Hopfield neural‐network‐based dynamic resource allocation scheme for non‐real‐time traffic in wireless networks

Dynamic resource allocation (DRA) plays a fundamental role in current and future wireless networks, including 3G systems. In this paper, a scheduling DRA scheme for non‐real‐time (NRT) packet services in wireless system is proposed based on the use of Hopfield neural networks (HNN). The scheme exploits the fast response time of HNN for solving NP optimization problems and has been particularized for the downlink transmission in a UMTS system, although it could be easily extended to any other radio access technology. The new DRA scheme follows a delay‐centric approach, since it maximizes the overall system resource utilization while minimizing the packet delay. Simulation results confirm that the proposed HNN‐based DRA scheme is effective in supporting different types of NRT services, while achieving efficient utilization of radio resources. Copyright © 2008 John Wiley & Sons, Ltd.     

User-centric energy efficiency fairness in backscatter-assisted wireless powered communication network

In order to address the unfair user-centric energy efficiency (EE) problem caused by channel difference in the backscatter-assisted wireless powered communication network,a resource allocation scheme was proposed.Firstly,a mixed integer nonconvex fractional programming problem was formulated to maximize the minimum user-centric EE,subject to the quality of service and energy-causality constraints.Based on the generalized fractional programming theory,the original problem was transformed into a mixed integer nonconvex subtraction problem.With the aid of the slack variable,the proof by contradiction,the auxiliary variable and the mixed integer nonconvex subtraction problem were further transformed into an equivalent convex problem.Finally,an iterative algorithm was proposed to obtain the optimal solutions.Computer simulations validated the quick convergence of the proposed iterative algorithm,and that the developed resource allocation scheme efficiently guarantees the fairness among users in terms of EE.     

User QoE Influenced Spectrum Trade,Resource Allocation,and Network Selection

In future wireless networks, we envision more dynamic telecommunication paradigm, where the dynamics may be translated into dynamic service offerings and user profiles etc. We further expect that the wireless communication market will be influenced when the user-centric network selection vision is realized. By the user-centric network selection vision, we mean that users will be free to select any available network operator or service provider on short term contractual basis. This dictates that operators will compete for their share of a common user pool on much smaller time quanta when compared with the current long term user contacts with the operators. One intuitive strategy of operators will be to incentivize users by offering different QoS and service price offers. As the operators’ offers are influenced by their incurring costs. This necessitates to study the market behavior at different levels and investigate the operator and user behavior at these level. In this paper, we categorize and position the communication players and model the interaction between players at different levels. We introduce the learning aspects in the interaction and investigate the equilibrium strategies of involved stake-holders i.e., users and operators. We also model the utility functions of all the involved stake-holders. We also examine the risk-sensitive utility functions in order to cover both risk-seeking and risk-averse in the user QoEs. We implement the user-centric approach and compare it against our proposed network-centric resource utilization and call blocking.     

Pricing and power control for joint network-centric and user-centric radio resource management

Objectives of most radio resource-management schemes can be classified as either user centric or network centric. User-centric schemes try to maximize the interests of individual users, while network-centric schemes optimize collective metrics for all users. These two types of resource management tend to result in qualitatively different resource allocations (with, sometimes, very different degrees of fairness). In this paper, we consider the joint optimization of both user-centric and network-centric metrics. Specifically, we use a utility function (measured in units of bits per Joule) as the user-centric metric, and for the network-centric counterpart, we consider a function of the sum of the throughputs of users in the network. The user-centric measure reflects the individual user's throughput, as well as the battery energy (transmit power) consumed to achieve it. The network-centric measure reflects the total revenue derived by the usage of network resources. We introduce an explicit pricing mechanism to mediate between the user-centric and network-centric resource-management problems. Users adjust their power in a distributed fashion to maximize the difference between their utilities and their payments (measured as a product of the unit price and throughput). The network adjusts the unit price in order to maximize its revenue (measured as the sum of the individual payments). We show that the distributed user-centric power control results in a unique Nash equilibrium. Our numerical results indicate that there exists a unique unit price that maximizes the revenue of the network. We also derive a semianalytical, computationally simple, and highly accurate approximation to the optimal solution. Our results show that while users with better channels receive better qualities of service, as usual (e.g., as in waterfilling), they also make proportionally higher contributions to the network revenue.     

Flexible design of cognitive radio wireless systems

The goal of this article is to show how many challenging unsolved resource allocation problems in the emerging field of cognitive radio (CR) networks fit naturally either in the game theoretical paradigm or in the more general theory of VI. This provides us with all the mathematical tools necessary to analyze the proposed equilibrium problems for CR systems (e.g., existence and uniqueness of the solution) and to devise distributed algorithms along with their convergence properties.     

Kalman Filter-Trained Recurrent Neural Equalizers for Time-Varying Channels

Recurrent neural networks have been successfully applied to communications channel equalization because of their modeling capability for nonlinear dynamic systems. Major problems of gradient-descent learning techniques commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. This paper presents decision-feedback equalizers using a recurrent neural network trained with Kalman-filtering algorithms. The main features of the proposed recurrent neural equalizers, using the extended Kalman filter and the unscented Kalman filter, are fast convergence and good performance using relatively short training symbols. Experimental results for various time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.     

Diversity reservation ALOHA

A multiaccess protocol suitable for packet satellite broadcast channels is introduced in this paper. The proposed protocol, called diversity reservation ALOHA (DRA) combines the essential features of announced retransmission random access (ARRA). diversity ALOHA, and a reservation scheme with a slotted ALOHA reservation channel to achieve the goals of improved delay-throughput characteristics and high channel utilization. Two versions of DRA protocols are analyzed. In scheme 1, called DRA with reservation cleared (DRA-RC), no global queue information is required for a user to access the channel. In this version, even a successful transmission of reservation requests will be cleared from the system if it causes overflow. The second scheme, called DRA with reservation delayed (DRA-RD), establishes a global queue distributed at users' terminals, so that users with successful transmission of reservation requests are assured of access rights although transmission may be delayed to later fames. Important system performance measures, such as activity-factor-throughput characteristics, for DRA-RC and DRA-RD are compared with those of ARRA. Examples of numerical results show that with system parameters properly chosen, substantial increases in attainable channel utilization are possible. Furthermore, the use of diversity for reservation allows a smaller packet delay especially under light traffic conditions. The analysis is validated in light of the simulation results.     

Organizing reserve channels for superior admission control performance

Good execution of seamless handoffs is key to the quality of service perceived by mobile service subscribers. Giving priority to handoff requests is a strategy commonly considered to guarantee low connection loss when performing handoffs. Often, priority systems can be implemented by adopting a pool of reserved (or guard) resources that is available only for calls or transactions being handed over. Most studies published thus far consider that fixed allocation of the guard resources is being employed. This work considers sharing the reserve channels among cells through the use of a particular method of dynamic resource allocation (DRA). It is shown here that the use of DRA on the pool of guard channels affords a great reduction in the probability of handoff failure without affecting the value of the new call blocking probability, thereby improving the quality of service provided by the cellular operator and also increasing system utilization.     

Multiresolution learning paradigm and signal prediction

Current neural network learning processes, regardless of the learning algorithm and preprocessing used, are sometimes inadequate for difficult problems. We present a new learning concept and paradigm for neural networks, called multiresolution learning, based on multiresolution analysis in wavelet theory. The multiresolution learning paradigm can significantly improve the generalization performance of neural networks     

Dynamic Resource Allocation for Beyond 3G Cellular Networks

The Multicarrier CDMA Transmission Techniques for Integrated Broadband Cellular Systems (MATRICE) project addresses a candidate solution for a Beyond 3G (B3G) air-interface based on Multi-Carrier Code Division Multiple Access (MC-CDMA). It investigates dynamic resource allocation strategies at the Medium Access Control (MAC) layer to support the transport of Internet Protocol (IP) packets over the air-interface in a cost effective manner and maximise the cell capacity with a target QoS. A candidate Dynamic Resource Allocation (DRA) protocol architecture is proposed that is based on cross-layer signalling to provide reactive resource allocation according to the fast channel and traffic variations. In-line with B3G expectations, the proposed DRA handles a very large number of users with inherent flexibility and granularity necessary to support heterogeneous traffic, and still with limited complexity. Thanks to the modular architecture of the DRA, various scheduling policies are investigated and compared in terms of capacity and reactivity to the system environment. Simulation results have shown that the MATRICE system has the potential to deliver broadband heterogenous services in a cost-effective manner, and emerge as a propespective candidate air-interface for B3G cellular networks.     

基于膜量子布谷鸟搜索的双通道网络频谱资源分配

无线双通道Ad Hoc网络中, 有效分配簇间码分频谱资源是提高资源利用效率的关键技术之一.综合考虑子簇码分频谱资源需求和分配公平性, 给出了簇间码分频谱资源分配数学模型, 并转换为以最大化码分频谱资源效益和分配公平性为多目标的受约束离散优化问题.结合膜结构、量子计算和布谷鸟搜索算法, 提出一种新的离散组合优化算法——膜量子布谷鸟搜索算法.该算法使用量子鸟窝表征问题潜在解, 利用布谷鸟寻窝产卵的演化方法在基础膜中寻求单目标最优解, 通过膜间信息共享和非支配解等级排序求出具有多目标最优解的表层膜Pareto前端解集.仿真结果证明, 与经典优化算法相比, 该算法不仅能够同时求解单目标和多目标最优解, 而且具有更优的收敛性能, 能更好地实现码分频谱资源效益最优化.     

Channel allocation for GPRS

Based on the GSM radio architecture, the general packet radio service (GPRS) provides users data connections with variable data rates and high bandwidth efficiency. In the GPRS service, allocation of physical channels is flexible, i.e., multiple channels can be allocated to a user. We propose four algorithms for the GPRS radio resource allocation: fixed resource allocation (FRA), dynamic resource allocation (DRA), fixed resource allocation with queue capability (FRAQ), and dynamic resource allocation with queue capability (DRAQ). We develop analytic and simulation models to evaluate the performance for these resource allocation algorithms in terms of the acceptance rate of both GPRS packet data and GSM voice calls. Our study indicates that DRAQ (queuing for both new and handoff calls) outperforms other algorithms     

A Neural Network-Based Connectivity Management for Mobile Computing Environment

The mobile computing environment experiences wireless problems and suffers from limited bandwidth, which leads to frequent disconnections. This has posed a challenge in maintaining user-to-user connectivity in the mobile computing environment. In this paper, we propose a neural network (NN) based connectivity management for mobile computing environment to maintain the mobile user-to-user connectivity throughout the transaction. Here the connectivity management maintains the status information of mobile hosts at the base station to handle frequent disconnection of mobile hosts (MHs), which occur because of hand-offs and interruptions. The disconnection of an MH because of wireless problems is called interruption, and the disconnection due to MH crossing the cell boundary is called hand-off. The neural networks are trained with respect to the status information to provide an intelligent decision for the connectivity management. The simulation results demonstrate that the proposed technique performs well in terms of percentage acceptance of disconnections and resource utilization (bandwidth and buffer) for the volatile mobile computing environment. It is also observed that the intelligent decision by neural network has improved the performance of the system.     

Direction of arrival estimation based on phase differences usingneural fuzzy network

A new high-resolution direction of arrival (DOA) estimation technique using a neural fuzzy network based on phase difference (PD) is proposed. The conventional DOA estimation method such as MUSIC and MLE, are computationally intensive and difficult to implement in real time. To attack these problems, neural networks have become popular for DOA estimation. However, the normal neural networks such as the multilayer perceptron (MLP) and radial basis function network (RBFN) usually produce the extra problems of low convergence speed and/or large network size (i.e., the number of network parameters is large). Also, the may to decide the network structure is heuristic. To overcome these defects and take use of neural learning ability, a powerful self-constructing neural fuzzy inference network (SONFIN) is used to develop a new DOA estimation algorithm. By feeding the PDs of the received radar-array signals, the trained SONFIN can give high-resolution DOA estimation. The proposed scheme is thus called PD-SONFIN. This new algorithm avoids the need of empirically determining the network size and parameters in normal neural networks due to the powerful on-line structure and parameter learning ability of SONFIN. The PD-SONFIN can always find itself an economical network size in the fast learning process. Our simulation results show that the performance of the new algorithm is superior to the RBFN in terms of convergence accuracy, estimation accuracy, sensitivity to noise, and network size     

Quantum Particle Swarm Optimization for Electromagnetics

A new particle swarm optimization (PSO) technique for electromagnetic applications is proposed. The method is based on quantum mechanics rather than the Newtonian rules assumed in all previous versions of PSO, which we refer to as classical PSO. A general procedure is suggested to derive many different versions of the quantum PSO algorithm (QPSO). The QPSO is applied first to linear array antenna synthesis, which is one of the standard problems used by antenna engineers. The performance of the QPSO is compared against an improved version of the classical PSO. The new algorithm outperforms the classical one most of the time in convergence speed and achieves better levels for the cost function. As another application, the algorithm is used to find a set of infinitesimal dipoles that produces the same near and far fields of a circular dielectric resonator antenna (DRA). In addition, the QPSO method is employed to find an equivalent circuit model for the DRA that can be used to predict some interesting parameters like the Q-factor. The QPSO contains only one control parameter that can be tuned easily by trial and error or by suggested simple linear variation. Based on our understanding of the physical background of the method, various explanations of the theoretical aspects of the algorithm are presented.     

Dynamic resource allocation in adaptive wireless OFDMA systems

Modern wireless orthogonal frequency division multiple access (OFDMA) systems incorporate dynamic resource allocation (DRA), adaptive modulation and coding (AMC), and power control (PC) to exploit multiuser diversity and achieve higher system throughput. In the literature, only a few proposed algorithms deal with the contiguous DRA problem according to which a contiguous collection of resources can be allocated to each user. This paper formulates this high complexity problem, provides a suitable decision metric and a simple yet efficient solution. The proposed algorithm allocates in each step a contiguous collection of resources to the pending user that leads to the highest estimated correctly received number of bits. Simulation results show that, in this way, considerably improved performance can be achieved in terms of overall system throughput, spectral efficiency, and served traffic. Copyright © 2010 John Wiley & Sons, Ltd.     

Intelligent feedback control-based adaptive resource management forasynchronous, decentralized real-time systems

Presents intelligent feedback control techniques for adaptive resource management in asynchronous, decentralized real-time systems. We propose adaptive resource management techniques that are based on feedback control theory and are designed using the intelligent control design paradigm. The controllers solve resource allocation problems that arise during run-time adaptation using the classic proportional-integral-derivative (PID) control functions and fuzzy logic. We study the performance of the controllers through simulation. The simulation results indicate that the controllers produce low missed deadline ratios and resource utilizations during high-workload situations     

以用户为中心的可见光通信协作传输中训练资源分配算法

以用户为中心的可见光通信协作传输是近年来出现的新架构,这导致虚拟小区之间出现重叠。为避免导频污染问题,每个虚拟小区中的光接入点(AP)或者虚拟小区中选择相同AP的用户发送的训练序列应该是正交的。针对可见光通信中以用户为中心的协作网络,研究训练资源的正交分配问题,提出了一种新的导频分配算法,联合导频分配和用户选择问题,以期最大限度地增加虚拟小区内可被接入的用户数。分析和仿真结果表明,该导频分配方案可以有效改善导频污染问题,提高训练资源利用率,并且相比已有的导频分配方案,性能有所改进。     

Generative Neural Network Based Spectrum Sharing Using Linear Sum Assignment Problems

Spectrum management and resource allocation(RA)problems are challenging and critical in a vast number of research areas such as wireless communications and computer networks.The traditional approaches for solving such problems usually consume time and memory,especially for large-size problems.Recently different machine learning approaches have been considered as potential promising techniques for combinatorial optimization problems,especially the generative model of the deep neural networks.In this work,we propose a resource allocation deep autoencoder network,as one of the promising generative models,for enabling spectrum sharing in underlay device-to-device(D2D)communication by solving linear sum assignment problems(LSAPs).Specifically,we investigate the performance of three different architectures for the conditional variational autoencoders(CVAE).The three proposed architecture are the convolutional neural network(CVAECNN)autoencoder,the feed-forward neural network(CVAE-FNN)autoencoder,and the hybrid(H-CVAE)autoencoder.The simulation results show that the proposed approach could be used as a replacement of the conventional RA techniques,such as the Hungarian algorithm,due to its ability to find solutions of LASPs of different sizes with high accuracy and very fast execution time.Moreover,the simulation results reveal that the accuracy of the proposed hybrid autoencoder architecture outperforms the other proposed architectures and the state-of-the-art DNN techniques.     

异构无线网络基于业务特性的动态无线资源管理（英文）

The traffic with tidal phenomenon in Heterogeneous Wireless Networks(HWNs)has radically increased the complexity of radio resource management and its performance analysis.In this paper,a Simplified Dynamic Hierarchy Resource Management(SDHRM)algorithm exploiting the resources dynamically and intelligently is proposed with the consideration of tidal traffic.In network-level resource allocation,the proposed algorithm first adopts wavelet neural network to forecast the traffic of each sub-area and then allocates the resources to those sub-areas to maximise the network utility.In connection-level network selection,based on the above resource allocation and the pre-defined QoS requirement,three typical network selection policies are provided to assign traffic flow to the most appropriate network.Furthermore,based on multidimensional Markov model,we analyse the performance of SDHRM in HWNs with heavy tailed traffic.Numerical results show that our theoretical values coincide with the simulation results and the SDHRM can improve the resource utilization.