Energy-efficient data gathering remains a fundamental challenge in Underwater Wireless Sensor Networks (UWSNs) due to the inherent limitations of multi-hop communication, which often result in excessive energy depletion near the sink, premature network partition, and degraded data collection performance. This paper proposes GA-UPNR (Genetic Algorithm Approach for UGV Path and Node Routing), a novel shoreside data collection framework that decomposes the problem into two optimization subproblems. The first genetic algorithm constructs energy-balanced multi-hop routing trees from surface nodes to distributed shoreside nodes to maximize network lifetime. The second genetic algorithm determines the optimal set of stopping points for an Unmanned Ground Vehicle (UGV), minimizing the travel distance required to collect data from all shoreside nodes. The two algorithms operate independently and are executed sequentially, providing a scalable solution for efficient data retrieval. Simulation results with 500 nodes indicate that GA-UPNR-Routing achieves longer network lifetime and higher connectivity compared to BMBS, SS-Dijkstra, and MS-Dijkstra. Specifically, GA-UPNR-Routing achieves a network lifetime of 40.14 rounds, in contrast to 21.16, 17.01, and 5.25 rounds for BMBS, MS-Dijkstra, and SS-Dijkstra, respectively. For the UGV stopping point selection, GA-UPNR-Path requires an average of 9.37 stops, whereas MCF and NPS require 11.07 and 105.61 stops, respectively. These results suggest that the GA-UPNR framework is suitable for scalable data collection in long-term marine monitoring applications.

The problem of computing batches of matrix multiplications in distributed computing systems with stragglers is studied. Unlike existing works in the literature, the matrices in a batch are assumed to be sparse, and the sparsity levels for matrices in different batches can be different. A novel coding scheme, called generalized sparse code (GSC), is proposed, in which the matrices are partitioned into smaller chunks that are re- grouped and encoded by respective sparse codes. The expected runtime of the proposed GSC scheme is analyzed, based on which a task assignment problem associated with the proposed GSC is formulated and solved. The solution follows the reverse water-filling principle, by which an efficient worker assignment algorithm whose worst-case time complexity equal to the total number of workers can be developed. Simulation results validate the advantage of the proposed GSC over four existing schemes, including entangled polynomial codes (EP), generalized cross-subspace alignment (GCSA), Lagrange coded computing (LCC) codes and factored Luby transform (FLT) codes at all sparsity levels. As a potential application of the proposed GSC, the problem of computing a batch of matrix multiplications with similarity is discussed.

An image may be disturbed by impulse noise during transmission or acquisition. To effectively restore the disturbed image is important for the applications of image processing. This study aims at enhancing the disturbed images by using the convolutional neural network (CNN) to identify similar patterns for the restoration of noisy pixels. In the training phase, each noisy pixel is analysed and compared with the noise‐free image to find the closest neighbouring pixels. The pixels in a local window form a micro‐pattern. All the captured micro‐patterns, whose centre pixel is noisy, become a dataset for the training of a position CNN. The closest neighbouring pixel of a noisy image to the centre one of the noise‐free image at the same position of each micro‐pattern is selected to be the target. In the enhancement phase, a noisy micro‐pattern, where the centre pixel is noisy, is input into the trained position CNN. The top N pixels are recognised and averaged to replace the grey level of the centre pixel. An enhanced pixel is obtained. The experimental results show that the position CNN can well recognise the similar neighbouring pixels and effectively enhance the noisy pixels in an image disturbed by salt‐and‐pepper noise.