Abstract:In this paper, the dependency of wave spectrum estimation on the analysis window orientation for X-band marine radar data is investigated. The investigation is made for the case where both wind wave and swell components are present. Our results show that more accurate and reliable wave spectrum estimates are obtained when one analysis window for each component (wind waves or swell) is used and is oriented in the up-wave direction of that component. The final wave spectrum estimation is found by averaging the output of those analysis windows. Since the direction of wind waves and swell is not known a priori, a new method is proposed to recursively determine the number and orientation of analysis windows. This method is referred to as the Adaptive Recursive Positioning Method (ARPM). For validation, ocean wave spectra are estimated from X-band marine radar field data using the ARPM and the standard method, using uniformly distributed analysis windows. The results from both methods are compared to ground truth wave spectra acquired using waverider buoy data. Results have shown that the ARPM produces a 10% improvement in the agreement (represented by the correlation coefficient) between the ground truth wave spectrum and the marine radar estimated wave spectrum. This improvement is reflected in a 15% to 30% accuracy enhancement in the wave period estimation and 6 • in peak wave direction. The ARPM not only increases the accuracy of wave period and direction estimation, but it also increases the method's reliability by producing a lower error standard deviation. Although these improvements are at the price of extra computational time, it has been found that this overhead is acceptable since it is far from the upper bound that requires offline analysis.

No abstract

No abstract

In this paper, a new method is proposed to estimate surface current using X-band nautical radar. This method is referred to as the Hybrid Least Squares (HLS) method. The HLS combines the existing Iterative Least Squares (ILS) and Normalized Scalar Product (NSP) methods. The HLS is designed to inherit the short ILS computational time and the high NSP reliability. For validation, the proposed method was applied on a number of simulated X-band nautical radar image sets. Results have shown that the HLS produces accurate estimates while maintaining very short computational time.

No abstract