The large-scale coverage of LEO satellite networks and the enhancement of onboard satellite computing resources have emerged as a pivotal solution for meeting the computing and routing demands of remote sensing (RS) tasks in areas without ground network coverage. Nonetheles, efficiently leveraging LEO satellite network resources effective computation offloading and relay routing presents substantial challenges. In this study, we delve into the joint optimization of task offloading and routing path selection within LEO satellite networks, aiming to maximize long-term user quality of service while adhering to energy constraints. We propose a novel integrated modeling scheme for relay routing and computation offloading of RS tasks in LEO satellite networks, effectively reducing the scale of system decision problems. Furthermore, we devise a multi-step reward aggregation Deep Q-Learning (DQN-MCAR) algorithm for computing-aware routing, effectively addressing the sequential learning problem among the computing-aware routing decisions of subtasks. Finally, theoretical analysis confirms that our proposed algorithm has lower complexity, and the superiority of our scheme is validated by extensive simulation experiments.

As numerous latency-sensitive applications have emerged with the popularization of 5G networking, accurate and rapid end-to-end latency measurement has come to play an essential role in network fault diagnosis and optimization. Although the Bloom hash-based timestamp aggregation method has been reported to be scalable and efficient, two shortcomings that reduce its accuracy have yet to be fully considered: frequent hash collisions and its fixed measurement interval. To address these challenges, we construct an end-to-end network latency measurement framework named Cuckoo Hash Adjustive Table exchange (CHAT). By employing an improved cuckoo filter, we decrease the number of hash collisions to assess the latency more accurately. Moreover, CHAT adjusts the receiver-side measurement interval dynamically based on a gain indicator, maximizing the total number of valid packets used for latency estimation. Additionally, the proposed measurement framework minimizes the number of packets transferred over links to avoid interfering with the end-to-end latency measurement in an actual network. Finally, extensive experiments on simulations and a practical real-world environment show the effectiveness and applicability of CHAT.