With the rapid growth of Sentinel-1 synthetic aperture radar (SAR) data, how to exploit Sentinel-1 imagery and achieve effective and robust marine surveillance are crucial problems. In this paper, we present the OpenSARShip, a dataset dedicated to Sentinel-1 ship interpretation. The OpenSARShip, providing 11 346 SAR ship chips integrated with automatic identification system messages, owes five essential properties: specificality, large scale, diversity, reliability, and public availability. These properties make sure that the OpenSARShip achieves its objectives. The first is to provide researchers a benchmark dataset to develop applicable and adaptive ship interpretation algorithms and push the performance ceilings of data analysis. The other is to provide a dataset for performing application-oriented quality assessment for Sentinel-1 imagery, which can boost their applications in a targeting way. The construction and the organization of the OpenSARShip are discussed, which show the inside of the dataset and ensure the essential properties. The elaborate geometric and scattering analyses, the benchmark for classification, and the imagery applicability assessment by using the OpenSARShip all demonstrate the applicability and potential of the dataset.

Ship detection from remote sensing images can provide important information for maritime reconnaissance and surveillance and is also a challenging task. Although previous detection methods including some advanced ones based on deep convolutional neural network expertize in detecting horizontal or nearly horizontal targets, they cannot give satisfying detection results for arbitrary-oriented ship detection. In this letter, we introduce a novel ship detection system that can detect arbitrary-oriented ships. In this method, a rotated region proposal networks (R ² PN) is proposed to generate multiorientated proposals with ship orientation angle information. In R ² PN, the orientation angles of bounding boxes are also regressed to make the inclined ship region proposals generated more accurately. For ship discrimination, a rotated region of interest pooling layer is adopted in the following classification subnetwork to extract discriminative features from such inclined candidate regions. The proposed whole ship detection system can be trained end to end. Experimental results conducted on our rotated ship data set and HRSD2016 benchmark demonstrate that our proposed method outperforms state-of-the-art approaches for the arbitrary-oriented ship detection task.

We investigate sensing-assisted beamforming for vehicle-to-infrastructure (V2I) communication by exploiting integrated sensing and communications (ISAC) functionalities at the roadside unit (RSU). The RSU deploys a massive multi-input-multi-output (mMIMO) array at mmWave. The pencil-sharp mMIMO beams and fine range-resolution implicate that the point-target assumption is impractical, as the vehicle’s geometry becomes essential. Therefore, the communication receiver (CR) may never lie in the beam, even when the vehicle is accurately tracked. To tackle this problem, we consider the extended target with two novel schemes. For the first scheme, the beamwidth is adjusted in real-time to cover the entire vehicle, followed by an extended Kalman filter to predict and track the position of CR according to resolved scatterers. An upgraded scheme is proposed by splitting each transmission block into two stages. The first stage is exploited for ISAC with a wide beam. Based on the sensed results at the first stage, the second stage is dedicated to communication with a pencil-sharp beam, yielding significant communication improvements. We reveal the inherent tradeoff between the two stages in terms of their durations, and develop an optimal allocation strategy that maximizes the average achievable rate. Finally, simulations verify the superiorities of proposed schemes over state-of-the-art methods.

In order for an orthogonal frequency division multiplexing (OFDM)-based radar and communication (RadCom) system to perform well, accurate range and velocity estimation needs to be realized. To achieve this goal, the intrapulse and intersubcarrier Doppler effects that are both typically ignored in conventional methods must be considered and properly utilized. In this paper, we first conceive of a complete receiving model considering the intrapulse and intersubcarrier Doppler effects. Then, an alternating projection - maximum likelihood (AP-ML) method is developed based on this complete model. Considering the heavy computational burden of the AP-ML method, the use of the expectation maximization (EM) algorithm is introduced to reduce the computational complexity. Simultaneously, the computational cost of the EM algorithm can be further reduced by using the reduced-rank technique. Consequently, better estimation results can be obtained with this reduced-rank EM algorithm when there are no approximation errors. Moreover, an analytical expression of the Cramer-Rao bound (CRB) is provided for the complete receiving model. The mutual impacts between the radar and communication functions are analyzed in detail. Finally, excellent experimental results illustrate the superiority of our proposed methods.

The detection and recognition of ships holds crucial significance across various domains, including transportation, fishing, smuggling prevention, and rescue operations. To achieve accurate and efficient ship detection and recognition, remote sensing technologies, specifically, optical and synthetic aperture radar (SAR) sensors, have been widely employed. However, each sensor modality presents its own limitations, such as cloud cover in optical images and interference in SAR images, resulting in restricted data availability and false alarms during detection. This article provides a comprehensive summary of recent advancements in the fusion of optical and SAR images for ship detection and recognition. Meanwhile, the fusion of remote sensing images and in situ data, such as automatic identification system (AIS) data, is briefly summarized, considering significant benefits and technical trends. This article highlights the immense potential of integrating these two sensor modalities to overcome individual challenges and enhance overall system performance. One of the primary challenges addressed in this work pertains to accurately matching ship targets between optical and SAR images when ships change positions. However, recently developed multimodal fusion networks and transfer learning techniques have the potential to tackle the challenges, even with unmatched optical and SAR image pairs. In addition to summarizing existing techniques, the article delves into future perspectives on fusion-based ship detection and recognition. In conclusion, this survey highlights the current state of optical and SAR image fusion techniques, pinpoints existing challenges, and outlines future research prospects, emphasizing the significance of fusion-based approaches in shaping the future of maritime applications.