Browse Prior Art Database

Transmission of Audio Data from Aircraft to Ground Station for Investigation and Analysis

IP.com Disclosure Number: IPCOM000236198D
Publication Date: 2014-Apr-11
Document File: 7 page(s) / 173K

Publishing Venue

The IP.com Prior Art Database

Related People

Bin Dong: INVENTOR

Abstract

Flight recorders are installed in an aircraft for recording voice and data information onboard the aircraft. This information can be used later for investigation and analysis purposes in case of accidents or other significant incidents. Two such flight recorders are: a Flight Data Recorder (FDR) and a Cockpit Voice Recorder (CVR). The FDR records all flight condition related information including performance, health, direction, altitude etc., and the CVR records audio in the aircraft using microphones installed in cockpit or pilot headsets. The recorded audio data is converted to text using speech recognition techniques. This text data, in sync with FDR data, is transmitted to ground tracking stations. This paper provides a method for accurately recognizing the speech and converting it into text in accordance with a proposed language model. Also presented is the use of wireless cellular technology for transfer of CVR audio data to a ground station database post aircraft landing or upon a pre-defined event and reception of upgraded language model configuration files at the aircraft before subsequent flight of the aircraft.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 24% of the total text.

Page 01 of 7

Transmission of Audio Data from Aircraft to Ground Station for Investigation and Analysis

Bin Dong

ABSTRACT

Flight recorders are installed in an aircraft for recording voice and data information onboard the aircraft. This information can be used later for investigation and analysis purposes in case of accidents or other significant incidents. Two such flight recorders are: a Flight Data Recorder (FDR) and a Cockpit Voice Recorder (CVR). The FDR records all flight condition related information including performance, health, direction, altitude etc., and the CVR records audio in the aircraft using microphones installed in cockpit or pilot headsets. The recorded audio data is converted to text using speech recognition techniques. This text data, in sync with FDR data, is transmitted to ground tracking stations. This paper provides a method for accurately recognizing the speech and converting it into text in accordance with a proposed language model. Also presented is the use of wireless cellular technology for transfer of CVR audio data to a ground station database post aircraft landing or upon a pre-defined event and reception of upgraded language model configuration files at the aircraft before subsequent flight of the aircraft.


1. INTRODUCTION

In spite of sustained efforts in modern times to make air travel the safest ever, aviation accidents do occur. Often these accidents present a fresh set of challenges for all stakeholders to mitigate such disasters in future. Hence analysis and investigation for all aviation incidents, more so for accidents, is conducted thoroughly. As the trend goes, focus of attention in the analysis and investigation phase, post an aviation disaster shifts


Page 02 of 7

to the 'Black Box'. Commercial airplanes are required to be equipped with a Cockpit Voice Recorder (CVR) and a Flight Data Recorder (FDR). Each of these two instruments are commonly referred to as the 'Black Box'. It often holds the vital clues that can help air crash investigators find out what happened just before the crash or what caused the crash. The FDR records operating parameters such as the time, altitude, airspeed, and direction from the avionics systems on the aircraft. The CVR records what the crew says and monitors any sounds that occur within the cockpit. This recorded flight data and cockpit audio data ideally should be transmitted to ground tracking stations during flight, after landing of the aircraft, or upon a pre-defined event, which then could be used for tracking or investigation purposes. Though data compression techniques are applied, there are significant cost overheads (memory and/or bandwidth requirements) involved in storing and transmitting the audio signals. Hence audio signals captured by the CVR are converted to text using speech recognition techniques so that they can be stored onboard the aircraft or transmitted to the ground stations as text data. Since this textual representation of the cockpit audio sig...