Publication Detail

Underwater Wireless Video Transmission using Acoustic OFDM

Jordi Ribas
118 pp.
MITSG 09-37
$100.00 DOM / $120.00 INT

The current project aims to design and implement an acoustic OFDM system for underwater video transmissions. The thesis work combines a theoretical part, whose objective is to choose the appropriate techniques to deal with the characteristics of the targeted channel, and a practical part regarding the system deployment and experimental tests.

The theoretical research has focused on (1) video compression techniques, (2) OFDM modulation and detection techniques, (3) channel estimation algorithms, and (4) Peak to Average Ratio (PAR) reduction techniques for OFDM signals.

Considering (1), the standard MPEG-4 compression technique has been chosen. Specifically, the Very Low Bit-rate Video (VLBV) layer provides algorithms and tools for applications operating at bit rates typically between 5 and 64 kbps, supporting image sequences with low spatial resolution (typically up to CIF resolution) and low frame rates (typically up to 15 Hz).

About (2), the chosen techniques have been a coherent detector for underwater acoustic channels, known as Low complexity OFDM detector for underwater acoustic channels, and a differential detector. The first algorithm has a block by block dependency and has proved to be inappropriate in situations with high acceleration and channel change over time, so the second algorithm has been designed to eliminate this dependency to deal with fast varying channels.

In the realm of (3), the decision has been to use a noise variance reducing algorithm, known as Sparse channel estimation. The main idea that allows the implementation of such technique in the current context is that the channel response in the time domain for underwater acoustic channels is proved to be generally sparse and, as a result, some samples of the estimated response on the receiver side can be truncated to reduce the noise.

The PAR reduction techniques (4) aim at reducing the peak to average power ratio of the signal to avoid spikes in the time domain, so as to make a better use of the power (which is one of the main problems of the OFDM systems) and better operate the amplifying circuitry on the transmitter side. The techniques that have been used include the following: clipping and filtering of the OFDM signal, symbol interleaving, in band tone reservation, and out of band tone insertion.

The practical part of the project consists of the implementation of the selected techniques, and the following sets of experiments: software simulations, in-air experiments, and underwater tests.

The implemented OFDM system has been developed in MATLAB due to the advantages that this application has in order to debug the program, tune the chosen parameters, and analyze results. The MATLAB scripts generate a wavefile which then is transmitted to the hardware from a laptop, using a Software Defined Radio (SDR) platform that allows the use of scripts programmed in Python. Once the complementary script records the wave on the receiving laptop from the receiving side hardware, another set of MATLAB scripts demodulate and detect the data.

The key hardware components that have been used during the in-air testbed experiments include a speaker, a microphone, the data acquisition system, and two laptop computers. The main components needed for the underwater experiments are the transducer and the hydrophone, the data acquisition system and two laptop computers. Raw videos have been compressed using the MPEG-4 technique prior to transmission.

The simulations that have been performed aim at proving the correct implementation of all the algorithms by coupling the generated OFDM signal to a channel simulation and noise addition algorithm, that artificially introduces some of the real channel effects into the signal.

The in-air experiments' goal is to further test the system by adding other real channel effects, such as motion, and confirming that the algorithm performance is the expected. In addition, a comparison between the in-air conditions and the underwater ones is pursued.

Finally, the underwater experiments have been performed to finally prove the usefulness of the coupled MPEG-4 compression and OFDM modulation system to convey high data rates for video transmissions. The limits of the system have been reached by increasing the transmission distance and trying different sets of parameters for the OFDM system.

A data rate of 151 kbps has been achieved in good channel conditions, and 91 kbps has been the limit otherwise. Both bit rates are suffcient to allow the transmission of real-time low quality video, at frame rates between 4 and 20 fps depending on the resolution and the MPEG-4 compression ratio.

type: Full theses / dissertations

Parent Project

Project No.: 2009-R/RCM-26
Title: Wireless Underwater Video Transmission

Other publications by this author (as Lead)