In the fog network composed of dense terminal devices and fog servers, how to reduce the system energy consumption during task offloading is a challenging problem. To solve this problem, this article first formulates the energy consumption minimization problem of partial task offloading under delay constraints with dynamic voltage scaling (DVS) and transmission power control (TPC) techniques. Second, this problem was decomposed into two subproblems to solve: 1) the partial task offloading problem with optimal energy consumption under known matching between the terminal device and fog server and 2) the optimal matching problem between terminal devices and fog servers. For the first subproblem, the optimal solution is obtained through theoretical derivation, and the EOPCO-S algorithm is proposed to solve it. For the second subproblem, we transform the original problem into a weighted bipartite graph matching problem and propose the Kuhn–Munkres-based EOPCO-M algorithm to solve it. Finally, numerical simulations are carried out to verify the theoretical derivation and the effectiveness of the proposed algorithms. Experimental results show that the proposed algorithm can significantly reduce the energy consumption of fog networks compared with several baseline algorithms.

KSII Transactions on Internet and Information Systems TIIS Paper Details

Low earth orbit Internet of things (LEO-IoT) networks offer global coverage and low-latency communication, making them attractive for large-scale IoT deployments. However, rapidly varying satellite connectivity and uneven, burst ground traffic lead to unstable routing performance, resulting in fluctuating delays and increased packet loss. To address these challenges, we propose a delay-aware routing optimization (DARO) algorithm that integrates traffic prediction and distributed control. A CNN-BiLSTM-Attention model is developed to capture spatial-temporal traffic patterns, enhancing the accuracy of dynamic traffic prediction. A closed-form end-to-end delay model is derived to characterize the effect of routing decisions on network latency. The routing problem is then formulated as a delay and packet loss minimization task and solved using multi-agent proximal policy optimization (MAPPO), enabling each satellite to adapt its routing strategy based on local observations and a shared critic. Simulation results show that DARO improves prediction accuracy by 19.05% to 76.39%, reduces packet loss by 27.35% to 90.76%, and lowers end-to-end delay by 2.43% to 56.04%, demonstrating its effectiveness in dynamic LEO-IoT environments.

Existing dependent task offloading approaches mainly rely on centralized algorithms and are typically constrained to single-objective optimization. In addition, offloading fairness among fog servers remains insufficiently addressed. To tackle these challenges, this paper proposes a novel distributed method that integrates Gale–Shapley Matching (GSM) with Improved Particle Swarm Optimization (IPSO), whose goal is to maximize the offloading gain on the user side and the fairness on the fog server side. Specifically, we firstly formulate an offloading gain maximization problem for dependent task offloading in multi-user and multi-fog computing scenarios, where the offloading gain is defined by jointly optimizing energy consumption and latency of task offloading. Then, we propose a GSM preference list construction method that considers user offloading gain, the number of offloading acceptances by fog servers and channel quality. Finally, GSM is employed to achieve stable user–fog associations, and IPSO is applied to optimize subtask offloading decisions. Experimental results show that, compared with existing algorithms, this method increases the average offloading gain by 9%; and improves the average fairness by 16.6% with minimal loss of user offloading gain. Therefore, the proposed algorithm can jointly optimize the latency and energy consumption of user tasks while ensuring excellent long-term fairness among fog servers.

Wildfire activity has intensified globally, yet the microphysical properties of black carbon (BC) in smoke plumes remain poorly constrained in remote sensing retrievals, limiting accurate assessments of aerosol climate forcing. This study develops an integrated framework combining AERONET measurements with satellite-based fire emission data to characterize BC mixing states under varying combustion intensities. Using a core-shell Mie model jointly constrained by multi-wavelength single-scattering albedo (440–1020 nm) and absorption Ångström exponent (AAE) calculated specifically from 675–1020 nm to minimize brown carbon interference, we retrieve BC core size, shell thickness, and shell-to-core ratio across four emission levels. Results reveal systematic variations with fire intensity: high-intensity combustion produces bimodal SSA distributions (peaks at 0.85 and 0.90) and narrows BC core size distributions (155–190 nm for C emsiion Level 4 versus 140–200 nm for C emsiion Level 1). Joint AAE-SSA constraints yield shell-to-core ratios of 2.9–3.8, significantly higher than SSA-only retrievals and consistent with field measurements. Regional analysis demonstrates that humidity exerts stronger control over coating thickness than emission intensity. Sensitivity tests excluding 440 nm SSA produce non-physical solutions under low-intensity conditions, confirming the value of short-wavelength constraints. This empirically grounded framework advances BC microphysical characterization from routine optical measurements, reducing key uncertainties for aerosol-climate interaction assessments.