JPH0342928A - Encoding device - Google Patents

Encoding device

Info

Publication number
JPH0342928A
JPH0342928A JP17751389A JP17751389A JPH0342928A JP H0342928 A JPH0342928 A JP H0342928A JP 17751389 A JP17751389 A JP 17751389A JP 17751389 A JP17751389 A JP 17751389A JP H0342928 A JPH0342928 A JP H0342928A
Authority
JP
Japan
Prior art keywords
signal
band
maximum value
sum signal
difference signal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP17751389A
Other languages
Japanese (ja)
Other versions
JPH0756959B2 (en
Inventor
Koji Takeno
浩司 竹野
Michiyo Goto
道代 後藤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Panasonic Holdings Corp
Original Assignee
Matsushita Electric Industrial Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Matsushita Electric Industrial Co Ltd filed Critical Matsushita Electric Industrial Co Ltd
Priority to JP17751389A priority Critical patent/JPH0756959B2/en
Publication of JPH0342928A publication Critical patent/JPH0342928A/en
Publication of JPH0756959B2 publication Critical patent/JPH0756959B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Landscapes

  • Signal Processing For Digital Recording And Reproducing (AREA)
  • Analogue/Digital Conversion (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)
  • Transmission Systems Not Characterized By The Medium Used For Transmission (AREA)

Abstract

PURPOSE:To improve the efficiency of encoding by determining the number of bits to be encoded adaptively with a maximum value and compressing the encoded signal although the number of encoded bits is determined indiscriminately with the maximum value before. CONSTITUTION:An input signal is divided into four bands by a frequency band divider 1 and stored, band by band, in a storage device 2 for a constant time signal. Then a maximum value detector 3 detects maximum values of the signal, band by band, and a 2nd encoder encodes the signal with six bits, band by band, and outputs the encoded signals. A threshold value determination device 6 determines threshold values, band by band, by using constants of the respective bands stored previously in a constant storage device 5 and the maximum values which are detected, band by band, by the maximum value detector 3. Then the signal stored in the storage device 2 for the constant time is encoded by the 2nd encoder 7 by using the threshold values of said bands and outputted. Namely, the 2nd encoder encodes only parts where the maximum values exceed the threshold values, so the efficiency is higher than before.

Description

【発明の詳細な説明】 産業上の利用分野 本発明は、音楽および音声信号を圧縮して伝送する符号
化装置に関するものである。
DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an encoding device for compressing and transmitting music and audio signals.

従来の技術 近年、テレビ電話などのように伝送路の容量が限られた
媒体を用いて、より高品質な音声や画像を伝送するため
に信号圧縮技術を用いた符号化装置が考案されてきてい
る。
Conventional technology In recent years, encoding devices using signal compression technology have been devised to transmit higher quality audio and images using media with limited transmission channel capacity, such as videophones. There is.

以下図面を参照しながら上述した従来の符号化装置の一
例について説明する。
An example of the above-mentioned conventional encoding device will be described below with reference to the drawings.

第4図は従来の符号化装置の構成を示すものである。第
4図において、41は周波数帯域分割器、42は記憶器
、43は最大値検出器、44は定数記憶器、45は第1
の符号化器、46は第2の符号化器で、図示のごとく接
続されている。
FIG. 4 shows the configuration of a conventional encoding device. In FIG. 4, 41 is a frequency band divider, 42 is a memory, 43 is a maximum value detector, 44 is a constant memory, and 45 is a first
encoder, 46 is a second encoder, connected as shown.

以上の様に構成された符号化装置について、以下周波数
帯域分割数を24とした場合の動作を説明する。まず入
力信号を周波数帯域分割器41で、24帯域に帯域分割
し、各帯域毎の信号を記憶器42で一定時間記憶する。
The operation of the encoding device configured as described above when the number of frequency band divisions is 24 will be described below. First, the input signal is divided into 24 bands by the frequency band divider 41, and the signal for each band is stored in the storage unit 42 for a certain period of time.

つぎに記憶器42に記憶された信号より最大値検出器4
3で各帯域毎に最大値を検出し、第1の符号化器で各帯
域毎に6ビツトで符号化し出力する。また記憶器42で
一定時間記憶された信号は、定数記憶器44にあらかじ
め記憶された各帯域毎に定められたビット数を用いて各
帯域毎に、第2の符号化器46で符号化して出力する。
Next, the maximum value detector 4 detects the signal stored in the memory 42.
3, the maximum value is detected for each band, and the first encoder encodes it with 6 bits for each band and outputs it. Further, the signal stored for a certain period of time in the storage device 42 is encoded by a second encoder 46 for each band using a predetermined number of bits for each band, which is stored in advance in the constant storage device 44. Output.

なお、最大値があらかじめ定められた値を超えない場合
は一定時間記憶された信号は符号化されない。
Note that if the maximum value does not exceed a predetermined value, the signal stored for a certain period of time is not encoded.

(例えば、「ロー ビットレート コーディング オブ
 ハイクラオリティ オーディオ シグナルズ アン イントロダクション トウ ザ マスカム シス
テム (Low bit−rate coding of h
ighquality audi。
(For example, "Low bit-rate coding of high-quality audio signals"
ighquality audi.

signals An  1ntroduction  to  the
  MASCAM  5ystea+)  J  、 
イービーニー レビュー テクニカル ナンバー230
(E B U Review Technical、 
No、230)  シイ、タイー・シイ、シュトル・エ
ム、リンク(G、Theile。
signals An introduction to the
MASCAM 5ystea+) J,
Ebenee Review Technical Number 230
(E B U Review Technical,
No, 230) See, Taiy See, Stru M, Link (G, Theile.

G、5La11. M、Link)共著 71−94頁
)発明が解決しようとする課題 しかしながら、上記のような構成では、最大値があらか
じめ定められた一定値を超えた場合は、定数記憶器によ
って定められたビット数で符号化しなければならず、最
大値が一定値を少し超えても一律に符号化ビット数が決
まり圧縮効率が悪いという課題を有していた。
G, 5La11. (Co-authored by M, Link) pp. 71-94) Problems to be Solved by the Invention However, in the above configuration, if the maximum value exceeds a predetermined constant value, the number of bits determined by the constant memory However, even if the maximum value slightly exceeds a certain value, the number of encoding bits is fixed uniformly, resulting in poor compression efficiency.

本発明は上記課題に鑑み、最大値により符号化するビッ
ト数を適応的に決め、効率よく圧縮符号化する符号化装
置を提供するものである。
In view of the above problems, the present invention provides an encoding device that adaptively determines the number of bits to be encoded based on the maximum value and performs compression encoding efficiently.

課題を解決するための手段 上記課題を解決するために本発明の符号化装置は、入力
信号を複数帯域に分割する周波数帯域分割器と、その出
力信号を一定時間記憶する記憶器と、一定時間記憶され
た信号の最大値を検出する最大値検出器と、帯域毎にあ
らかじめ定められた定数を記憶している定数記憶器と、
その定数と最大値を用いて最低周波数帯域では自己帯域
の最大値を最低周波数帯域用に決まっている定数で除算
し他の周波数帯域では各帯域より1帯域低い周波数帯域
の最大値を帯域毎に決まっている定数で除算して帯域毎
のしきい値を決定するしきい値決定器と、最大値を符号
化し、符号化信号を出力する第1の符号化器と、一定時
間記憶された信号をしきい値を用いて符号化し符号化信
号を出力する第2の符号化器とを備えたものである。
Means for Solving the Problems In order to solve the above problems, the encoding device of the present invention includes a frequency band divider that divides an input signal into a plurality of bands, a memory that stores the output signal for a certain period of time, a maximum value detector that detects the maximum value of the stored signal; a constant memory that stores a predetermined constant for each band;
Using that constant and maximum value, in the lowest frequency band, divide the maximum value of the own band by the constant determined for the lowest frequency band, and in other frequency bands, divide the maximum value of the frequency band one band lower than each band for each band. a threshold determiner that determines a threshold value for each band by dividing by a predetermined constant; a first encoder that encodes the maximum value and outputs an encoded signal; and a signal that is stored for a certain period of time. and a second encoder that encodes the code using a threshold value and outputs a coded signal.

作用 本発明は上記した構成によって、従来最大値によって一
律に符号化ビット数が決まっていたのに対し、最大値に
より符号化するビット数を適応的に決めて、符号化信号
を圧縮することにより符号化効率を改善することとなる
According to the above-described configuration, the present invention adaptively determines the number of bits to be encoded based on the maximum value and compresses the encoded signal, whereas conventionally the number of encoded bits is uniformly determined based on the maximum value. This will improve encoding efficiency.

実施例 以下、本発明の第1の実施例の符号化装置について、図
面を参照しながら説明する。
Embodiment Hereinafter, an encoding apparatus according to a first embodiment of the present invention will be explained with reference to the drawings.

第1図は本発明の第1の実施例における符号化装置の構
成を示すものである。第1図において、lは周波数帯域
分割器、2は記憶器、3は最大値検出器、4は第1の符
号化器、5は定数記憶器、6はしきい値決定器、7は第
2の符号化器で、図示のごとく接続されている。
FIG. 1 shows the configuration of an encoding device in a first embodiment of the present invention. In FIG. 1, l is a frequency band divider, 2 is a memory, 3 is a maximum value detector, 4 is a first encoder, 5 is a constant memory, 6 is a threshold value determiner, and 7 is a maximum value detector. Two encoders are connected as shown.

以上の様に構成された符号化装置について、以下周波数
分割数を4とした場合の動作を説明する。
The operation of the encoding device configured as described above when the number of frequency divisions is 4 will be described below.

まず入力信号を周波数帯域分割器1で4帯域に帯域分割
し、各帯域毎に記憶器2で一定時間信号を記憶する。つ
ぎに記憶器2に記憶された信号より、最大値検出器3で
各帯域毎に最大値を検出し、第2の符号化器で各帯域毎
に6ビツトで符号化し出力する。
First, an input signal is divided into four bands by a frequency band divider 1, and the signal is stored for a certain period of time in a storage device 2 for each band. Next, a maximum value detector 3 detects the maximum value for each band from the signal stored in the memory 2, and a second encoder encodes the signal with 6 bits for each band and outputs it.

つぎにしきい値決定器6で、定数記憶器5にあらかじめ
記憶された各帯域毎の定数と、最大値検出器3で各帯域
毎に検出した最大値とを用い、最低周波数帯域では自己
帯域の最大値を最低周波数帯域用に決まっている前記定
数で除算し、他の周波数帯域は各帯域より1帯域低い周
波数帯域の最大値を、帯域毎に定まっている定数で除算
して、各帯域毎にしきい値を決定する。つぎに記憶器2
で一定時間記憶された信号は、しきい値決定器6によっ
て決まった各帯域毎のしきい値を用い、第2の符号化器
7で符号化して出力する。
Next, the threshold determiner 6 uses the constant for each band prestored in the constant storage 5 and the maximum value detected for each band by the maximum value detector 3. The maximum value is divided by the constant determined for the lowest frequency band, and for other frequency bands, the maximum value of the frequency band one band lower than each band is divided by the constant determined for each band. Determine the threshold value. Next, memory device 2
The signal stored for a certain period of time is encoded by the second encoder 7 using the threshold value for each band determined by the threshold value determiner 6 and output.

従来の符号化器は、最大値があらかじめ定められた一定
値を超えた場合は定数記憶器によって定められたビット
数で符号化していたのに対し、本実施例の第2の符号化
器は最大値がしきい値を超えた部分のみを符号化するこ
とにより従来より効率のよい符号化器の実現が可能とな
る。
In the conventional encoder, when the maximum value exceeds a predetermined constant value, it is encoded with the number of bits determined by the constant storage device, whereas the second encoder of this embodiment encodes with the number of bits determined by the constant memory. By encoding only the portion where the maximum value exceeds the threshold value, it is possible to realize a more efficient encoder than in the past.

以上の様に第1の実施例によれば、入力信号を複数帯域
に分割する周波数帯域分割器と、その出力信号を一定時
間記憶する記憶器と、一定時間記憶された信号を最大値
を検出する最大値検出器と、帯域毎にあらかじめ定めら
れた定数を記憶している定数記憶器と、その定数と最大
値を用いて最低周波数帯域では自己帯域の最大値を最低
周波数帯域用に決まっている定数で除算し、他の周波数
帯域では各帯域より1帯域低い周波数帯域の最大値を帯
域毎に定まっている定数で除算して帯域毎のしきい値を
決定するしきい値決定器と、最大値を符号化し、符号化
信号を出力する第1の符号化器と、一定時間記憶された
信号をしきい値を用いて符号化し、符号化信号を出力す
る第2の符号化器とを備えたものであり、最大値により
符号化ビット数を適応的に決めることにより、効率のよ
い符号化器を実現することができる。
As described above, according to the first embodiment, there is a frequency band divider that divides an input signal into multiple bands, a memory that stores the output signal for a certain period of time, and a maximum value of the signal that is stored for a certain period of time. a maximum value detector that stores a predetermined constant for each band, and a constant memory that stores a predetermined constant for each band, and uses the constant and maximum value to determine the maximum value of the own band in the lowest frequency band a threshold determiner that determines a threshold value for each band by dividing the maximum value of a frequency band one band lower than each band by a constant determined for each band; A first encoder that encodes a maximum value and outputs an encoded signal; and a second encoder that encodes a signal stored for a certain period of time using a threshold value and outputs an encoded signal. By adaptively determining the number of encoding bits based on the maximum value, an efficient encoder can be realized.

なお、本実施例において、帯域分割数を4としているが
、一般に任意の分割数でよい。また、第1の符号化器を
6ビツトとしているが一般に任意のビット数でよい。
Note that in this embodiment, the number of band divisions is four, but generally any number of divisions may be used. Further, although the first encoder is 6 bits, generally any number of bits may be used.

以下、本発明の第2の実施例の符号化装置について、図
面を参照しながら説明する。
An encoding device according to a second embodiment of the present invention will be described below with reference to the drawings.

第2図は、本発明の第2の実施例における符号化装置の
構成を示すものである。第2図において21は加算器、
22は減算器、23は和信号用倍数器、24は差信号用
倍数器、25は和信号用周波数帯域分割器、26は差信
号用周波数帯域分割器、27は和信号用記憶器、28は
差信号用記憶器、29は和信号用最大値検出器、30は
差信号用最大値検出器、31は和信号用第1の符号化器
、32は差信号用第1の符号化器、33は定数記憶器、
34はしきい値決定器、35は和信号用第2の符号化器
、36は差信号用第2の符号化器で、図示のごとく接続
されている。
FIG. 2 shows the configuration of an encoding device in a second embodiment of the present invention. In FIG. 2, 21 is an adder;
22 is a subtracter, 23 is a multiplier for sum signal, 24 is a multiplier for difference signal, 25 is a frequency band divider for sum signal, 26 is a frequency band divider for difference signal, 27 is a memory for sum signal, 28 29 is a maximum value detector for sum signals, 30 is a maximum value detector for difference signals, 31 is a first encoder for sum signals, and 32 is a first encoder for difference signals. , 33 is a constant memory,
34 is a threshold value determiner, 35 is a second encoder for sum signals, and 36 is a second encoder for difference signals, which are connected as shown.

以上の様に構成された符号化装置について、以下周波数
分割数を4とし、和信号用記憶器と差信号用記憶器の記
憶時間を共に10m5とした場合の動作を説明する。
The operation of the encoding apparatus configured as described above will be described below when the number of frequency divisions is 4 and the storage time of both the sum signal memory and the difference signal memory is 10 m5.

2種類の入力信号を、音楽信号の左チャンネル及び右チ
ャンネル信号とし、左チャンネル信号をし、右チャンネ
ル信号をRとする。つぎに、加算器21によって(L+
R) 、減算器22によって(L−R)とし、それぞれ
和信号用倍数器23、差信号用倍数器24によって、1
/2倍することとしくL+R)/2.  (L−R)/
2となる。
Two types of input signals are a left channel signal and a right channel signal of a music signal, where the left channel signal is the left channel signal and the right channel signal is R. Next, the adder 21 adds (L+
R) and (LR) by the subtracter 22, and 1 by the sum signal multiplier 23 and the difference signal multiplier 24, respectively.
/2 times L+R)/2. (L-R)/
It becomes 2.

この和信号と、差信号をそれぞれ和信号用周波数帯域分
割器25と、差信号用周波数帯域分割器26で4帯域に
帯域分割し、それぞれ各帯域毎に信号をIon’s記憶
する和信号用記憶器27と、差信号用記憶器28で記憶
する。
The sum signal and the difference signal are divided into four bands by the sum signal frequency band divider 25 and the difference signal frequency band divider 26, respectively, and the signal is stored in Ion's for each band. It is stored in the memory 27 and the difference signal memory 28.

つぎに各帯域毎に和信号用最大値検出器29と、差信号
用最大値検出器30で、和信号用記憶器27と、差信号
用記憶器28で記憶された各帯域毎の信号より和信号用
最大値と差信号用最大値とを検出し、和信号用第2の符
号化器35と、差信号用第2の符号化器36で各帯域毎
に6ビツトで符号化し出力する。
Next, the maximum value detector 29 for sum signal and the maximum value detector 30 for difference signal are used for each band to calculate the signal for each band stored in the sum signal storage 27 and difference signal storage 28. The maximum value for the sum signal and the maximum value for the difference signal are detected, and encoded with 6 bits for each band by the second encoder 35 for the sum signal and the second encoder 36 for the difference signal and output. .

つぎにしきい値決定器34で、定数記憶器33にあらか
じめ記憶された各帯域毎の定数と、和信号用最大値検出
器29、差信号用最大値検出器30で各帯域毎に検出し
た最大値とを用い、最低周波数帯域では自己帯域の最大
値を最低周波数帯域用に決まっている定数で除算し、他
の周波数帯域は各帯域より1帯域低い周波数帯域の最大
値を、帯域毎に定まっている定数で除算して求めた各帯
域毎の、それぞれの和信号によるしきい値と、差信号に
よるしきい値を比較し、大きい方の値をそれぞれ各帯域
毎の和信号および差信号に共通なしきい値をとする。
Next, the threshold value determiner 34 uses the constant for each band prestored in the constant storage 33 and the maximum value detected for each band by the maximum value detector 29 for sum signal and the maximum value detector 30 for difference signal. For the lowest frequency band, divide the maximum value of the own band by a constant determined for the lowest frequency band, and for other frequency bands, divide the maximum value of the frequency band one band lower than each band, and divide it by the constant determined for the lowest frequency band. Compare the threshold value based on the sum signal and the threshold value based on the difference signal for each band, which are calculated by dividing by a constant, and use the larger value as the sum signal and difference signal for each band, respectively. Let a common threshold be.

つぎに和信号用記憶器27、差信号用記憶器28で10
−3間記憶された信号を、しきい値決定器34によって
決まって各帯域毎のしきい値を用い、和信号用第2の符
号化器35、差信号用第2の符号化器36でそれぞれ符
号化して出力する。
Next, the sum signal memory 27 and the difference signal memory 28 are 10
The signals stored for 3-3 days are processed by a second encoder 35 for sum signals and a second encoder 36 for difference signals, using threshold values for each band determined by a threshold determiner 34. Each is encoded and output.

従来の符号化器は、2種類の入力については1種類づつ
独立に符号化していたのに対し、本実施例の和信号用第
2の符号化器と差信号用第2の符号化器は、2種類の入
力より相関を利用してしきい値を決定し符号化すること
により従来より効率のよい符号化器が可能となる。
While the conventional encoder independently encodes two types of input, one type each, the second encoder for the sum signal and the second encoder for the difference signal of this embodiment , by determining a threshold value using correlation from two types of inputs and performing encoding, a more efficient encoder than before can be achieved.

以上の様に第2の実施例によれば、2種類の信号を同時
に入力し、各入力信号から和信号をつくる加算器と、差
信号をつくる減算器と、その加算器と減算器の出力をそ
れぞれ定数倍する和信号用倍数器と差信号用倍数器と、
その和信号用倍数器と差信号用倍数器の出力をそれぞれ
周波数軸上で複数帯域に分割する和信号用周波数帯域分
割器と差信号用周波数帯域分割器と、その和信号用周波
数帯域分割器と差信号用周波数帯域分割器の出力信号を
それぞれ一定時間記憶する和信号用記憶器と差信号用記
憶器と、その和信号用記憶器で記憶された信号より和信
号用最大値を検出する和信号用最大値検出器と、差信号
用記憶器で記憶された信号より差信号用最大値を検出す
る差信号用最大値検出器と、帯域毎にあらかじめ定めら
れた定数を記憶している定数記憶器と、その定められた
定数と和信号用最大値検出器と差信号用最大値検出器の
それぞれの出力を用い、最低周波数帯域では自己帯域の
最大値を最低周波数帯域用に決まっている定数で除算し
、他の周波数帯域は各帯域より1帯域低い周波数帯域の
最大値を帯域毎に定まっている定数で除算し、各帯域毎
に求まった値の大きい方を各帯域毎の和信号および差信
号に共通のしきい値とするしきい値決定器と、和信号用
最大値を符号化し、符号化信号を出力する和信号用第1
の符号化器と、差信号用最大値を符号化し、符号化信号
を出力する差信号用第1の符号化器と、和信号用記憶器
で記憶された信号をしきい値を用いて符号化し、符号化
信号を出力する和信号用第2の符号化器と、差信号用記
憶器で記憶された信号をしきい値を用いて符号化し符号
化信号を出力する差信号用第2の符号化器とを備えたも
のであり、最大値がしきい値を超えた部分のみを符号化
することにより従来より効率のよい圧縮符号化が実現で
きると共に、入力間の相関を利用して、効率のよい符号
化器を実現することができる。
As described above, according to the second embodiment, there is an adder that inputs two types of signals simultaneously and creates a sum signal from each input signal, a subtracter that creates a difference signal, and the outputs of the adder and subtracter. a sum signal multiplier and a difference signal multiplier that multiply each by a constant,
A sum signal frequency band divider and a difference signal frequency band divider that divide the outputs of the sum signal multiplier and difference signal multiplier into multiple bands on the frequency axis, respectively, and the sum signal frequency band divider A sum signal memory and a difference signal memory each storing the output signals of the frequency band divider for the difference signal and the frequency band divider for a certain period of time, and detecting the maximum value for the sum signal from the signals stored in the sum signal memory. A maximum value detector for the sum signal, a maximum value detector for the difference signal that detects the maximum value for the difference signal from the signal stored in the memory for the difference signal, and a constant predetermined for each band is stored. Using a constant memory, its determined constant, and the respective outputs of the maximum value detector for the sum signal and the maximum value detector for the difference signal, the maximum value of the self band is determined for the lowest frequency band in the lowest frequency band. For other frequency bands, divide the maximum value of the frequency band one band lower than each band by a constant determined for each band, and calculate the larger value for each band as the sum of each band. a threshold value determiner that uses a common threshold value for the signal and the difference signal; and a first threshold value determiner for the sum signal that encodes the maximum value for the sum signal and outputs the encoded signal.
a first encoder for difference signals that encodes the maximum value for difference signals and outputs an encoded signal, and encodes the signal stored in the sum signal storage using a threshold value. a second encoder for the sum signal which encodes the signal stored in the storage unit for the difference signal using a threshold and outputs the encoded signal; By encoding only the portion where the maximum value exceeds the threshold, it is possible to realize compression encoding that is more efficient than conventional methods, and by using the correlation between inputs, An efficient encoder can be realized.

なお、本実施例において、帯域分割数を4としているが
、一般に任意の分割数でよい。また、本実施例では最大
値を求める信号を10−8としているがこれに限らなく
ても同様にして符号化効率を改善できる。
Note that in this embodiment, the number of band divisions is four, but generally any number of divisions may be used. Further, in this embodiment, the signal for obtaining the maximum value is set to 10-8, but the coding efficiency can be similarly improved without being limited to this.

以下本発明の第3の実施例の符号化装置について、図面
を参照しながら説明する。
A coding device according to a third embodiment of the present invention will be described below with reference to the drawings.

第3図は、本発明の第3の実施例における符号化装置の
構成を示すものである。第3図において、21は加算器
、22は減算器、23は和信号用倍数器、24は差信号
用倍数器、25は和信号用周波帯域分割器、26は差信
号用周波数帯域分割器、27は和信号用記憶器、28は
差信号用記憶器、31は和信号用第1の符号化器、32
は差信号用第1の符号化器、35は和信号用第2の符号
化器、36は差信号用第2の符号化器、37は和信号用
有効上位桁検出器、38は差信号用有効上位桁検出器、
39は定数記憶器、40はしきい値決定器では、図示の
ごとく接続されている。
FIG. 3 shows the configuration of an encoding device in a third embodiment of the present invention. In FIG. 3, 21 is an adder, 22 is a subtracter, 23 is a multiplier for sum signal, 24 is a multiplier for difference signal, 25 is a frequency band divider for sum signal, and 26 is a frequency band divider for difference signal. , 27 is a memory for sum signals, 28 is a memory for difference signals, 31 is a first encoder for sum signals, 32
is a first encoder for difference signals, 35 is a second encoder for sum signals, 36 is a second encoder for difference signals, 37 is an effective upper digit detector for sum signals, and 38 is a difference signal. Effective upper digit detector for
39 is a constant storage device, and 40 is a threshold value determiner, which are connected as shown in the figure.

以上の様に構成された符号化装置について、以下周波数
分割数を4とし、和信号用記憶器と差信号用記憶器の記
憶時間を共に10m5とした場合の動作を説明する。
The operation of the encoding apparatus configured as described above will be described below when the number of frequency divisions is 4 and the storage time of both the sum signal memory and the difference signal memory is 10 m5.

2つの入力から、加算器21、減算器22で和信号、差
信号を求め、和信号用記憶器27、差信号用記憶器28
の出力を得るまでは、第2の実施例における動作と同様
である。
From the two inputs, an adder 21 and a subtracter 22 calculate a sum signal and a difference signal, and a sum signal memory 27 and a difference signal memory 28
The operation is the same as in the second embodiment until the output of is obtained.

つぎに、和信号用有効上位桁検出器37と、差信号用有
効上位桁検出器38で、それぞれ和信号用記憶器27、
差信号用記憶器28により記憶された10a+s間の信
号から和信号用最大値と和信号用最大値を求め、その各
最大値の有効上位桁を検出し、和信号用筒2の符号化器
35と、差信号用第2の符号化器36で各帯域毎に符号
化し出力する。ここでは、入力信号として16ビツトの
データを用いていることにすると、たかだか4ビツトで
符号化できる。
Next, the effective upper digit detector 37 for the sum signal and the effective upper digit detector 38 for the difference signal are connected to the sum signal memory 27,
The maximum value for the sum signal and the maximum value for the sum signal are obtained from the signal between 10a+s stored in the storage unit 28 for the difference signal, the effective upper digit of each maximum value is detected, and the encoder of the sum signal cylinder 2 35 and a second encoder 36 for difference signals, each band is encoded and output. Here, if 16-bit data is used as the input signal, it can be encoded with at most 4 bits.

つぎに、しきい値決定器40で、定数記憶器39にあら
かじめ記憶された各帯域毎の定数と、和信号用有効上位
桁検出器37、差信号用有効上位桁検出器38で各帯域
毎に検出した和信号用最大値の和信号用有効上位桁と差
信号用最大値の差信号用有効上位桁を用い、最低周波数
帯域では自己帯域の最大値の有効上位桁から最低周波数
帯域用に決まっている定数を減算し、他の周波数帯域は
各帯域より1帯域低い周波数帯域の最大値の有効上位桁
から帯域毎に定まっている定数を減算して求めた各帯域
毎の、それぞれ和信号によるしきい値と、差信号による
しきい値を比較し、大きい方の値をそれぞ各帯域毎の和
信号および、差信号に共通なしきい値とする。和信号用
記憶器27と、差信号用記憶器28に1On+s間記憶
された信号を、しきい値決定器40によって決まった各
帯域毎とのしきい値を用い、和信号用筒2の符号化器3
5と、差信号用第2の符号化器36でそれぞれしきい値
の値だけ右方向にビットシフトをしたのち符号化して出
力する。
Next, the threshold determiner 40 uses constants for each band prestored in the constant storage 39, the effective upper digit detector 37 for the sum signal, and the effective upper digit detector 38 for the difference signal for each band. Using the effective upper digits for the sum signal of the maximum value for the sum signal and the effective upper digits for the difference signal of the maximum value for the difference signal detected in The sum signal for each band is obtained by subtracting a fixed constant, and for other frequency bands, subtracting the constant fixed for each band from the effective upper digit of the maximum value of the frequency band one band lower than each band. The threshold value based on the difference signal is compared with the threshold value based on the difference signal, and the larger value is set as the common threshold value for the sum signal and the difference signal for each band. The signals stored in the sum signal storage 27 and the difference signal storage 28 for 1On+s are determined by the sign of the sum signal tube 2 using the threshold value for each band determined by the threshold value determination device 40. Maker 3
5 and a second difference signal encoder 36, the bits are shifted to the right by the threshold value, and then encoded and output.

従来の符号化器では、第2の符号化器で一定時間記憶さ
れた信号を除算したのち符号化していたのに対し、本実
施例の和信号用筒2の符号化器と差信号用第2の符号化
器は、2種類の入力より相関を利用してしきい値を決定
し符号化することにより従来より効率のよい符号化器が
可能となると共に、第2の符号化器により、しきい値を
用いてビットシフトを行ない一定時間記憶された信号を
符号化することにより従来より演算量の少ない符号化器
が可能となる。
In the conventional encoder, the second encoder divides the signal stored for a certain period of time and then encodes it. The second encoder uses correlation from two types of inputs to determine a threshold value and performs encoding, thereby making it possible to create a more efficient encoder than before. By performing bit shift using a threshold value and encoding a signal stored for a certain period of time, an encoder with a smaller amount of calculation than conventional ones can be realized.

以上の様に第3の実施例によれば、2種類の信号を同時
に入力し、各入力信号から和信号をつくる加算器と、差
信号をつくる減算器と、その加算器と減算器の出力をそ
れぞれ定数倍する和信号用倍数器と差信号用倍数器と、
その和信号用倍数器と差信号用倍数器の出力をそれぞれ
周波数軸上で複数帯域に分割する和信号用周波数帯域分
割器と差信号用周波数帯域分割器と、その和信号用周波
数帯域分割器と差信号用周波数帯域分割器の出力信号を
それぞれ一定時間記憶する和信号用記憶器と差信号用記
憶器と、その和信号用記憶器で記憶された信号より求め
た和信号用最大値より有効上位桁を検出する和信号用有
効上位桁検出器と、差信号用記憶器で記憶された信号よ
り求めた差信号用最大値より有効上位桁を検出する差信
号用有効上位桁検出器と、帯域毎にあらかじめ定められ
た定数を記憶している定数記憶器と、その定められた定
数と和信号用有効上位桁検出器と差信号用有効上位桁検
出器のそれぞれの出力を用い、最低周波数帯域では自己
帯域の最大値の有効上位桁より最低周波数帯域用に決ま
っている定数を減算し、他の周波数帯域は各帯域より1
帯域低い周波数帯域の最大値の有効上位桁より、帯域毎
に定まっている定数を減算し、各帯域毎に求まった値の
大きい方を各帯域毎の和信号および差信号に共通のしき
い値とするしきい値決定器・と、和信号用有効上位桁を
符号化し符号化信号を出力する和信号用筒1の符号化器
と、差信号用有効上位桁を符号化し符号化信号を出力す
る差信号用第1の符号化器と、和信号用記憶器で記憶さ
れた信号をしきい値を用いて符号化し符号化信号を出力
する和信号用筒2の符号化器と、差信号用記憶器で記憶
された信号をしきい値を用いて符号化し符号化信号を出
力する差信号用第2の符号化器とを備えたものであり、
最大値がしきい値を超えた部分のみを符号化することに
より従来より効率のよい符号化器が実現できると共に、
2種類の入力間の相関を利用して効率のよい符号化器を
実現できる。また、第2の符号化器により、一定時間記
憶された信号をしきい値を用いてビットシフトをしたの
ち符号化することにより従来より演算量の少ない符号化
器を実現できる。
As described above, according to the third embodiment, there is an adder that inputs two types of signals simultaneously and creates a sum signal from each input signal, a subtracter that creates a difference signal, and the outputs of the adder and subtracter. a sum signal multiplier and a difference signal multiplier that multiply each by a constant,
A sum signal frequency band divider and a difference signal frequency band divider that divide the outputs of the sum signal multiplier and difference signal multiplier into multiple bands on the frequency axis, respectively, and the sum signal frequency band divider and a sum signal memory and a difference signal memory that store the output signals of the frequency band divider for the difference signal for a certain period of time, respectively, and the maximum value for the sum signal obtained from the signals stored in the sum signal memory. An effective upper digit detector for the sum signal detects the effective upper digit, and an effective upper digit detector for the difference signal detects the effective upper digit from the maximum value for the difference signal obtained from the signal stored in the difference signal memory. , using a constant memory that stores predetermined constants for each band, the determined constants, and the respective outputs of the effective upper digit detector for the sum signal and the effective upper digit detector for the difference signal, For frequency bands, subtract the constant determined for the lowest frequency band from the effective upper digit of the maximum value of the own band, and for other frequency bands, subtract 1 from each band.
Subtract a constant determined for each band from the effective upper digit of the maximum value of the lower frequency band, and use the larger value found for each band as the common threshold for the sum signal and difference signal for each band. a threshold value determiner for encoding the effective upper digits for the sum signal and outputting an encoded signal, an encoder for the sum signal cylinder 1 that encodes the effective upper digits for the difference signal and outputting the encoded signal. a first encoder for the difference signal, a second encoder for the sum signal that encodes the signal stored in the sum signal storage using a threshold value and outputs an encoded signal, and a second encoder for the sum signal; a second encoder for difference signals that encodes the signal stored in the storage device using a threshold value and outputs a encoded signal;
By encoding only the portion where the maximum value exceeds the threshold, a more efficient encoder than before can be realized, and
An efficient encoder can be realized by utilizing the correlation between two types of inputs. Further, by bit-shifting the signal stored for a certain period of time using a threshold value and then encoding it using the second encoder, it is possible to realize an encoder with a smaller amount of calculation than the conventional encoder.

なお、本実施例において、帯域分割数を4としているが
、一般に任意の分割数でよい。また、本実施例では最大
値を求める信号を10m5としているがこれに限らなく
ても同様にして符号化効率を改善できる。
Note that in this embodiment, the number of band divisions is four, but generally any number of divisions may be used. Further, in this embodiment, the signal for which the maximum value is determined is 10m5, but the encoding efficiency can be similarly improved without being limited to this.

発明の効果 以上のように本発明の符号化装置は、入力信号を複数帯
域に分割する周波数帯域分割器と、その出力信号を一定
時間記憶する記憶器と、一定時間記憶された信号の最大
値を検出する最大値検出器と、帯域毎にあらかじめ定め
られた定数を記憶している定数記憶器と、その定数と最
大値を用いて最低周波数帯域では自己帯域の信号の最大
値を最低周波数帯域用に決まっている定数で除算し、他
の周波数帯域では各帯域より1帯域低い周波数帯域の信
号の最大値を帯域毎に定まっている定数で除算して帯域
毎のしきい値を決定するしきい値決定器と、最大値を符
号化し、符号化信号を出力する第1の符号化器と、一定
時間記憶された信号をしきい値を用いて符号化し、符号
化信号を出力する第2の符号化器とを備えているので、
最大値がしきい値を超えた部分のみを符号化することに
より従来の効率のよい符号化器を実現することができる
Effects of the Invention As described above, the encoding device of the present invention includes a frequency band divider that divides an input signal into a plurality of bands, a memory that stores the output signal for a certain period of time, and a maximum value of the signal stored for a certain period of time. a maximum value detector that detects the maximum value, a constant memory that stores a predetermined constant for each band, and a constant memory that stores a predetermined constant for each band. For other frequency bands, the maximum value of the signal in a frequency band one band lower than each band is divided by a constant determined for each band to determine the threshold value for each band. a threshold determiner, a first encoder that encodes a maximum value and outputs an encoded signal, and a second encoder that encodes a signal stored for a certain period of time using a threshold and outputs an encoded signal. Since it is equipped with an encoder of
A conventional efficient encoder can be realized by encoding only the portion where the maximum value exceeds the threshold.

また本発明の符号化装置は、2種類の信号を同時に入力
し、各入力信号から和信号をつくる加算器と、差信号を
つくる減算器と、その加算器と減算器の出力をそれぞれ
定数倍する和信号用倍数器と差信号用倍数器と、その和
信号用倍数器と差信号用倍数器の出力をそれぞれ周波数
軸上で複数帯域に分割する和信号用周波数帯域分割器と
差信号用周波数帯域分割器と、その和信号用周波数帯域
分割器と差信号用周波数帯域分割器の出力信号をそれぞ
れ一定時間記憶する和信号用記憶器と差信号用記憶器と
、その和信号用記憶器で記憶された信号より和信号用最
大値を検出する和信号用最大値検出器と、差信号用記憶
器で記憶された信号より差信号用最大値を検出する差信
号用最大値検出器と、帯域毎にあらかじめ定められた定
数を記憶している定数記憶器と、その定められた定数と
和信号用最大値検出器と差信号用最大値検出器のそれぞ
れの出力を用い、最低周波数帯域では自己帯域の最大値
の最低周波数帯域用に決まっている定数で除算し、他の
周波数帯域は各帯域より1帯域低い周波数帯域の最大値
を帯域毎に定まっている定数で除算し、各帯域毎に求ま
った値の大きい方を各帯域毎の和信号および差信号に共
通のしきい値とするしきい値決定器と、和信号用最大値
を符号化し符号化信号を出力する和信号用第1の符号化
器と、差信号用最大値を符号化し、符号化信号を出力す
る差信号用第1の符号化器と、和信号用記憶器で記憶さ
れた信号をしきい値を用いて符号化し符号化信号を出力
する和信号用第2の符号化器と、差信号用記憶器で記憶
された信号をしきい値を用いて符号化し、符号化信号を
出力する差信号用第2の符号化器とを備えているので、
最大値がしきい値を超えた部分のみを符号化することに
より従来より効率のよい圧縮符号化が実現できると共に
、2種類の入力間の相関を利用して効率のよい符号化器
を実現することができる。
The encoding device of the present invention also includes an adder that inputs two types of signals simultaneously and creates a sum signal from each input signal, a subtracter that creates a difference signal, and the outputs of the adder and subtracter are each multiplied by a constant. A frequency band divider for the sum signal and a frequency band divider for the difference signal that divides the outputs of the sum signal multiplier and the difference signal multiplier into multiple bands on the frequency axis, respectively. A frequency band divider, a sum signal memory and a difference signal memory, each storing the output signals of the sum signal frequency band divider and the difference signal frequency band divider for a certain period of time, and the sum signal memory. A maximum value detector for the sum signal detects the maximum value for the sum signal from the signal stored in the storage device for the difference signal, and a maximum value detector for the difference signal detects the maximum value for the difference signal from the signal stored in the storage device for the difference signal. , using a constant memory that stores constants determined in advance for each band, and the outputs of the constants and the maximum value detector for the sum signal and the maximum value detector for the difference signal, the lowest frequency band is determined. Then, divide the maximum value of the own band by a constant determined for the lowest frequency band, and for other frequency bands, divide the maximum value of the frequency band one band lower than each band by a constant determined for each band, A threshold determiner that uses the larger value obtained for each band as a common threshold for the sum signal and difference signal for each band, and a sum signal that encodes the maximum value for the sum signal and outputs the encoded signal. a first encoder, a first encoder for a difference signal that encodes the maximum value for a difference signal and outputs an encoded signal, and a signal stored in a memory for a sum signal using a threshold value; a second encoder for the sum signal that encodes the signal stored in the storage unit for the difference signal and outputs the encoded signal; and a second encoder for the difference signal that encodes the signal stored in the storage unit for the difference signal using a threshold value and outputs the encoded signal. Since it is equipped with two encoders,
By encoding only the part where the maximum value exceeds the threshold, it is possible to realize compression encoding that is more efficient than before, and it also realizes an efficient encoder by using the correlation between two types of inputs. be able to.

さらに本発明の符号化装置は、2種類の信号を同時に入
力し、各入力信号から和信号をつくる加算器と、差信号
をつくる減算器と、その加算器と減算器の出力をそれぞ
れ定数倍する和信号用倍数器と差信号用倍数器と、その
和信号用倍数器と差信号用倍数器の出力をそれぞれ周波
数軸上で複数帯域に分割する和信号用周波数帯域分割器
と差信号用周波数帯域分割器と、その和信号用周波数帯
域分割器と差信号用周波数帯域分割器の出力信号をそれ
ぞれ一定時間記憶する和信号用記憶器と差信号用記憶器
と、その和信号用記憶器で記憶された信号より求めた和
信号用最大値より有効上位桁を検出する和信号用有効上
位桁検出器と、差信号用記憶器で記憶された信号より求
めた差信号用最大値より有効上位桁を検出する差信号用
有効上位桁検出器と、帯域毎にあらかじめ定められた定
数を記憶している定数記憶器と、その定められた定数と
和信号用有効上位桁検出器と差信号用有効上位桁検出器
のそれぞれの出力を用い、最低周波数帯域では自己帯域
の最大値の有効上位桁より最低周波数用に決まっている
定数を減算し、他の周波数帯域は各帯域より1帯域低い
周波数帯域の最大値の有効上位桁より帯域毎に定まって
いる定数を減算し、各帯域毎に求まった値の大きい方を
各帯域毎の和信号および差信号に共通のしきい値とする
しきい値決定器と、和信号用有効上位桁を符号化し符号
化信号を出力する和信号用第1の符号化器と、差信号用
有効上位桁を符号化し符号化信号を出力する差信号用筆
1の符号化器と、和信号用記憶器で記憶された信号をし
きい値を用いて符号化し符号化信号を出力する和信号用
第2の符号化器と、差信号用記憶器で記憶された信号を
しきい値を用いて符号化し符号化信号を出力する差信号
用筆2の符号化器とを備えているので、最大値がしきい
値を超えた部分のみを符号化することにより従来より効
率のよい符号化器が実現できると共に、入力間の相関を
利用して効率のよい符号化器を実現できる。また、第2
の符号化器により、しきい値を用いてビットシフトを行
ない一定時間記憶された信号を符号化することにより従
来より演算量の少ない符号化器を実現できる。
Furthermore, the encoding device of the present invention includes an adder that inputs two types of signals simultaneously and creates a sum signal from each input signal, a subtracter that creates a difference signal, and multiplies the outputs of the adder and subtracter by a constant. A frequency band divider for the sum signal and a frequency band divider for the difference signal that divides the outputs of the sum signal multiplier and the difference signal multiplier into multiple bands on the frequency axis, respectively. A frequency band divider, a sum signal memory and a difference signal memory, each storing the output signals of the sum signal frequency band divider and the difference signal frequency band divider for a certain period of time, and the sum signal memory. A sum signal effective upper digit detector that detects the effective upper digits from the maximum value for the sum signal calculated from the signal stored in the memory, and an effective upper digit detector for the sum signal that detects the effective upper digit from the maximum value for the difference signal calculated from the signal stored in the memory for the difference signal. An effective high-order digit detector for difference signals that detects high-order digits, a constant memory that stores predetermined constants for each band, the determined constants, an effective high-order digit detector for sum signals, and a difference signal. Using the respective outputs of the effective upper digit detectors, in the lowest frequency band, subtract a constant determined for the lowest frequency from the effective upper digit of the maximum value of the own band, and for other frequency bands, subtract one band lower than each band. Subtract a constant determined for each band from the effective upper digit of the maximum value of the frequency band, and use the larger value found for each band as the common threshold for the sum signal and difference signal for each band. a threshold determiner, a first encoder for the sum signal that encodes the effective upper digits for the sum signal and outputs the encoded signal, and a first encoder for the difference signal that encodes the effective upper digits for the difference signal and outputs the encoded signal. A first encoder, a second encoder for sum signal which encodes the signal stored in the memory for sum signal using a threshold value and outputs an encoded signal, and a memory for difference signal. Since it is equipped with a difference signal encoder 2 that encodes the stored signal using a threshold value and outputs an encoded signal, only the part whose maximum value exceeds the threshold value is encoded. By doing so, it is possible to realize a more efficient encoder than in the past, and it is also possible to realize a more efficient encoder by using the correlation between inputs. Also, the second
By performing bit shifting using a threshold value and encoding a signal stored for a certain period of time, it is possible to realize an encoder with a smaller amount of calculation than conventional encoders.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は本発明の第1の実施例における符号化装置の構
成図、第2図は本発明の第2の実施例における符号化装
置の構成図、第3図は本発明の第3の実施例における符
号化装置の構成図、第4図は従来の符号化装置の構成図
である。 1.41・・・・・・周波数帯域分割器、2,42・・
・・・・記憶器、3,43・・・・・・最大値検出器、
4.45・・・・・・第1の符号化器、5・・・・・・
定数記憶器、6・・・・・・しきい値決定器、7,46
・・・・・・第2の符号化器、21・・・・・・加算器
、22・・・・・・減算器、23・・・・・・和信号用
倍数器、24・・・・・・差信号用倍数器、25・・・
・・・和信号用周波数帯域分割器、26・・・・・・差
信号用帯域分割器、27・・・・・・和信号用記憶器、
28・・・・・・差信号用記憶器、29・・・・・・和
信号用最大値検出器、30・・・・・・差信号用最大値
検出器、31・・・・・・和信号用第1の符号化器、3
2・・・・・・差信号用筆1の符号化器、33.39.
44・・・・・・定数記憶器、34.40・・・・・・
しきい値検出器、35・・・・・・和信号用第2の符号
化器、36・・・・・・差信号用筆2の符号化器、37
・・・・・・和信号用有効上位桁検出器、38・・・・
・・差信号用有効上位桁検出器。
FIG. 1 is a block diagram of an encoding device in a first embodiment of the present invention, FIG. 2 is a block diagram of an encoding device in a second embodiment of the present invention, and FIG. 3 is a block diagram of a coding device in a third embodiment of the present invention. FIG. 4 is a block diagram of a conventional encoding device. 1.41...Frequency band divider, 2,42...
...Memory device, 3,43... Maximum value detector,
4.45...first encoder, 5...
Constant storage device, 6...Threshold value determination device, 7, 46
... Second encoder, 21 ... Adder, 22 ... Subtractor, 23 ... Multiplier for sum signal, 24 ... ...Difference signal multiplier, 25...
. . . frequency band divider for sum signal, 26 . . . band divider for difference signal, 27 . . . storage unit for sum signal,
28... Memory for difference signal, 29... Maximum value detector for sum signal, 30... Maximum value detector for difference signal, 31... first encoder for sum signal, 3
2...Encoder of difference signal brush 1, 33.39.
44...Constant memory, 34.40...
Threshold detector, 35...Second encoder for sum signal, 36...Brush 2 encoder for difference signal, 37
...Effective upper digit detector for sum signal, 38...
...Effective upper digit detector for difference signal.

Claims (3)

【特許請求の範囲】[Claims] (1)入力信号を周波数軸上で複数帯域に分割する周波
数帯域分割器と、前記周波数帯域分割器の出力信号を一
定時間記憶する記憶器と、前記一定時間記憶された信号
の最大値を検出する最大値検出器と、前記帯域毎にあら
かじめ定められた定数を記憶している定数記憶器と、前
記あらかじめ定められた定数と、前記最大値を用いて前
記帯域毎のしきい値を決定するしきい値決定器と、前記
最大値を符号化し、符号化信号を出力する第1の符号化
器と、前記一定時間記憶された信号を前記しきい値を用
いて符号化し、符号化信号を出力する第2の符号化器と
を備え、前記しきい値決定器は、最低周波数帯域では自
己帯域の最大値を最低周波数帯域用に決まっている前記
定数で除算し、他の周波数帯域は各帯域より1帯域低い
周波数帯域の最大値を、帯域毎に定まっている前記定数
で除算し各帯域のしきい値とすることを特徴とする符号
化装置。
(1) A frequency band divider that divides the input signal into multiple bands on the frequency axis, a memory that stores the output signal of the frequency band divider for a certain period of time, and detects the maximum value of the signal stored for the certain period of time. a constant storage device storing a predetermined constant for each band; and determining a threshold value for each band using the predetermined constant and the maximum value. a threshold value determiner, a first encoder that encodes the maximum value and outputs an encoded signal, and encodes the signal stored for a certain period of time using the threshold value and outputs an encoded signal. a second encoder that outputs an output, and the threshold value determiner divides the maximum value of the own band by the constant determined for the lowest frequency band in the lowest frequency band, and divides the maximum value of the own band by the constant determined for the lowest frequency band, and An encoding device characterized in that the maximum value of a frequency band one band lower than the band is divided by the constant determined for each band to obtain a threshold value for each band.
(2)2種類の信号を同時に入力し、前記2種類の入力
信号から和信号をつくる加算器と、差信号をつくる減算
器と、前記加算器の出力を定数倍する和信号用倍数器と
、前記減算器の出力を定数倍する差信号用倍数器と、前
記和信号用倍数器の出力を周波数軸上で複数帯域に分割
する和信号用周波数帯域分割器と、前記差信号用倍数器
の出力を周波数軸上で複数帯域に分割する差信号用周波
数帯域分割器と、前記和信号用周波数帯域分割器の出力
信号を一定時間記憶する和信号用記憶器と、前記差信号
用周波数帯域分割器の出力信号を一定時間記憶する差信
号用記憶器と、前記和信号用記憶器で記憶された信号よ
り和信号用最大値を検出する和信号用最大値検出器と、
前記差信号用記憶器で記憶された信号より差信号用最大
値を検出する差信号用最大値検出器と、前記帯域毎にあ
らかじめ定められた定数を記憶している定数記憶器と、
前記あらかじめ定められた定数と前記和信号用最大値と
差信号用最大値を用いて前記帯域毎のしきい値を決定す
るしきい値決定器と、前記和信号用最大値を符号化し、
符号化信号を出力する和信号用第1の符号化器と、前記
差信号用最大値を符号化し、符号化信号を出力する差信
号用第1の符号化器と、前記和信号用記憶器で記憶され
た信号を、前記しきい値を用いて符号化し、符号化信号
を出力する和信号用第2の符号化器と、前記差信号用記
憶器で記憶された信号を、前記しきい値を用いて符号化
し、符号化信号を出力する差信号用第2の符号化器とを
備え、前記しきい値決定器は、前記和信号用最大値検出
器と前記差信号用最大値検出器の出力それぞれに対し、
最低周波数帯域では自己帯域の最大値を最低周波数帯域
用に決まっている前記定数で除算し、他の周波数帯域は
各帯域より1帯域低い周波数帯域の最大値を、帯域毎に
定まっている前記定数で除算し、前記各帯域毎に求まっ
た値の大きい方を前記各帯域毎の和信号および、差信号
に共通のしきい値とすることを特徴とする符号化装置。
(2) an adder that simultaneously inputs two types of signals and creates a sum signal from the two types of input signals, a subtracter that creates a difference signal, and a sum signal multiplier that multiplies the output of the adder by a constant; , a difference signal multiplier that multiplies the output of the subtracter by a constant, a sum signal frequency band divider that divides the output of the sum signal multiplier into a plurality of bands on the frequency axis, and the difference signal multiplier. a frequency band divider for a difference signal that divides the output of the frequency band divider into a plurality of bands on the frequency axis; a memory for a sum signal that stores the output signal of the frequency band divider for a sum signal for a certain period of time; and a frequency band for the difference signal. a difference signal memory that stores the output signal of the divider for a certain period of time; a sum signal maximum value detector that detects the sum signal maximum value from the signal stored in the sum signal memory;
a difference signal maximum value detector that detects a difference signal maximum value from the signal stored in the difference signal storage; a constant storage that stores a predetermined constant for each band;
a threshold determiner that determines the threshold value for each band using the predetermined constant, the maximum value for the sum signal, and the maximum value for the difference signal, and encodes the maximum value for the sum signal,
a first encoder for a sum signal that outputs an encoded signal; a first encoder for a difference signal that encodes the maximum value for the difference signal and outputs the encoded signal; and a memory for the sum signal. A second encoder for sum signal encodes the signal stored in the storage device using the threshold value and outputs a coded signal; a second encoder for a difference signal that encodes using a value and outputs an encoded signal, and the threshold value determiner includes a maximum value detector for the sum signal and a maximum value detector for the difference signal. For each output of the device,
For the lowest frequency band, divide the maximum value of the own band by the constant determined for the lowest frequency band, and for other frequency bands, divide the maximum value of the frequency band one band lower than each band by the constant determined for each band. , and the larger value obtained for each band is used as a common threshold for the sum signal and the difference signal for each band.
(3)2種類の信号を同時に入力し、前記2種類の入力
信号から和信号をつくる加算器と、差信号をつくる減算
器と、前記加算器の出力を定数倍する和信号用倍数器と
、前記減算器の出力を定数倍する差信号用倍数器と、前
記和信号用倍数器の出力を周波数軸上で複数帯域に分割
する和信号用周波数帯域分割器と、前記差信号用倍数器
の出力を周波数軸上で複数帯域に分割する差信号用周波
数帯域分割器と、前記和信号用周波数帯域分割器の出力
信号を一定時間記憶する和信号用記憶器と、前記差信号
用周波数帯域分割器の出力信号を一定時間記憶する差信
号用記憶器と、前記和信号用記憶器で記憶された信号よ
り和信号用最大値を検出し、前記和信号用最大値の有効
上位桁を検出する和信号用有効上位桁検出器器と、前記
差信号用記憶器で記憶された信号より差信号用最大値を
検出し、前記差信号用最大値の有効上位桁を検出する差
信号用有効上位桁検出器と、前記帯域毎にあらかじめ定
められた定数を記憶している定数記憶器と、前記あらか
じめ定められた定数と、前記和信号用有効上位桁と前記
差信号用有効上位桁を用いて前記帯域毎のしきい値を決
定するしきい値決定器と、前記和信号用有効上位桁を符
号化し、符号化信号を出力する和信号用第1の符号化器
と、前記差信号用有効上位桁を符号化し、符号化信号を
出力する差信号用第1の符号化器と、前記和信号用記憶
器で記憶された信号を、前記しきい値を用いて符号化し
、符号化信号を出力する和信号用第2の符号化器と、前
記差信号用記憶器で記憶された信号を、前記しきい値を
用いて符号化し、符号化信号を出力する差信号用第2の
符号化器とを備え、前記和信号用有効上位桁検出器は、
前記和信号用最大値から和信号用有効上位桁を検出し、
前記差信号用有効上位桁検出器は、前記差信号用最大値
から差信号用有効上位桁を検出し、前記しきい値決定器
は、前記和信号用有効上位桁検出器と前記差信号用有効
上位桁検出器の出力それぞれに対し、最低周波数帯域で
は、自己帯域の有効上位桁から最低周波数帯域用に決ま
っている前記定数を減算し、他の周波数帯域は各帯域よ
り1帯域低い周波数帯域の有効上位桁から帯域毎に定ま
っている前記定数を減算し、前記各帯域毎に求まった値
の大きい方を前記各帯域毎の和信号および、差信号に共
通のしきい値とすることを特徴とする符号化装置。
(3) an adder that simultaneously inputs two types of signals and creates a sum signal from the two types of input signals, a subtracter that creates a difference signal, and a sum signal multiplier that multiplies the output of the adder by a constant; , a difference signal multiplier that multiplies the output of the subtracter by a constant, a sum signal frequency band divider that divides the output of the sum signal multiplier into a plurality of bands on the frequency axis, and the difference signal multiplier. a frequency band divider for a difference signal that divides the output of the frequency band divider into a plurality of bands on the frequency axis; a memory for a sum signal that stores the output signal of the frequency band divider for a sum signal for a certain period of time; and a frequency band for the difference signal. A difference signal memory that stores the output signal of the divider for a certain period of time, and a sum signal maximum value from the signals stored in the sum signal memory, and an effective upper digit of the sum signal maximum value. an effective upper digit detector for the sum signal, which detects the maximum value for the difference signal from the signal stored in the memory for the difference signal, and detects the effective upper digit of the maximum value for the difference signal; using a high-order digit detector, a constant memory storing a predetermined constant for each band, the predetermined constant, the effective high-order digit for the sum signal, and the effective high-order digit for the difference signal. a threshold value determiner for determining the threshold value for each band; a first encoder for the sum signal that encodes the effective upper digits for the sum signal and outputs an encoded signal; and a first encoder for the sum signal; A first encoder for a difference signal encodes the effective upper digits and outputs an encoded signal, and encodes the signal stored in the sum signal memory using the threshold value and outputs an encoded signal. a second encoder for a sum signal that outputs a sum signal, and a second encoder for a difference signal that encodes a signal stored in the difference signal storage using the threshold value and outputs an encoded signal. the effective upper digit detector for the sum signal,
detecting the effective upper digit for the sum signal from the maximum value for the sum signal;
The effective upper digit detector for the difference signal detects the effective upper digit for the difference signal from the maximum value for the difference signal, and the threshold value determiner detects the effective upper digit for the difference signal from the effective upper digit detector for the sum signal. For each output of the effective upper digit detector, in the lowest frequency band, the constant determined for the lowest frequency band is subtracted from the effective upper digit of the own band, and for other frequency bands, the frequency band is one band lower than each band. The constant determined for each band is subtracted from the effective upper digits of Characteristic encoding device.
JP17751389A 1989-07-10 1989-07-10 Encoder Expired - Fee Related JPH0756959B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP17751389A JPH0756959B2 (en) 1989-07-10 1989-07-10 Encoder

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP17751389A JPH0756959B2 (en) 1989-07-10 1989-07-10 Encoder

Publications (2)

Publication Number Publication Date
JPH0342928A true JPH0342928A (en) 1991-02-25
JPH0756959B2 JPH0756959B2 (en) 1995-06-14

Family

ID=16032223

Family Applications (1)

Application Number Title Priority Date Filing Date
JP17751389A Expired - Fee Related JPH0756959B2 (en) 1989-07-10 1989-07-10 Encoder

Country Status (1)

Country Link
JP (1) JPH0756959B2 (en)

Also Published As

Publication number Publication date
JPH0756959B2 (en) 1995-06-14

Similar Documents

Publication Publication Date Title
KR100307596B1 (en) Lossless coding and decoding apparatuses of digital audio data
RU94046112A (en) METHOD FOR REDUCING THE DATA NUMBER WHEN TRANSMITTING AND / OR ACCUMULATING DIGITAL SIGNALS COMING FROM MULTIPLE RELATED CHANNELS
US5856797A (en) Data encoding device, data decoding device, data encoding method and data decoding method
US20060153392A1 (en) Method and apparatus for encoding and decoding multi-channel signals
JP5035459B2 (en) Audio signal transmission device, audio signal reception device, and audio signal transmission system
JPH061916B2 (en) Band division encoding / decoding device
JPH0342928A (en) Encoding device
JPS62114342A (en) Digital signal transmission system high efficiently, high quality
KR960012477B1 (en) Adaptable stereo digital audio coder & decoder
JPH03127516A (en) Coder
JPH0779251B2 (en) Subband filter and subband coding method
JP2006171771A (en) Voice coding method and voice decoding method
JP2006171770A (en) Voice coding method and decoding method
JPH0779250B2 (en) Subband filter and subband coding method

Legal Events

Date Code Title Description
LAPS Cancellation because of no payment of annual fees