CONTENTS

 1     Scope
 2     References
 3     Terms and definitions
 4     Abbreviations
 5     Test signals
 6     Characteristics of echo cancellers
        6.1     General
        6.2     Purpose, operation and environment
        6.3     External enabling/disabling
        6.4     Tests and requirements for performance with input signals applied to the send and receive paths
                  6.4.1     Transmission performance
                  6.4.2     Echo canceller performance
 7     Characteristics of an echo canceller tone disabler
        7.1     General
        7.2     Detector characteristics
                  7.2.1     Phase reversal detection
        7.3     Guardband characteristics
                  7.3.1     Noise tolerance
        7.4     Holding-band characteristics
        7.5     Operate time
        7.6     False operation due to speech currents
        7.7     False operation due to data signals
        7.8     Release time
        7.9     Other considerations
 8     NLPs for use in echo cancellers
        8.1     Scope
        8.2     General principles and guidelines
                  8.2.1     Function
                  8.2.2     Suppression threshold
                  8.2.3     Control of NLP activation
                  8.2.4     Frequency limits of control paths
                  8.2.5     Signal attenuation below threshold level
                  8.2.6     Testing of NLPs
Annex A – Description of an echo canceller reference tone disabler
Annex B – Description of a Reference NLP
        B.1     General
        B.2     Suppression threshold (TSUP)
        B.3     Static characteristics of activation control
        B.4     Dynamic characteristics of activation control
        B.5     Frequency limits of control paths
        B.6     Testing
Annex C – Composite source signals for testing of speech echo cancellers – Signal, description and analysis
        C.1     Introduction
        C.2     Composite Source Signal – general considerations
                  C.2.1     General description of the different sequences
                  C.2.2     Calculation and analysis using a composite source signal
        C.3     Bandlimited Composite Source Signal with speech like power density spectrum – practical realization for measurements of echo cancellers
                  C.3.1     Composite Source Signal for single talk
                  C.3.2     Bandlimited Composite Source Signal to simulate double talk
        C.4     Appropriate analyses to determine convergence characteristics of speech echo cancellers using the Composite Source Signal
                  C.4.1     Calculation in the frequency domain
                  C.4.2     Calculation in the time domain
                  C.4.3     Level calculations according to the active speech level P.56
Annex D – Echo-path models for testing of speech echo cancellers
        D.1     Introduction
                  D.1.1     CSS or white noise input
                  D.1.2     Tone(s) input
        D.2     Echo path models from network hybrid simulator
        D.3     Echo path models measured from telephone networks in North America
        D.4     Echo path models measured from telephone networks in Europe
Appendix I – Guidance for application of echo cancellers
        I.1        Scope
        I.2        Echo control in the PSTN
                  I.2.1     Echo suppressors
                  I.2.2     Echo cancellers
                  I.2.3     Responsibilities of modem manufacturers and end users
        I.3        Application rules and operational constraints
        I.4        Effect of cancellers on voice and data services
        I.5        High-level speech
                  I.5.1     Introduction
                  I.5.2     Double talk detection and activity detection
                  I.5.3     Effects of low bit rate coders
                  I.5.4     Effects of a non-linear echo path
                  I.5.5     Guidelines for Rout usage in echo cancellers
        I.6        Network and service evolutionary considerations
                  I.6.1     Bit transparency of echo cancellers
                  I.6.2     Convergence speed
                  I.6.3     Acoustic echo control and environments
                  I.6.4     Comfort Noise
        I.7        Considerations regarding echo canceller performance during double talk
                  I.7.1     Introduction
                  I.7.2     Double talk parameters
                  I.7.3     Analysis of technical parameters that influences performance under double talk conditions
                  I.7.4     Conducting the double talk Tests 3A and 3B without inhibiting the adaptation
                  I.7.5     Subjective and objective echo canceller testing
        I.8        Guidelines on the use of parameters for testing echo cancellers
        I.9        Bibliography
Appendix II – Measurement methods for characteristics of echo paths
       II.1     Introduction
       II.2     Measurement procedure
                 II.2.1     Measurement set-up
                 II.2.2     Computation of the echo path impulse response
       II.3     Analysis of echo-path characteristics
                 II.3.1     Dispersion time
                 II.3.2     Magnitude response of echo path
       II.4     Examples of echo path measurements from real networks
                 II.4.1     Echo-path characteristics from measurements in North America
                 II.4.2     Echo path characteristics from measurements in Europe
       II.5     Conclusions
       II.6     Bibliography
Appendix III – Multiple tail circuits