takes the maximum value are stored. This is repeated for a plurality of sampling sequences of the signal for estimating the frequency. Here, each sampling series is time series data composed of a plurality of sampling values. After accumulating data by this iterative process, a histogram of kmax is taken and its mode value (mode) kmax-mode is determined. The mode value is a value of kmax that has the highest number of appearances. The frequency corresponding to this frequency number kmax-mode,

Among the plurality of obtained frequency estimation values, those included in a certain frequency range including this frequency fo are selected and an average value thereof is obtained. This is adopted as the final frequency estimation value.
[0015]
As described above, by selecting and averaging the frequency estimation values having values close to the frequency fo to obtain the final frequency estimation value, even if the signal S / N ratio is small, the spectrum level of the signal is high. Since only the estimated values when dominant over the noise are selected and the estimated values at other times are ignored, an accurate frequency estimated value can be obtained.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
The present invention will be described with reference to the drawings. FIG. 1 is a view showing a ship equipped with a tide meter to which the present invention is applied, and FIG. 2 is a block diagram of the tide meter. A transmitter / receiver 11 as three ultrasonic transducers is provided on the bottom of the ship 1, and the ultrasonic beam 2 is emitted in three directions and the reflected wave from the sea floor and water mass of the ultrasonic beam 2 is reflected. Receive. A transmission driver 19 is connected to the three

transducers

11a, 11b, and 11c via

traps

12a, 12b, and 12c, respectively. The transmission driver repeatedly supplies an electrical pulse signal to the transducer 11. Further, frequency converters 13a to 13c, amplifiers 14a to 14c, and A / D converters 15a to 15c are connected to the traps 12a to 12c as reception systems. The trap 12 is a circuit that prevents a high-level signal input from the transmission drive unit 19 from entering the reception system and transmits a low-level reflected signal received by the transducer 11 to the reception system. A frequency converter 13 connected to the trap 12 converts the frequency of the received signal to a low value. By reducing the frequency, the sampling rate of the A / D converter 15 at the subsequent stage can be reduced. Even if the frequency is converted, the Doppler shift is preserved. The frequency-converted signal is amplified by the amplifier 14 and then converted to digital data (quantized discrete time signal) by the A / D converter 15. This digital data is input to the arithmetic unit 16.
[0017]
The signal output from the frequency converter 13 is input to the seabed detection unit 20. The seabed detection unit 20 determines the seabed depth based on the time difference between the reflected waveforms received by the transducers 11a to 11c.
[0018]
The calculation unit 16 generates a waveform of a portion corresponding to the seabed depth and a waveform of a portion corresponding to one or more tidal current measurement depths of the signal (frequency converted digital signal) received by the transducer 11. The frequency is estimated by performing DFT on the extracted waveform. The frequency estimation of the reflected wave for each measurement depth is performed separately for each received signal of each of the transducers 11a to 11c. Then, based on the estimated frequency of each measurement depth, a ground speed that is a speed vector with respect to the seabed of the ship and a water speed that is a speed vector with respect to a water mass at a predetermined depth of the ship are obtained. Then, by subtracting the water velocity from the ground velocity, a tidal velocity that is a velocity vector with respect to the seabed of the water mass is calculated. The calculated ground speed and tidal current speed are input to the display unit 17. The display unit 17 converts the ground speed and tidal current speed into a predetermined display format (for example, arrow vector display or numerical value / orientation display), develops the pattern on the screen, and outputs it to the monitor 18. When the monitor 18 displays this pattern, the user can recognize the ground speed and the tidal current speed.
[0019]
With reference to FIGS. 3 to 5, the frequency estimation processing operation of the reflected wave reception signal in the calculation unit 16 will be described. FIG. 3 is a flowchart showing the frequency estimation processing operation, FIG. 4 is a diagram showing a processing result accumulation table set in the arithmetic unit 16, and FIG. 5 is a diagram showing an amplitude spectrum and kmax histogram obtained by DFT.
[0020]
The processing operation of FIG. 3 is an operation of processing the sampling sequence of each received signal by repeatedly transmitting and receiving ultrasonic pulses, and for each of the above-mentioned transducers 11a to 11c, and the seabed depth and one or more measurements. The processing operation is repeatedly executed for each waveform of the portion corresponding to the depth.
First, 1 is set to a counter m indicating the number of times of data collection, that is, the number of times of transmission / reception of transmitted / received ultrasonic pulses (s10). N samples (x (0) to x (N-1)) of sampling data of the waveform section corresponding to the measurement depth or the seafloor depth to be measured for tidal currents are read (s11). As shown, an appropriate window function w (n) such as a Hanning window is multiplied (s12).
[0021]
g (n) = w (n) .x (n) (n = 0, 1,..., N−1)
This data string {g (n)} is DFT transformed according to the formula [1] (s13). An FFT algorithm may be used for this DFT conversion.
[0022]
A frequency estimation value f (m) is calculated using the value of G (k) subjected to the DFT transformation (s14). There are several methods for this frequency estimation method. For example, the Tabei / Ueda method described above may be used. This estimated frequency value is recorded in a processing result accumulation table (see FIG. 4) set in the calculation unit 16. Then, the frequency number k when the amplitude spectrum | G (k) | takes the maximum value is obtained and set as kmax (m) (s15). This kmax (m) is also recorded in the processing result accumulation table. If the S / N ratio of the received signal is large at this time, the spectrum is as shown in FIG. 5A, and kmax takes a value corresponding to the signal frequency. If the S / N ratio is small, the spectrum is as shown in FIG. B), and kmax may take a value unrelated to the signal frequency. The above data collection process is repeated for M received signals (s16 → s11), and then the process proceeds to the counting operation of s18.
[0023]
In s18, a histogram of M kmax (m) (m = 1 to M) stored in the processing result accumulation table is taken (see FIG. 5C), and the mode value kmax-mode is determined. Then, fo = kmax−mode · (fs / N) corresponding to the mode value is calculated, and M frequency estimated values f (m) (m = 1 to M) calculated in the operation of s14 are calculated. Among them, only values included in a certain frequency range including the frequency fo are selected (s19). Then, an average value of the selected frequency estimation values is calculated, and this average value is output as the frequency of the received signal (s20).
[0024]
Since the ultrasonic beam actually transmitted from the transmitter / receiver 11 has a certain divergence angle, for example, the incident angle with respect to the seabed is not constant, and the Doppler shift received by the reflected wave at the seabed has a width. By selecting an estimated value in the frequency range corresponding to the beam divergence angle, an effective estimated value can be used without omission.
[0025]
As described above, by using the method of the present invention, even when the S / N ratio is small enough to make the received signal spectrum appear and hide as a noise spectrum, the effective signal is selected from a plurality of estimated values. It is possible to estimate an accurate frequency. By applying this to a tide meter or Doppler sonar, an accurate Doppler shift frequency can be determined, and accurate speed detection can be performed.
[0026]
In the above embodiment, one ultrasonic transducer is used as the transmitter and the receiver, but the transmitter and the receiver may be provided separately.
[0027]
In the above embodiment, an example in which the present invention is applied to a tide meter and a Doppler sonar mounted on a ship has been described. However, the present invention is not limited to this field, and is applicable to various measurement fields including frequency estimation processing. Applicable.
[0028]
【The invention's effect】
As described above, according to the present invention, it is possible to ignore an estimated value greatly different from the true value, which can be caused by the influence of noise. Therefore, an accurate frequency estimated value can be obtained even if the S / N ratio of the signal is small. .
[Brief description of the drawings]
FIG. 1 is a diagram showing a ship equipped with a tide meter to which the present invention is applied. FIG. 2 is a block diagram of the tide meter. FIG. 3 is a flowchart showing an operation of a calculation unit of the tide meter. FIG. 5 is a diagram illustrating an example of a processing result accumulation table set in the calculation unit. FIG. 5 is a diagram illustrating an example of a spectrum and a histogram calculated by the calculation unit. FIG. 6 is a diagram illustrating a general frequency estimation method based on DFT. Explanation of]
11a to 11c: transducers, 16: arithmetic unit

Claims

A signal on which a noise component other than the signal component is superimposed is input, the signal is sampled, analog-to-digital converted and a discrete Fourier transform is performed to obtain a spectrum, and a frequency estimation value of the signal component is obtained based on the spectrum. In the method
Obtaining and storing a frequency estimation value based on the spectrum for each of a plurality of sampling series of the signal, and counting the frequency number frequency that is the maximum amplitude of the spectrum for the plurality of sampling series,
Of the frequency estimation values obtained for the plurality of sampling series, only the frequency estimation values included in a certain frequency range including the frequency corresponding to the frequency number at which the frequency has the maximum value are selected, and the frequency estimation values thus selected are selected. A frequency estimation method, wherein an average value is determined as a frequency estimation value of the signal.

A signal on which a noise component other than the signal component is superimposed is input, the signal is sampled, analog-to-digital converted and a discrete Fourier transform is performed to obtain a spectrum, and a frequency estimation value of the signal component is obtained based on the spectrum. In the device
Storage means for obtaining and storing a frequency estimation value based on the spectrum for each of a plurality of sampling series of the signal;
The frequency of the frequency number having the maximum amplitude of the spectrum is counted for the plurality of sampling sequences, and the frequency number corresponding to the frequency number at which the frequency has the maximum value among the frequency estimation values obtained for the plurality of sampling sequences is constant. Frequency determination means for selecting only the frequency estimation values included in the frequency range, and determining the average value of the selected frequency estimation values as the frequency estimation value of the signal;
A frequency estimation apparatus comprising:

In a Doppler sonar that receives a reflected wave from an object of an ultrasonic signal transmitted from a transmitter by a wave receiver and measures a relative velocity of the ship with respect to the object based on a Doppler shift frequency of the reflected wave ,
Analog-to-digital conversion means for converting the reflected wave into a digital signal;
Spectral decomposition means for obtaining a spectrum by performing a discrete Fourier transform on the digital signal;
Frequency estimation means for estimating a Doppler shift frequency of the reflected wave based on the spectrum;
Maximum amplitude number detecting means for detecting a maximum amplitude number that is a frequency number that is the maximum amplitude of the spectrum;
For a plurality of reflected waves obtained by a plurality of times of transmission, the frequency of the maximum amplitude number is obtained, and a frequency estimation value included in a certain frequency range including the frequency corresponding to the frequency number at which the frequency has the maximum value Frequency determining means for selecting only the frequency estimated values, and using the average value of the selected frequency estimated values as a determined frequency estimated value,
Doppler sonar characterized by comprising.

The reflected wave from the bottom and water mass of the ultrasonic signal transmitted from the transmitter is received by the receiver, and the ground speed and water speed are calculated based on the Doppler shift frequency of each reflected wave, and these are subtracted. In the tide meter to obtain the tidal velocity by
Water bottom reflected wave analog-digital conversion means for converting the reflected wave from the water bottom into a digital signal;
Water bottom reflected wave spectrum decomposing means for obtaining a spectrum of the water bottom reflected wave by performing a discrete Fourier transform on the digital signal;
A bottom reflected wave frequency estimating means for estimating a frequency of the bottom reflected wave based on a spectrum of the bottom reflected wave;
Water bottom reflected wave maximum amplitude number detecting means for detecting a water bottom reflected wave maximum amplitude number which is a frequency number which is the maximum amplitude of the spectrum of the water bottom reflected wave;
For a plurality of bottom reflected waves obtained by a plurality of times of transmission, the frequency of the bottom bottom reflected wave maximum amplitude number is obtained and included in a certain frequency range including the frequency corresponding to the frequency number at which the frequency has the maximum value. Only the bottom reflected wave frequency estimation value to be selected, and the bottom reflected wave frequency determining means using the average value of the selected bottom reflected wave frequency estimated values as the determined bottom reflected wave frequency estimated value,
A water mass reflected wave analog-digital conversion means for converting a reflected wave from the water mass into a digital signal;
Water mass reflected wave spectrum decomposing means for obtaining a spectrum of the water mass reflected wave by discrete Fourier transform of the digital signal;
Water mass reflected wave frequency estimating means for estimating the frequency of the water mass reflected wave based on the spectrum of the water mass reflected wave;
Water mass reflected wave maximum amplitude number detecting means for detecting a water mass reflected wave maximum amplitude number which is a frequency number which is the maximum amplitude of the spectrum of the water mass reflected wave;
For a plurality of water mass reflected waves obtained by multiple transmissions, the water mass reflection included in a certain frequency range including the frequency corresponding to the frequency number at which the frequency of the water mass reflected wave maximum amplitude number takes the maximum value. A water mass reflected wave frequency determining means that selects only the wave frequency estimated value, and uses the average value of the selected water mass reflected wave frequency estimated values as a determined water mass reflected wave frequency estimated value;
A tide meter characterized by comprising