Search

This paper proposes a flight data-driven model to identify air traffic flow patterns based on multi-layer cluster analysis for identifying and characterizing traffic flow patterns in the airspace of the Beijing-Tianjin-Hebei multi-airport system. The spatial and temporal trends of flight data are mined to capture the air traffic flow characteristics, and the busyness of waypoints help air traffic managers to understand the reuse status of runways and airspace for capacity planning decision support in the complex airspace of Beijing-Tianjin-Hebei.

Hepatocellular carcinoma is one of the most common malignant tumors in the world with complex etiology, high tumor heterogeneity, and low efficacy of treatment. The establishment of an animal model that is close to the clinical situation of hepatocellular carcinoma and can be successfully modeled many times is of great significance to the study of the pathogenesis, diagnosis and treatment of hepatocellular carcinoma. We used Diethylnitrosamine (DEN) to induce hepatocellular carcinoma in rodents and compared four models of DEN-induced hepatocellular carcinoma. Group C (Control): rats were fed a standard laboratory rat diet and freely drank normal water. Group P (Peritoneal injection): rats were administered an IP injection (50 mg/kg/week) between 5 and 23 weeks after birth. Ten microliter of DEN solution would be injected per g of rat. Group O (DEN-Fed group): rats were allowed unrestricted access to water contaminated with 0.01 % DEN between the ages of 7 and 15 weeks. 0.2 mL of DEN drinking water was consumed per gram of rat. Group P+O (Combined peritoneal injection and Oral feeding): rats were administered an IP injection (50 mg/kg) at weeks 3 and 5 post-birth, and they freely drank water contaminated with 0.012 % DEN between weeks 7 and 14 post-birth. We used an ultrasound scan, biochemical testing, haematoxylin, and eosin staining, Masson staining, Wolf scarlet staining, Ki67, CD34, a-SMA, CD8, and CD68 staining to compare between groups. Liver dissection and ultrasound scan showed that compared to other groups, the liver of Group P+O was darker in color, with more grey-white cancer nodules and larger localized tumors, and the structure of the tumors was slightly disorganized, with the elastography hardness of the middle lobe and the right lobe was slightly increased. The alanine aminotransferase and total bilirubin of Group P+O were higher than those of Group O but lower than those of Group P. Haematoxylin and eosin staining showed that the tumors of Group P+O were large, with large tumor cords and pseudo-glandular, the degree of differentiation was medium and surrounded by more fatty lesions. We conclude that combined DEN treatment is more effective, stable, and has the advantage of multiple modalities, leading to faster tumor formation. [ABSTRACT FROM AUTHOR]

Device-to-device (D2D) communication is a promising wireless communication technology which can effectively reduce the traffic load of the base station and improve the spectral efficiency. The application of intelligent reflective surfaces (IRS) in D2D communication systems can further improve the throughput, but the problem of interference suppression becomes more complex and challenging due to the introduction of new links. Therefore, how to perform effective and low-complexity optimal radio resource allocation is still a problem to be solved in IRS-assisted D2D communication systems. To this end, a low-complexity power and phase shift joint optimization algorithm based on particle swarm optimization is proposed in this paper. First, a multivariable joint optimization problem for the uplink cellular network with IRS-assisted D2D communication is established, where multiple DUEs are allowed to share a CUE's sub-channel. However, the proposed problem considering the joint optimization of power and phase shift, with the objective of maximizing the system sum rate and the constraints of the minimum user signal-to-interference-plus-noise ratio (SINR), is a non-convex non-linear model and is hard to solve. Different from the existing work, instead of decomposing this optimization problem into two sub-problems and optimizing the two variables separately, we jointly optimize them based on Particle Swarm Optimization (PSO). Then, a fitness function with a penalty term is established, and a penalty value priority update scheme is designed for discrete phase shift optimization variables and continuous power optimization variables. Finally, the performance analysis and simulation results show that the proposed algorithm is close to the iterative algorithm in terms of sum rate, but lower in power consumption. In particular, when the number of D2D users is four, the power consumption is reduced by 20%. In addition, compared with PSO and distributed PSO, the sum rate of the proposed algorithm increases by about 10.2% and 38.3%, respectively, when the number of D2D users is four. [ABSTRACT FROM AUTHOR]