Techniques for adaptive video multicasts through acoustic channels in underwater environments

An apparatus for underwater communications comprising:

• an acoustic transceiver;
• at least one processor;
• and
• at least one memory including one or more sequences of instructions,
• the at least one memory and the one or more sequences of instructions configured to, with the at least one processor, cause the apparatus to perform at least the following,
• a. retrieving from a computer-readable medium first data that indicates a model that comprises a neural network encoder that outputs a value for a class of Modulating and Coding Schemes (MCS), which model is trained on a training set including for each instance an input value for a quality metric, an input value indicating one of a plurality of a scalable video coding (SVC) structures for that quality metric, and an input value of acoustic channel information such that reconstructed video features based on the class of MCS output from the model are sufficiently similar to original video features for a particular purpose;
• b. determining second data that indicates a current quality value for the quality metric and a current channel value for the acoustic channel information and a current SVC value indicating one of the plurality of SVC modules;
• c. generating third data that indicates a current MCS value for the class of MCS, which value is output by the encoder of the first data operating on the second data;
• and
• d. causing a current video stream to be transmitted as an underwater acoustic signal based on the second data and the third data.

A method for underwater acoustic communications, comprising:

• a. training automatically on a processor a model that comprises a neural network encoder that outputs a value for a class of Modulating and Coding Schemes (MCS), which model is trained on a training set including for each instance an input value for a quality metric, an input value indicating one of a plurality of scalable video coding (SVC) structures for that quality metric, and an input value of acoustic channel information such that reconstructed video features based on the class of MCS output from the model are sufficiently similar to original video features for a particular purpose;
• b. sending first data that indicates the model to a processor on an underwater device that comprises an underwater acoustic transceiver, wherein the underwater device is configured to perform at least the steps of:
• i. determining second data that indicates a current quality value for the quality metric and a current channel value for the acoustic channel information and a current SVC value indicating one of the plurality of SVC modules;
• ii. generating third data that indicates a current MCS value for the class of MCS, which value is output by the encoder of the first data operating on the second data;
• and
• iii. causing a current video stream to be transmitted as an underwater acoustic signal based on the second data and the third data.

A non-transitory computer-readable medium carrying one or more sequences of instructions for underwater communications, wherein execution of the one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of:

• a. retrieving from a computer-readable medium first data that indicates a model that comprises a neural network encoder that outputs a value for a class of acoustic physical layer video protocol (AC), which model is trained on a training set including for each instance an input value for a quality metric, an input value indicating one of a plurality of application layer video coding (VC) structures for that quality metric, and an input value of acoustic channel information such that reconstructed video features based on the AC output from the model are sufficiently similar to original video features for a particular purpose;
• b. determining second data that indicates a current quality value for the quality metric and a current channel value for the acoustic channel information and a current VC value indicating one of the plurality of VC structures;
• c. generating third data that indicates a current AC value for the class of AC, which value is output by the encoder of the first data operating on the second data;
• and
• d. causing a current video stream to be transmitted as an underwater acoustic signal based on the second data and the third data.