Establishing Message Reliability and Security in An underwater Command link

Publisher Website
Google Scholar
Abstract
Multipath, reverberation, and natural and man made noise limit the data rate and reliability of underwater acoustic command and telemetry links. This paper describes signal coding, gating, and receiver design techniques used in the RATAC(R) acoustic command and control system. Probability relationships for correct and error commands are derived and performance predictions made as a function of range. Test data on an offshore drilling platform is compared to theoretical performance predictions. 1.0 INTRODUCTION Acoustic telemetry techniques for control of underwater devices are limited by the constraints of the acoustic medium and its boundaries, and the sources of noise which are present in the underwater environment. Reliable command operation of these devices is a stringent requirement due to their inaccessibility and the slow communication data rates. Errors and undesired commands must be avoided in all types of noise and interference. This paper will discuss the signal processing and coding techniques, and the predicted and tested performance of the Raytheon RATAC (R) * acoustic command and control system. The system provides multichannel capability, and is designed to minimize the probability of a false command caused by natural or man made noise, or propagation effects. This is accomplished through the use of frequency diversity, time coding, and receiver gating. Test results at a drilling platform installation are included. 2.0 THE UNDERWATER ACOUSTIC ENVIRONMENT The background noise environment consists of several natural and man made noise sources. The most important natural noise sources are the usual sea state noise and the noise due to raindrops striking the surface.1 These spectra are shown in Figure 1. Biological noise sources are primarily concentrated at low frequencies, with the exception of snapping shrimp, which produces a noise level about equal to the rain noise spectrum.2 Interference due to porpoise and whale calls, which may appear similar to tones or frequency slides, is minimized by signal frequency coding and time gating techniques. Man made noises are generally localized, and generate most of their energy in the low frequency regions of the spectrum. The noise spectra produced by normal operation on a drilling platform have been chosen as a worst case. Two spectra are shown in Figure 1 for drilling platform noise.3 & 4 The first is the peak values of measured noise level from the CUSS I drilling barge; and the second is the noise level measured at a distance of 70 yards from a fixed leg platform, and referred to a standard reference distance of one yard. Spherical spreading was assumed as a worst case. The difference in measured levels is due to the presentation of peak values in the first case for the noise spectra, and the different platform noise characteristics. These curves include the maximum expected noise levels, with the exception of isolated events such as explosive pipe cutting. The noise spectrum consists of many line components as well as a broadband background. The signal decoding technique for a telemetry system must there for be able to discriminate against line frequency components as well as the more common white Gaussian noise background generally assumed for computational convenience.

This publication has 0 references indexed in Scilit: