Shallow-water acoustic communication channel, referred to as a multipath-limited channel, produces inter-symbol interference that poses a significant obstacle to reliable communication. Accordingly, signal-to-multipath ratio (SMR), rather than signal-to-noise ratio (SNR), becomes an important factor affecting communication performance. However, it is difficult to estimate SMR from measured communication data, especially at higher frequency (>10 kHz) because many arrivals scattered from rough ocean boundaries produce a significant intrapath time spreading, which acts as random noise in communication. In this paper, the energy fraction of the channel impulse response existing in one symbol duration is proposed as a parameter for estimating the quality of shallow-water communication channels. This parameter is compared with the bit-error-rate performance for data acquired in shallow water off the south coast of Korea, where the water depth is 45 m and the bottom consists of sandy clay sediment. The results imply that the energy fraction in one symbol duration may be used as a parameter for describing shallow-water communication channels and applied to the quick decision of a symbol or bit rate in a shallow-water field for reliable underwater communication.

Acoustic Doppler current profilers (ADCPs) were developed to acquire water current velocities, as well as depth-dependent echo intensities. As the backscattering strength of an underwater object can be estimated from the measured echo intensity, the ADCP can be used to estimate plankton populations and distributions. In this study, the backscattering strength of bubble clusters in a water tank was estimated using the commercial ADCP as a proof-of-concept. Specifically, the temporal variations in the backscattering strength and the duration of bubble existence were quantitatively evaluated. Additionally, the PDSL (population density spectrum level) and VF (void fraction) of the artificial bubbles were characterized based on the obtained distribution characteristics using a PDPA (phase Doppler particle analyzer).

Measurements of bottom backscattering strengths in a frequency range of 6-14 kHz were made on the shallow water off the southern Gyeonggi Bay in Yellow Sea in May 2013, as part of the KIOST-HYU joint acoustics experiment. Geological surveys for the experimental area were performed using multi-beam echo sounder, sparker system, and grab sampling to investigate the bottom topography, sub-bottom profile and composition of surficial sediment, respectively. In this paper, the backscattering strengths as a function of grazing angle (in range of 28° ~ 69°) were estimated and compared to the predictions obtained by Lambert's law and APL-UW scattering model. Finally, the effects of geoacoustic parameters corresponding to the experimental area on the backscattering strengths are discussed.

Recently, neural network-based deep learning techniques have been actively applied to detect underwater objects in sonar (sound navigation and ranging) images. However, unlike optical images, acquiring sonar images is extremely time- and cost-intensive, and therefore securing sonar data and conducting related research can be rather challenging. Here, a side-scan sonar was used to obtain sonar images to detect underwater objects off the coast of the Korean Peninsula. For the detection experiments, we used an underwater mock-up model with a similar size, shape, material, and acoustic characteristics to the target object that we wished to detect. We acquired various side-scan sonar images of the mock-up object against the background of mud, sand, and rock to account for the different characteristics of the coastal and seafloor environments of the Korean Peninsula. To construct a detection network suitable for the obtained sonar images from the experiment, the performance of five types of feature extraction networks and two types of optimizers was analyzed. From the analysis results, it was confirmed that performance was achieved when DarkNet-19 was used as the feature extraction network, and ADAM was applied as the optimizer. However, it is possible that there are feature extraction network and optimizer that are more suitable for our sonar images. Therefore, further research is needed. In addition, it is expected that the performance of the modified detection network can be more improved if additional images are obtained.

Time reversal processes have been used to improve communication performance in the severe underwater communication environment characterized by significant multipath channels by reducing inter-symbol interference and increasing signal-to-noise ratio. In general, the performance of the time reversal is strongly related to the behavior of the q-function, which is estimated by a sum of the autocorrelation of the channel impulse response for each channel in the receiver array. The q-function depends on the complexity of the communication channel, the number of channel elements and their spacing. A q-function with a high side-lobe level and a main-lobe width wider than the symbol duration creates a residual ISI (inter-symbol interference), which makes communication difficult even after time reversal is applied. In this paper, we propose a new parameter, E q , to describe the performance of time reversal communication. E q is an estimate of how much of the q-function lies within one symbol duration. The values of E q were estimated using communication data acquired at two different sites: one in which the sound speed ratio of sediment to water was less than unity and one where the ratio was higher than unity. Finally, the parameter E q was compared to the bit error rate and the output signal-to-noise ratio obtained after the time reversal operation. The results show that these parameters are strongly correlated to the parameter E q .