As user equipments (UEs) and infrastructure evolve, the network architecture becomes increasingly heterogeneous, presenting challenges in managing diverse communication technologies and varied network demands. Recognizing these challenges, the integration of unmanned aerial vehicles (UAVs) and device-to-device (D2D) communication emerges as an effective strategy to enhance network capacity and address the complexities of a heterogeneous network. In this context, mobile edge computing (MEC), facilitated by servers at the network edge, significantly reduces data processing latency and alleviates backhaul congestion. This paper proposes a UAV and D2D-enabled heterogeneous MEC framework that supports two offloading modes for UEs: the first is to offload tasks to UAVs equipped with computing servers, and the second is to offload tasks to nearby idle UEs via D2D links. Unlike existing schemes, the selection of offloading modes is determined by the network load, which improves resource utilization. To further reduce computing latency, we formulate an optimization problem that jointly considers UAV trajectory planning, resource management, and task offloading policy. Since the problem is non-convex and challenging to solve in polynomial time, we decompose it into a joint resource optimization and an offload scheme optimization. Based on this, an optimization algorithm based on block descent and potential game is proposed to obtain the optimal solution. Simulation results indicate that compared to typical UAV-based MEC without D2D, our solution can reduce latency by about 20%. In comparison to equivalent resource allocation, the proposed algorithm can reduce latency by about 25%.

The K-means algorithm has been extensively investigated in the field of text clustering because of its linear time complexity and adaptation to sparse matrix data. However, it has two main problems, namely, the determination of the number of clusters and the location of the initial cluster centres. In this study, we propose an improved K-means++ algorithm based on the Davies-Bouldin index (DBI) and the largest sum of distance called the SDK-means++ algorithm. Firstly, we use the term frequency-inverse document frequency to represent the data set. Secondly, we measure the distance between objects by cosine similarity. Thirdly, the initial cluster centres are selected by comparing the distance to existing initial cluster centres and the maximum density. Fourthly, clustering results are obtained using the K-means++ method. Lastly, DBI is used to obtain optimal clustering results automatically. Experimental results on real bank transaction volume data sets show that the SDK-means++ algorithm is more effective and efficient than two other algorithms in organising large financial text data sets. The F-measure value of the proposed algorithm is 0.97. The running time of the SDK-means++ algorithm is reduced by 42.9% and 22.4% compared with that for K-means and K-means++ algorithms, respectively.

A new payload concept has been proposed: the microwave imager combined active/passive (MICAP). MICAP is a combined one-dimensional microwave interferometric radiometer that operates at 1.4, 6.9, 18.7, and 23.8 GHz and L-band (1.26 GHz) scatterometer. It has the capability to simultaneously remotely sense sea surface salinity (SSS), sea surface temperature (SST), and wind speed. MICAP will be a candidate payload onboard the Ocean Salinity Satellite led by the State Oceanic Administration of China to monitor SSS and reduce geophysical errors caused by surface roughness and SST. To provide an “all-weather” estimation of SSS with high accuracy from space, the errors of the simultaneous retrieval of multiparameters using MICAP are analyzed, and noise levels and stability requirements of instruments are estimated. Preliminary analysis shows that MICAP can provide SSS with an accuracy of 1 psu for single measurement and 0.1 psu over the global ocean for 200 × 200 km resolution pixels and one month at middle and low latitudes with default instrument noises (0.1 K, 0.3 K and 0.3 K for the L, C, and K band radiometers, respectively, and 0.1 dB for the L-band scatterometer) while uncertainties of the drift corrections are less than the radiometer sensitivities.

Accurate server traffic prediction can help enterprises formulate network resource allocation strategies in advance and reduce the probability of network congestion. Traditional prediction models ignore the unique data characteristics of server traffic that can be used to optimize the prediction model, so they often cannot meet the long-term and high-precision prediction required by server traffic prediction. To solve this problem, this paper establishes a hybrid model ARIMA-LSTM-CF, which combines the advantages of linear and nonlinear models, as well as the periodic fluctuation characteristics of server traffic data obtained from banks. In addition, this paper also uses the optimized K-means clustering method to extract the traffic data of workdays and non workdays. The results show that the new hybrid model performs better than the single ARIMA and LSTM models in predicting the long-term trend of server traffic. RMSE (root mean square error) and MAE (mean absolute error) are reduced by 50%. R2 score index reached 0.64. The results show that the model can effectively extract the data characteristics of server traffic data, and the model has accurate and stable long-term prediction ability.

Due to the lack of spectrum resources, effective methods must be proposed to meet the needs of the Internet of Things (IoT) massive connectivity. In this paper, two effective techniques to improve spectral efficiency, namely, sparse code multiple access (SCMA) and wireless multicast, are used together for the IoT system to expand network capacity. Different from existing works, we introduce layered coding into the SCMA-based IoT system and consider two channel state information (CSI) feedback scenarios, which can make full use of the diversity gain of multiple receivers. The original multicast data are divided into one basic layer (BL) and several enhancement layers (ELs) by layered coding. IoT receivers with good channel gain can achieve better quality of service by receiving more data layers. In the transmission scheme with CSI, the physical resources allocated to the BL and ELs can be dynamically adjusted. In the transmission scheme without CSI, a fixed resource allocation scheme is performed because it is impossible to calculate the multicast rate of each codebook in real time. To further improve the system capacity, we have proposed resource allocation algorithms for these two schemes according to different requirements for computational complexity. The simulation results show that the proposed schemes can provide a higher system capacity than unicast SCMA and nonlayered multicast SCMA.

The derivation of ice water path (IWP) from microwave (MW) radiometer measurements is challenging. This study presents a deep learning framework for global retrieval of IWP using observations from the Microwave Humidity Sounder-II (MWHS-II) aboard the FengYun-3D (FY-3D) satellites. Two deep learning models, deep forest (DF21) and quantile regression neural network (QRNN), are constructed to detect ice cloud flags and retrieve IWP. By collocating MWHS-II observations with 2C-ICE, a joint product of CloudSat and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO), deep learning models learn the characteristics of IWP from MWHS-II brightness temperatures (BTs). The test results show that the MWHS-II channels provide more information on IWP than the MWHS channels, particularly the 89-GHz channel and the 118-GHz channels with an offset of $\geq 0.8$ GHz. Combining the QRNN and DF21 models, the IWP retrieval results in a root mean square error (RMSE) of 707.346 g/m2, mean absolute percentage error (MAPE) of 65.122%, mean bias error (MBE) of −104 g/m2, determination coefficient ( $R^{2}$ ) of 0.683, and Pearson correlation coefficient (PCC) of 0.831. Application of the models to MWHS-II observations of tropical cyclone CILIDA shows better agreement with 2C-ICE. All the datasets exhibit a similar feature on the monthly mean scale, but the magnitudes of IWP differ. Compared with GMI-GPROF, MODIS, and ERA5 IWP products, MWHS-II results are closest to 2C-ICE. Similar results are also shown for the zonal mean data. These results show that deep learning methods efficiently and probabilistically retrieve IWP from long-term observation data of MWHS/MWHS-II.