Voice over Packet (VoP)   -   Speech Codecs  |  Telephony Software  |  Network Layer  |  Data & Fax Modems


Back to Solutions Main Page




Site Map

Wireline Communications Solutions


VAD  |  DTMF  |  Echo Cancellation  | 


    Voice Activity Detection and AGC


Voice Activity Detection and AGC (Automatic Gain Control) algorithm module detects voice activity and adjusts the voice signal to a target power level. The AGC algorithm adjusts the voice power level to a user specified target within a user specified time constant. The AGC is integrated with the Voice Activity Detect algorithm to ensure that background noise or echo does not get amplified up to the specified target level.


  • Voice Activity detector classifies voice activity as "early" or "sustained" according to user specified parameters
  • Voice Activity detector adapts to different background ambient noise or line noise;
  • Available in integrated format with speech codec- G.722, G.726, G.729A

Metric for VAD (ADSP-2189)

 MIPS  0.6
 DM  120W
 PM  0.9KW


    DTMF Tone Detection

The Telephony software is used to place calls, monitor a call's progress and detect incoming calls on POTS (plain old telephone service) lines. In digital telephony where call control is directed by other means, DTMF dialing and call progress monitoring are required for secondary dialing.

The North American Telephony Platform is used for call control under North American administrations. The International Telephony Platform is used for call control under various International administrations.

When originating a telephone call, call progress tones are generated by the telephone network as the call sequences through the telephone network, The call progress signals include primary and secondary dial tone, busy, congestion, ring back and bong.

Telephone numbers can be dialed with tones (DTMF), pulses or adaptively. Adaptive dialing can be used when the capability of the telephone line is unknown, but tone dialing is preferred. Pulse dialing must be implemented with external hardware, usually a relay in the telephone circuitry.

Ring detection is used to detect valid incoming ring signals. Distinctive ring detection can be used to detect the different ring cadence. Distinctive ringing, an option available from the telephone company provides several different telephone numbers for a single physical line; incoming calls produce a distinct ring cadence for each dialed number.


  • Complete call progress monitoring (North American and International)
  • Per call time slot assignment and sample stream coding (A-Law, -Law or linear)
  • Hook Control and dialing procedures (North American and International)
  • International DAA support
  • Pulse, tone and adaptive dialing
  • DTMF Receiver (less than 150 hits on Bellcore test tape typical)
  • General Purpose Tone Detectors/Generators
  • Ring detect and distinctive ring detect
  • Multi-tasking environment compatible
  • Rate adaptation (V.110 and V.120) compatible


  • DAA interface with linear/PCM codec at standard sample rates
  • Direct interface to 8.0 kHz PCM data stream (A-Law or mu-law)
  • Distinct tone for Data/Facsimile/Voice available
  • Supported with Speech Coder Systems (G.722, G.729A, G.728 et al.)
  • supported with Line echo cancellation (G.165 compliant)
  • Multiple instances can be executed on a DSP / RISC

Availability:  on ADSP2189, TMS320C54 and other propriety cores.


    ITU-T G.165/G.168 Digital Echo Canceller


The Digital Echo Canceller (DEC) is a self-contained module that runs on family of Digital Signal Processors. The DEC is fully compliant with ITU G.165 and the relevant clauses of G.164 and V.25. It provides for a time span of 4 to 32ms. Echo Cancel Disable (ECD) tone (2100 Hz) is processed fully, however ECD processing can be bypassed to reduce the MIPS loading. The G.165 Non-Linear Processing (NLP) can also be bypassed.

The Line Echo Canceller is a self-contained module complying to ITU-T recommendations G.168. It provides variable process time span of 4 to 32ms, Switchable Non-Linear Processing (NLP) and tone detection.


  • Complies with ITU-T standards G.165/G.168(2000)
  • Simple user interface
  • Allows the user to run multiple instance of G.165


  • Used in conjunction with a standard speech codec for echo cancellation purposes
  • Operates on user defined frame size
  • Needs outer shell to set up input and output buffers
  • Available in Tail dependent and independent form for 16/32/64 ms

Typical Applications:

  • Dedicated single line echo cancellers
  • 4-Wire echo cancellations
  • Speaker phones


G.165 on ADSP2189

 MIPS  6.1*
 DM  360W
 PM  1.1KW
*MIPS increases with ECD enabled

G.168 on ADSP2189

 MIPS  6.5
 DM  400W
 PM  1.3KW

top             Nuntius Logo

Home Page   |   Company   |   Solutions   |   Technology   |   Employment   |   Contact   |   Site Map