The cognitive radio is growing an important interest in wireless communication study. A ad hoc network may take a master-slave tree-based structure some special applications. For master node with limited capability, slave nodes usually use the same frequency to access subnet managed by master. Each can acquire many frequencies for local spectrum sensing process. However, there be no common set of available every node. In this case, we should delete and keep other staying as possible. By...

Reinforcement learning methods are more and popular in machine fields, have achieved outstanding breakthroughs various applications combined with deep convolution neural networks. Deep networks excellent fitting abilities promote the prosperity of learning. However, we want to look inside model get some rules sometime, but it is a hard task for reinforcement now. Our fuzzy tree unites methods, decision tree, could extract explicit keep performance at same time. paper show that models based...

Cognitive Frequency Decision Making (CFDM) is a new application in cognitive radio ad hoc network with limited communication capability, and once solved by our algorithm Extreme Maximal Biclique Searcher (EMBS). In this paper, we extend the CFDM from one subnet to whole network, propose Common (CFS) find solution. CFS uses result of novel Weighted Frequent Itemset Mining (MWFIM) which mainly discussed paper also proposed us mine all maximal weighted frequent itemsets transaction database...

Backoff algorithm is a key component of contention-based MAC layer protocol. Numbers dynamic CW adjustment methods have been proposed for the optimal throughput and fairness. However, they only utilize half-feedback information may suffer from diverging problem. In this paper, we propose novel Full-feedback Contention Window Adjustment (FCWA) backoff algorithm. Simulation results demonstrated that FCWA provides remarkable performance improvement in terms short-term fairness, packet delay...

This paper formulates a new QoS model for tree-based cognitive radio network (CRN) which is under test communication in mountainous areas. The takes spectrum quality into account, results more complex decision involving 3-dimension information: users, spectrums and their combination optimization making. An optimal algorithm with correctness performance proofs proposed to search the best solution among all possible candidates. Each candidate form of extreme maximal bicliques (EMB) viewpoint...

Abstract With the development and wide applications of wireless communication technology, limited spectrum resources fixed allocation policy could no longer satisfy demand for communication. Just this reason, many become holes because they are allocated but not used. Cognitive radio is now becoming one most important techniques high utility these holes. If available to cognitive users abundant over a certain time, it worth consideration increase network throughputs by orthogonal multiplexing...

Utilizing service oriented architectures to enhance seamless collaborations and information sharing among nodes in mobile ad hoc networks with limited communication capability is a challenging task. In this paper novel tree-based discovery mechanism, able achieve high accuracy the process of discovery, proposed, which affords wireless overheads nearly linear number property-changing services offered by whole network. The mechanism suitable customized for typical networks. As properties such...

Cognitive radio (CR) attempts to improve spectrum utility by exploiting whitespaces in the spectral and time domains. However, different or domains may provide communication qualities. Distinguishing best among a large number of candidates is e xpensive terms energy has yet be fully studied literature. This paper presents sensing framework based on channel usability patterns mined from actual experimental data address this problem. In contrast prediction techniques that simply regard as idle...

Wireless networks will provide much more convenience for mobile users via heterogeneous wireless architectures and dynamic spectrum access techniques. However, also impose challenges on developers due to unknowing global electromagnetic environment information in each node, as well diverse QoS requirements of various applications. Therefore, both correctness reliability communication protocol become extremely difficult. In years, model checking has been advertised a way ensure the complex...

Backoff mechanism is a key component of contention-based medium access control (MAC) layer protocol. It has been shown that the backoff IEEE 802.11 standard may be very inefficient especially when network congested. Numbers methods have proposed to tune contention window (CW) with aim achieve optimal throughput in WLANs. However, mechanisms do not specifically address proper settings for variable packet length influence and CW diverging problem. This paper proposes novel four-way handshaking...

Uncertain big data has become increasingly important in our life, since uncertainties exist generation/acquisition, physical measurements and staling. Thus, this paper, we aim to create a unified mathematical model between relation uncertainties, which are both aspects of uncertain data. We first proposed the fusion uncertainty uncertainty, consists model. Then defined two fundamental forms for model, open closed one-side coupling respectively. After that picked part called discuss. studied...

Many tasks of complex data relation mining in social networks, bioinformatics and cognitive radios usually involve a process enumerating maximal bicliques from relationship graph which is generally represented by matrix. Currently most researchers focus on the large static keep unchanged through enumeration big mining. But situations, will be dynamically changed since either vertices or edges removed updated to reflect corresponding change real world. Therefore, it impossible employ those...

Communities are clusters of closely connected nodes in a social network. Detecting community structure can help us understand their network characteristics. Most popular algorithms based on modularity optimization, such as the SLM algorithm. The algorithm detect non-overlapping communities. However, communities real-world networks also overlap maybe belong to multiple clusters. Some models BIGCLAM be used discover overlapping structure, but it has some problems running detection large-scale...

How to determine an optimal topology is extremely important during the spectrum decision procedure for a cognitive radio network. In multi-transceiver network, transceivers of user will work in parallel mode, by which throughput network can be increased. Each working such mode should keep electromagnetic compatible (EMC) increase time-complexity dramatically. order solve this problem, approximate model with lower time complexity given. To evaluate reliability model, method EMC probability...

In this article we provide a cognitive system that combines the traditional spectrum sensing process and data mining together. The aims to enhance utilization using mining. It includes three aspects: pre-processing, results validation. experimental demonstrate feasibility of such model.