VoIP Equipment for Satellite Use

  • Status
    Ongoing
  • Status date
    2008-04-14
Objectives

The project addresses the difficulties of carrying VoIP (Voice over IP) traffic over shared satellite services (especially DVB-RCS) and will result in a commercially saleable device which allows telecom operators to deploy a range of satellite-based commercial services.

The platform for the development will be WTL's existing IPNx switch which is now successfully used by terrestrial telecoms operators. It will be made more suitable for satcom operators. This will involve adapting WTL's patented NOP (Network Optimisation Protocol) which is currently used to improve the bandwidth utilisation of VoIP traffic within a telephone network. The resultant product will then be suitable for use with VSAT or other low cost satellite services. Particular attention will be given to the applicability of DVB-RCS service for VoIP.
Challenges

The key issues are expected to be:


  1. The achievement of high audio quality whilst using very low amounts of satellite bandwidth. This must be verified by the use of an independent test laboratory,

  2. Jitter (delay variation) tolerance,

  3. Resilience to packet loss without serious degradation,

  4. Packet prioritisation of real time (voice) traffic,

  5. The effect of Bandwidth on Demand,

  6. Remote management of the equipment whilst using very little link capacity),

  7. The design of an algorithm for the software detection of DTMF digits,

  8. Design of a method for playing constant tones (to match the tones normally played to the user by a standard telephone network),

  9. Resource management of multiple simultaneous IP streams playing tones or messages to many concurrent users,

  10. Management of a successful live customer trial.
Benefits

The ability to use optimised VoIP over DVB-RCS will give long distance telecom operators many new opportunities. The lower cost of infrastructure and space segment is likely to be particularly appealing to less developed countries where telecom provision is still sparse.

Features


  1. VoIP over DVB-RCS: It will be necessary to create new algorithms for the ways in which IP voice packets are processed after they have been encoded from normal analogue speech and before they are transmitted over the IP network.

  2. Improved voice encoding: a major determinant of the voice quality delivered by a VoIP system is the voice encoding algorithm (codec) used. WTL have identified a new codec that is available from an existing technology partner which will be integrated into the NOP framework.

  3. WTL Equipment Management over Satellite: Current management tools must be extended to ensure that all operations can be carried out as effectively remotely as they can locally. It is important that remote management consumes a minimum of the satellite link capacity to achieve its objectives.

  4. IP-based IVR: It is proposed that an IP-based Interactive Voice Response (IVR) system will be developed that resides in the IPNx switch at the remote site. This task will mean replacing dedicated DSP-based hardware with a software-only solution.
Plan

The project is divided into 2 phases:


  • Phase 1 builds a DVB-RCS reference network and uses emulation tools to characterise the issues currently affecting the quality of VoIP over satellite.

  • Phase 2 implements the improvements identified in Phase 1 and culminates in a Beta Test installation using VoIP over DVBRCS with a live telecom operator.
Current status

Phase 1 of the project was successfully completed. This involved the creation of a real and an emulated VoIP over DVB-RCS network. Tests were carried out over these networks to benchmark performance and to indicate whether any improvements were required to WTL's NOP (Network Optimisation Protocol - a proprietary method of reducing the IP bandwidth consumed by VoIP calls). The test report has been published and the conclusions indicate a package of modifications for NOP that will improve voice quality over DVB-RCS links. The BDR meeting took place in May 2006 and formal End of Phase 1 occurred in September 2006.

A number of modifications were consequently made to the proposal for Phase 2 work which has now been accepted. The Go-Ahead for Phase 2 was given in November 2006 and the Work packages to design the modified product have been underway ever since. The next milestone is the Mid Term Review where the Design Documents will be delivered.