Beijing University of Posts and Telecommunications

About: Beijing University of Posts and Telecommunications is a education organization based out in Beijing, Beijing, China. It is known for research contribution in the topics: MIMO & Quality of service. The organization has 39576 authors who have published 41525 publications receiving 403759 citations. The organization is also known as: BUPT.

Resource Trading in Blockchain-Based Industrial Internet of Things

Abstract: Past few years have witnessed the compelling applications of the blockchain technique in our daily life ranging from the financial market to health care. Considering the integration of the blockchain technique and the industrial Internet of Things (IoT), blockchain may act as a distributed ledger for beneficially establishing a decentralized autonomous trading platform for industrial IoT (IIoT) networks. However, the power and computation constraints prevent IoT devices from directly participating in this proof-of-work process. As a remedy, in this treatise, the cloud computing service is introduced into the blockchain platform for the sake of assisting to offload computational task from the IIoT network itself. In addition, we study the resource management and pricing problem between the cloud provider and miners. More explicitly, we model the interaction between the cloud provider and miners as a Stackelberg game, where the leader, i.e., cloud provider, makes the price first, and then miners act as the followers. Moreover, in order to find the Nash equilibrium of the proposed Stackelberg game, a multiagent reinforcement learning algorithm is conceived for searching the near-optimal policy. Finally, extensive simulations are conducted to evaluate our proposed algorithm in comparison to some state-of-the-art schemes.

Tungsten disulfide saturable absorbers for 67 fs mode-locked erbium-doped fiber lasers.

Abstract: In this paper, we demonstrate 67 fs pulse emitting with tungsten disulfide (WS2) in mode-locked erbium-doped fiber (EDF) lasers. Using the pulsed laser deposition method, WS2 is deposited on the surface of the tapered fiber to form the evanescent field. The fiber-taper WS2 saturable absorber (SA) with the large modulation depth is fabricated to support the ultrashort pulse generation. The influences of the WS2 SA are analyzed through contrastive experiments on fiber lasers with or without the WS2 SA. The pulse duration is measured to be 67 fs, which is the shortest pulse duration obtained in the mode-locked fiber lasers with two dimensional (2D) material SAs. Compared to graphene, topological insulator, and other transition metal dichalcogenides (TMDs) SAs, results in this paper indicate that the fiber-taper WS2 SA with large modulation depth is a more promising photonic device in mode-locked fiber lasers with the wide spectrum and ultrashort pulse duration.

A perturbed particle swarm algorithm for numerical optimization

Abstract: The canonical particle swarm optimization (PSO) has its own disadvantages, such as the high speed of convergence which often implies a rapid loss of diversity during the optimization process, which inevitably leads to undesirable premature convergence. In order to overcome the disadvantage of PSO, a perturbed particle swarm algorithm (pPSA) is presented based on the new particle updating strategy which is based upon the concept of perturbed global best to deal with the problem of premature convergence and diversity maintenance within the swarm. A linear model and a random model together with the initial max-min model are provided to understand and analyze the uncertainty of perturbed particle updating strategy. pPSA is validated using 12 standard test functions. The preliminary results indicate that pPSO performs much better than PSO both in quality of solutions and robustness and comparable with GCPSO. The experiments confirm us that the perturbed particle updating strategy is an encouraging strategy for stochastic heuristic algorithms and the max-min model is a promising model on the concept of possibility measure.

A Survey on the Scalability of Blockchain Systems

Abstract: In the past decade, crypto-currencies such as Bitcoin and Litecoin have developed rapidly. Blockchain as the underlying technology of these digital crypto-currencies has attracted great attention from academia and industry. Blockchain has many good features, such as trust-free, transparency, anonymity, democracy, automation, decentralization and security. Despite these promising features, scalability is still a key barrier when the blockchain technology is widely used in real business environments. In this article, we focus on the scalability issue, and provide a brief survey of recent studies on scalable blockchain systems. We first discuss the scalability issue from the perspectives of throughput, storage and networking. Then, existing enabling technologies for scalable blockchain systems are presented. We also discuss some research challenges and future research directions for scalable blockchain systems.

Sketch-based image retrieval via Siamese convolutional neural network

Abstract: Sketch-based image retrieval (SBIR) is a challenging task due to the ambiguity inherent in sketches when compared with photos. In this paper, we propose a novel convolutional neural network based on Siamese network for SBIR. The main idea is to pull output feature vectors closer for input sketch-image pairs that are labeled as similar, and push them away if irrelevant. This is achieved by jointly tuning two convolutional neural networks which linked by one loss function. Experimental results on Flickr15K demonstrate that the proposed method offers a better performance when compared with several state-of-the-art approaches.