VoIP TTY Tone Segmentation for Packet Loss Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing VoIP systems are unreliable for TTY devices due to packet loss, which is not recoverable by current packet loss concealment algorithms, and existing solutions that improve reliability often preclude mixed-mode voice and TTY dialog.
Innovation Solution
Each acoustic tone in a Baudot character sequence is transmitted as a pair of data packets, with the first packet synchronized to the tone's start and the remainder encoded in the second packet, allowing for reconstruction of missing information at the receiving end by copying characteristics into unoccupied time slots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If TTY tones are transmitted as audio stream over VoIP, then voice communication efficiency is maintained, but packet loss causes character transmission errors
Solution Approach 1:
The patent segments each TTY character into multiple audio packets (typically 2-3 packets per character) with sequential numbering. This segmentation allows the receiving end to identify and reconstruct missing packets while maintaining the overall character integrity, thus improving reliability without sacrificing communication efficiency.
Solution Approach 2:
The patent implements preliminary error correction by adding redundant information to each packet, including character codes, packet numbers, and checksums before transmission. This preliminary action enables the receiving end to detect and correct errors without requiring retransmission, maintaining communication efficiency while improving reliability.
2Object-affected harmful factors
If packet loss concealment algorithms are used, then voice quality is maintained, but TTY character recovery is not achieved
Solution Approach 1:
The patent applies different processing methods to different types of audio data: voice packets use conventional packet loss concealment algorithms to maintain natural speech quality, while TTY packets use character-based error correction and reconstruction to ensure accurate character recovery. This local quality differentiation resolves the contradiction between voice quality and character reliability.
Solution Approach 2:
The patent introduces an intermediary TTY processing layer that sits between the audio stream and the TTY decoder. This intermediary layer identifies TTY packets, performs error correction, and reconstructs missing characters before passing data to the decoder, thereby enabling TTY character recovery without interfering with voice quality concealment.
3Reliability
If modem-type mechanisms are used for TTY transmission, then reliability is improved, but mixed-mode voice and TTY dialog is precluded
Solution Approach 1:
The patent merges TTY character transmission with voice communication by encoding TTY characters as audible tones within the same audio stream used for voice. The system combines voice packets and TTY packets into a single VoIP stream, allowing both voice and TTY to coexist and be transmitted simultaneously over the same channel, thus enabling mixed-mode communication while maintaining reliability through the error correction mechanisms.
Data Source
AI summary
The present invention is directed to a system for reliably transmitting characters encoded as audible signals using voice over Internet protocol communication systems. According to an embodiment of the present invention, tones within a sequence of tones encoding a textual character are transmitted using two data packets. The first data packet includes a segment of time having a starting boundary that is synchronized with the start of the encoded tone and extending for a time period equal to about one-half the total time period of the tone. The second packet encodes the remaining half of the audible tone. At the receiving end, data packets encoding a textual character are ordered according to the position occupied by the data encoded by the packet within the sequence of tones. If any unpaired packets of data are detected at the receiving end, the characteristics of the packets that are unpaired are copied into the time slot that would have been occupied by the missing packet. Accordingly, the present invention allows for the accurate transmission of textual characters over VoIP networks, even where a packet of data encoding part of a sequence of packets encoding a character is dropped.