Sanqing Hu

MAC protocol identification using support vector machines for cognitive radio networks

Cognitive radio is regarded as a potential solution to address the spectrum scarcity issue in wireless communication. In CR, an unlicensed network user (secondary user) is enabled to dynamically/adaptively access the frequency channels considering the current state of the external radio environment. In this article, we investigate the medium access control protocol identification for applications in cognitive MAC. MAC protocol identification enables CR users to sense and identify the MAC protocol types of any existing transmissions (primary or secondary users). The identification results will be used by CR users to adaptively change their transmission parameters in order to improve spectrum utilization, as well as to minimize potential interference to primary and other secondary users. MAC protocol identification also facilitates the implementation of communications among heterogeneous CR networks. In this article, we consider four MAC protocols, including TDMA, CSMA/CA, pure ALOHA, and slotted ALOHA, and propose a MAC identification method based on machine learning techniques. Computer simulations are performed to evaluate the MAC identification performance.

MAC protocol identification approach for implement smart cognitive radio

Cognitive radio (CR) is regarded as a solution to address the spectrum scarcity issue in wireless communications. In CR, an unlicensed network user is enabled to dynamically/adaptively access the frequency channels according to the current state of the radio environment and, therefore, CR refers to a smart radio as defined in an “observe-learn-decision” cognitive cycle. In this paper, we consider the medium access control (MAC) protocols as radio parameters in the cognitive cycle, and propose a new approach called MAC protocol identification to implement smart cognitive MAC. The MAC protocol identification approach enables CR users to sense and identify the network MAC protocol types. The identification results will be used by CR users to adaptively change their transmission parameters in order to improve the spectrum hole utilization efficiency, save energy, minimize the interference to the primary users, as well as facilitate communications among heterogeneous CR networks. To verify that the MAC protocol identification is feasible, we propose a MAC identification process based on machine learning techniques in this paper, and some experimental results are presented.

Energy-Efficiency-Based Optimal Relay Selection Scheme With a BER Constraint in Cooperative Cognitive Radio Networks

Due to the energy constraint of each cognitive user and potential secondary transmission errors, energy efficiency is very important for cognitive nodes in spectrum sensing and cooperative transmissions. In this paper, an energy awareness optimal relay selection (EAORS) scheme is proposed, in which the number of optimal relays is determined by a weighted objective function considering energy consumption in both spectrum sensing and the cooperative transmission phases, under the constraints of the global misdetection requirement and secondary cooperative transmission bit error rate (BER). In the proposed EAORS scheme, we investigate the tradeoff between detection accuracy and energy efficiency and construct the objective function as a weighted convex function of relay numbers with the consideration of global missing detection probability and cooperative transmission BER. This nonlinear convex problem is solved by applying numerical analysis to obtain the optimal relay numbers. Energy consumption analysis is conducted to show the effectiveness of the proposed EAORS scheme. Numerical results show that the proposed scheme outperforms a compressed sensing (CS)-based energy-efficient cooperative detection scheme and provides a tradeoff between detection accuracy and energy efficiency.

Slotted Aloha for cognitive radio users and its tagged user analysis

Cognitive radio (CR) is a promising dynamic spectrum access technique to improve the utilization of the licensed spectrum. In this paper, we investigate the medium access control (MAC) protocol for CR networks and propose a cognitive MAC using slotted Aloha to enable CR users to access spectrum holes in a licensed frequency band. In order to avoid the high computational complexities of traditional Markovian analysis, we investigate the tagged user analysis (TUA) based on queuing theories in CR scenarios and use TUA to analyze the proposed cognitive MAC performance. The theoretical derivations based on TUA are presented and computer simulations are performed to evaluate the CR network performance as well as TUA accuracy. The network throughput, packet blocking probability, queue length, packet waiting time and response time are evaluated.

Tagged user approach for finite-user finite-buffer S-Aloha analysis in AWGN and frequency selective fading channels

The tagged user approach (TUA) is shown to be an effective method based on queueing theories to analyze random multiple access systems and avoid the high computational complexities of Markovian analysis. In this paper, we investigate using the TUA to analyze a finite-user finite-buffer slotted Aloha system considering AWGN and frequency selective fading effects respectively. The theoretical derivations based on TUA are presented in this paper and computer simulations are performed to evaluate the TUA accuracy. Performance results of system throughput, packet blocking probability, queue length, packet waiting time and response time are presented.

A novel methodology of data analytics and modeling to evaluate LTE network performance

Big Data and LTE are two hot topics in their respective field. Is there a synergy effect between the two techniques? Traditionally various simulation based approaches are developed to study wireless network performance. However simulation has its own limitations. In this paper, the authors develop a methodology of data analytics and modeling to evaluate LTE network performance based upon traffic measurements and service growth trends. The overall model is finally validated and shows high accuracy, robustness, and reliability.

A study of LTE network performance based on data analytics and statistical modeling

Various simulation based approaches are developed to study wireless network performance and capacity traditionally. However, these simulation based approaches has its own limitations in supervised and unsupervised learning. Since big data techniques become more available, using big data techniques to understood the huge amount of the LTE network measurements and diagnosis data, evaluate and predict the LTE network capacity and performance becomes a very promising approach. This paper investigates the LTE network performance from the aspect of data analytics and statistical modeling. In this paper, we develop a methodology of data analytics and modeling to evaluate LTE network performance based on traffic measurements and service growth trends. A relational algorithm is developed to obtain the relationship between LTE network resources and LTE KPIs and a forecasting algorithm is developed to trend the network resource consumptions based on traffic and service growth. The numerical results show a high accuracy, robustness, and reliability of the proposed methodology.

Adaptive time division duplexing network with network coding embedded two-way relay

Using relays to receive and forward packets between a network base station and network users is a promising solution to address the cellular coverage issue. In this study, we consider time division duplexing (TDD) networks and explore the use of network coding for implementing the relays. Several transmission scenarios are investigated, including direction transmissions, one-way relay and two-way relay. In addition, we propose an adaptive TDD network, which enables the network to adaptively configure frame structures based on channel conditions to support direct transmissions, cooperative relay and two-hop relay (one-way and two-way). An opportunistic decoding scheme and a relay operation scheme are introduced to realise the proposed adaptive TDD network in a distributed manner. Both throughput and outage probability are derived. Performance evaluations show that the proposed adaptive TDD network based on network coding achieves the maximum throughput, as compared with one-way relay (without network coding), and extends the cell coverage range (compared with direct transmissions).

An Intelligent Device for Non-Restricted 3D Tracking and Drawing

Magnetic and optical methods for 3D movement detection have been studied in recent years. However, they all need environmental references to determine the targetpsilas spatial position and moving state. To breakthrough these limitations, a new non-restricted 3D measurement method based on accelerometer and the hardware design of this intelligent device are discussed in this paper. In order to improve the accuracy and efficiency of this system, we provide a method for data computation; a Kalman filter to reduce the noise when we process the accelerometer output; a new algorithm based on sensorpsilas non-linear character to compensate the acceleration after sampling; a second-order polynomial reconstruction algorithm to modify the velocity and displacement results. The device is assembled and tested in plenty of experiments in our laboratory. The results are showed in figures and special improvement is discussed in this paper.