A transmission system usually includes a transmitter, a receiver, and a transmission channel The main characteristics of a ‘shallow’ water transmission channel are ‘multipath’ effect and noise. Existing acoustic data transmission systems cannot be operated within this environment or they have considerably reduced data rate.

However, the indirect paths of an acoustic wave from the transmitter to the receiver experiences a number of reflections and loose energy; fortunately, the longer the way, the more attenuated it is. Then after a certain time delay, T, the strength on the different signals arriving at the receiver can be selected by using a power criterion.

Starting from this point, Ifremer has developed an original underwater data-transmitting system called the self-contained transmitting system; it is based on an original modulation scheme which fights against the above described problems and enables the user to transmit data at high data rate.

The self-contained transmitting system consists of two main parts- a transmitter subsystem and a receiver subsystem These two parts are symmetrical: the ‘transmitter’ achieves the coding of the numerical data and the modulation of the carrier, and the ‘receiver’ demodulates the acoustic wave and decodes the transmitted information This equipment can be considered as a sophisticated modulator/demodulator (modem) using an underwater acoustic link.

Protection against multipaths is ensured by an original modulation scheme which can be compared to the spread spectrum method. If the reliability of the transmission must be increased, coding of the data can be introduced but the data rate is reduced.

General principle

An acoustic wave is modulated using a linear law. The characteristics of this modulation are described in Figure 1 whereis initiated as soon as the synchronization frequency F8 is detected. This modulation has the same characteristics ¿F and To as the transmitted one but the frequency is shifted (see Fig 1).

A ‘beat frequency’ can then be created by using the received signal and the internally made reference.

The direct path will always be centred on the frequency¿F and the other paths, which are delayed because of longer paths, will be centred on frequencies¿F, different from ¿F It will be therefore possible to separate the direct path from the reflected ones by a narrow band filtering around ¿F

Table available in full paper
FSK linear modulation

Starting from the above described principle, a special modulation has been developed; it has been called FSK linear modulation Its characteristics are shown in Figure 2 The frequency of the transmitted acoustic carrier is switched from a slope, which represents the ‘0’ of the binary data, to another slope, which is the

‘1’ of the numerical data, depending on the value of the binary data to be transmitted.

After beating with the local reference, matched filters, centred on ¿F0 and ¿F1, are used to determine whether a ‘0’ or a ‘1’ has been transmitted.


The transmitter and the receiver subsystems are based on a set of cards called Ifremer 800 which use the well-known CMOS NSC 800 microprocessor.

This content is only available via PDF.
You can access this article if you purchase or spend a download.