Monitoring of Global Acoustic Transmissions: Signal Processing and Preliminary Data Analysis

• Main Author: Frogner, Gary R.
• Other Authors: NAVAL POSTGRADUATE SCHOOL MONTEREY CA
• Format: Text
• Language: English
• Published: 1991
• Subjects: Physical and Dynamic Oceanography; Acoustics; *GREENHOUSE EFFECT; COMPUTER PROGRAMS; SIGNAL PROCESSING; FREQUENCY; DATA PROCESSING; PEAK VALUES; TEMPERATURE; GLOBAL; MONITORING; REAL TIME; RESOLUTION; THESES; TIME; CODING; FEASIBILITY STUDIES; SIGNALS; HEATING; SOUND TRANSMISSION; ACOUSTIC CHANNELS; DEEP DEPTH; DOPPLER EFFECT; PATTERNS; HYDROPHONES; RECEPTION; ARRIVAL; TIME DOMAIN; SOUTH(DIRECTION); UNDERWATER SOUND; ISLANDS; INDIAN OCEAN; OCEAN BASINS; *GLOBAL TRANSMISSIONS; *ACOUSTIC TRANSMISSIONS; *SIGNAL PROCESSING; ACOUSTIC TOMOGRAPHY; ACOUSTIC VELOCITY; Heard Island; Indian
• Online Access: http://www.dtic.mil/docs/citations/ADA246572; http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA246572

A great deal of controversy exists concerning the possible global warming trend which may occur as a result of a documented increase in atmospheric greenhouse gasses. The 1991 Heard Island Feasibility Experiment tested the feasibility of using transmissions of acoustic energy through major ocean basins of the world to monitor spatially averaged global temperature trends. This thesis documents the Naval Postgraduate School's reception of the phase encoded signal transmitted from the Southern Indian Ocean, development of real-time signal processing software, and preliminary data analysis. Data, received from a 32-channel vertical array suspended in the deep sound channel off the coast of Monterey, CA, was processed using real-time capable software. Data processing to reduce noise, determine SNR, and remove the m-sequence coding was found to be quite sensitive to Doppler frequency shifts. Although the SNR of the raw data was only about -27.5 dB for individual hydrophones, the transmitted signal was detected in both the frequency and time domains. However, the maximum processed signal peak in the time domain had an SNR of only +9 dB which is insufficient for use in a long term global temperature monitoring project. As it provides inadequate arrival time resolution.