智能优化在多用户检测中的应用

为进一步提高离散混合蛙跳算法（DSFLA）的性能,将免疫算法和克隆选择理论分别与DSFLA相结合,提出了免疫蛙跳算法（IDSFLA）和克隆蛙跳算法（KDSFLA）,利用这两种智能算法得到两种新的多用户检测器。IDSFLA是在DSFLA的每一族内更新中,嵌入免疫算法,利用Hopfield神经网络（HNN）快速产生最优个体作为疫苗母本,提高算法的全局收敛能力;KDSFLA在族内更新中,利用克隆算法的消亡操作,淘汰适应度低的青蛙个体,保证最优个体的有效进化。仿真结果表明,所提出的两种多用户检测器,在误码率、收敛速度、系统容量、抗远近能力等方面都有显著改善。     

一种改进的遗传量子算法及其应用

针对遗传量子算法(0QA)在优化连续多蜂函数时易出现早熟现象，本文提出一种改进的遗传量子算法(IGQA)，其核心是在量子门更新过程进行改进的基础上，引入群体灾变和自适应搜索网格的策略。通过典型函数测试和FIR数字滤波器设计实例表明，IGQA的性能优于GQA和其它几种遗传算法，具有比GQA更快的收敛速度和更好的全局寻优能力，能有效地克服早熟现象。     

基于神经网络离散混合蛙跳算法的多用户检测

为进一步提高基于离散混合蛙跳算法（DSFLA）的多用户检测性能,提出一种基于DSFLA和神经网络相结合的神经网络离散混合蛙跳算法,并用于多用户检测。在DSFLA的每一族内更新中,随机选择若干只“青蛙”采用Hopfield神经网络的寻优更新策略,进行快速迭代,寻找全局最优。仿真结果证明,基于神经网络离散混合蛙跳算法的多用户检测器在误码率、收敛速度、系统容量、抗远近能力等方面都优于传统方法和一些应用优化算法的多用户检测器。     

MC-CDMA系统的神经鱼群算法检测器

为使人工鱼群算法在最短的时间内取得多用户检测问题的最优解,在MC-CDMA系统基础上设计一种神经网络人工鱼群算法。人工鱼的3种神经网络行为使神经网络人工鱼群算法在解决多用户检测该类组合优化问题时,减少搜索的随机性和任意性,加快原鱼群算法的收敛速度。仿真结果证明,该算法能够快速收敛,且其抗多址干扰能力和抗远近效应能力优于已有应用智能算法的多用户检测器。     

一种改进的暂态混沌神经网络多用户检测算法

分析了暂态混沌神经网络中的模拟退火函数和自反馈连接权值的敏感性,提出了一种基于模拟退火优化的自适应暂态混沌神经网络,具有较好的逃逸局部最优点的能力,并将其应用于DS/CDMA的多用户检测技术。仿真结果表明,基于模拟退火优化的自适应暂态混沌神经网络多用户检测算法,其误码率性能以及抗远近效应能力优于已有的神经网络多用户检测算法,并具有较好的信干比。     

抗时延敏感性跨层自适应资源分配方案*

为了对抗多用户OFDM系统中用户实时业务对时延的敏感性,提出一种利用Hopfield神经网络(HNN)算法的跨层自适应资源分配方案。该方案设置用户调度优先级时同时考虑物理层的信道状态信息,及媒体接入层的用户队列状态信息和等待时间等;采用HNN算法,最大化系统容量的同时降低了平均时延和丢包率。仿真结果表明,相比于传统资源分配方案,该方案可以有效保障用户的服务质量,并提高了系统的整体性能。     

基于免疫粒子群算法的多用户检测技术研究

将免疫系统的免疫机制引入到粒子群优化算法的设计中,模拟免疫系统、群集智能和神经网络的信息处理机制,提出了免疫粒子群优化算法。这种免疫粒子群算法结合了粒子群的近似全局优化能力和由Hopfield神经网络构成的免疫系统的快速信息处理机制,加快了算法的收敛速度,并提高了粒子群算法的全局收敛能力。然后利用此算法对CDMA系统的多用户检测性能改进问题进行实验研究,证明了本文的方法有较快的收敛速度,并且无论是抗多址干扰能力还是抗远近效应能力都优于传统方法和一些应用优化算法的多用户检测器。     

基于RBF神经网络混合遗传算法的多用户检测

提出了一种混合递阶遗传算法来同时训练RBF神经网络的结构和参数,引入了改进的染色体编码方案,用基于奇异值分解的最小二乘法计算网络输出层权值,提高了遗传搜索的效率,精简了网络结构.并用变学习速率梯度下降法优化遗传训练出的最优网络,应用到多用户检测中.仿真结果表明,新混合学习算法训练出的网络结构优于其它算法训练的网络结构,并且性能良好.     

基于免疫进化规划的多用户检测技术研究

把人工免疫系统和神经网络系统的信息处理机制引入到进化规划算法（EP），提出了免疫进化规划算法．所提IEP通过使用随机Hopfield神经网络制备疫苗构成新的免疫算子，把新的免疫算子结合到进化规划中，不仅加快了进化规划的收敛速度，并提高了进化规划的全局收敛能力．然后在CDMA系统利用此算法设计了新的多用户检测器．仿真结果证明了该方法能够快速收敛到全局最优解，并且无论抗多址干扰和抗远近效应能力都优于传统方法和一些应用优化算法的多用户检测器．     

一种基于隐训练序列的DS-UWB多用户检测算法研究

多用户检测技术目的是在传统检测技术的基础上,充分利用造成多址干扰的所有用户信号信息对多用户做联合检测,以有效地消除多址干扰和远近效应问题。基于隐训练序列的最小均方误差（MMSE）多用户检测算法是在基于完全训练序列的MMSE多用户检测算法基础上,通过对算法进行改进,将训练序列嵌入到用户信息序列中发送。该检测算法提高了频谱资源的利用率,仿真结果表明其性能接近于完全训练序列的MMSE多用户检测算法,并且算法复杂度较低,适于工程应用。     

Hysteretic HNN MAI-Cancellation for CDMA Signals in Multipath Fading Channels

Multiple-access interference cancellation using hysteretic Hopfield neural network receiver for direct sequence code-division multiple access (DS-CDMA) in multipath fading channels is investigated. It has been shown that by applying the phenomenon of “hysteresis” to the Hopfield neural network (HNN) detector, performance of this detector may be enhanced in all near-far situations for different number of multipath rays. Introducing the concept of Hysteresis into HNN has made this suboptimum CDMA detector even closer to the optimum multiuser CDMA detector. As shown by simulation results, the bit-error rate performance achieved by the Hysteretic Hopfield Neural Network detector outperforms the classical HNN detector with a good margin and is promising.     

Mutation Hopfield neural network and its applications

In this paper, a new operator is proposed to optimize the traditional Hopfield neural network (HNN). The key idea is to incorporate the global search capability of the Estimation of Distribution Algorithms (EDAs) into the HNN, which typically has a powerful local search capability and fast operation. On account of this property of the EDA, our proposed algorithm also exhibits a powerful global search capability. In addition, the possible infeasible solutions generated during the re-sampling period of the EDA are eliminated by the HNN. Therefore, the merits of both these methods are combined in a unified framework. The proposed model is tested on a numerical example, the max-cut problem. The new and optimized model yielded a better performance than certain traditional intelligent optimization methods, such as HNN, genetic algorithm (GA). The proposed mutation Hopfield neural network (MHNN) is also used to solve a practical problem, aircraft landing scheduling (ALS). Compared with first-come-first-served sequence, MHNN sequence reduces both total landing time and total delay.     

Fusion of artificial neural networks for learning capability enhancement: Application to medical image classification

Artificial neural network (ANN) is one of the commonly used tools for computational applications. The specific advantages of ANN are high accuracy, less convergence time, less computational complexity, and so forth. However, all these merits are not available in the same ANN. Even though back propagation neural (BPN) networks are accurate, their computational complexity is significantly high. BPN networks are also not stable. On the other hand, Hopfield neural network (HNN) is better than BPN in terms of computational efficiency. But the accuracy of HNN is low. In this work, a modified ANN is proposed to overcome this specific problem. The modified ANN is a fusion of BPN and HNN. The technical concepts of BPN and HNN are mixed in the training algorithm of the proposed back propagation‐Hopfield network (BPHN). The objective of this fusion is to improve the performance of conventional ANN. Magnetic resonance brain image classification experiments are used to analyse the proposed BPHN. Experimental results have suggested improvement in the learning process of the proposed BPHN. A comparative analysis with the conventional networks is performed to validate the performance of the proposed approach.     

Hopfield neural network implementation of the optimal CDMAmultiuser detector

We investigate the application of Hopfield neural networks (HNN's) to the problem of multiuser detection in spread spectrum/CDMA (code division multiple access) communication systems. It is shown that the NP-complete problem of minimizing the objective function of the optimal multiuser detector (OMD) can be translated into minimizing an HNN “energy” function, thus allowing to take advantage of the ability of HNN's to perform very fast gradient descent algorithms in analog hardware and produce in real-time suboptimal solutions to hard combinatorial optimization problems. The performance of the proposed HNN receiver is evaluated via computer simulations and compared to that of other suboptimal schemes as well as to that of the OMD for both the synchronous and the asynchronous CDMA transmission cases. It is shown that the HNN detector exhibits a number of attractive properties and that it provides a powerful generalization of a well-known and extensively studied suboptimal scheme, namely the multistage detector     

一种基于免疫和Hopfield神经网络的多峰值优化算法

分析了免疫算法和Hopfield神经网络的优缺点，提出了一种解决多峰值函数优化问题的混合算法。Hopfield神经网络易于硬件实现，具有简单、快速的优点，但是对初始值具有依赖性以及容易陷入局部极值。免疫算法具有识别多样性的特点，但搜索效率和精度不高。将两算法结合起来，优势互补。首先用免疫算法寻优，然后对所得具有全局多样性的解进行聚类分析，所得聚类中心作为Hopfield神经网络的初始搜索点，最后利用Hopfield神经网络逐个寻优。实验表明，该算法是一种有效的求解多峰函数优化问题的方法，与免疫算法相比，搜索效率和精度都较高。     

Integrating a scale-invariant feature of fractal geometry into the Hopfield neural network for super-resolution mapping

ABSTRACT

Super-resolution mapping (SRM) is a potential technique to improve image pattern recognition by predicting the spatial distribution of class composition at a sub-pixel scale. A number of SRM techniques have been reported in the past two decades. Most of the techniques are based on the assumption of spatial dependence. In this paper, a scale-invariant concept of fractal geometry is taking into account in the original Hopfield neural network (HNN) algorithm and a self-similar Hopfield neural network (SSHNN) is proposed which based on both spatial dependence and self-similarity in the fractal geometry. Both synthetic and real satellite images are used to test the performance of the SSHNN. The results show that by taking self-similarity into consideration, with a single image and no other additional data needed, the mapping accuracy of the SSHNN increases by up to 20% compared to the HNN.     

基于ATM交换结构的Hopfield神经网络调度算法

针对ATM交换结构，采用输入缓冲和每条入线在同一个时隙内可传送多于一个信元的策略，利用神经网络具有的实时性、高度并行处理能力和易于电路或光电技术实现等特点，提出了一种Hopfield神经网络调度算法。实验仿真比较表明，该方法不但大大提高了吞吐率，消除了队头阻塞造成的性能恶化，而且降低了信元丢失率和较大程度地降低了平均信元时延，提高了ATM交换结构的性能，实现了信元的优化调度。     

基于混沌粒子群算法和Hopfield网络求解TSP问题

针对Hopfield网络求解TSP问题经常出现局部最优解,该文将混沌粒子群算法(PSO)与之结合,提出一种基于混沌粒子群的Hopfield神经网络方法。通过实验将其与文献[5,8]以及"PSO+HNN"策略比较,验证了该文算法不仅能够以更大概率收敛到全局最优,而且耗时更少。     

基于混沌粒子群算法和Hopfield网络求解TSP问题

针对Hopfield网络求解TSP问题经常出现局部最优解,该文将混沌粒子群算法（PSO）与之结合,提出一种基于混沌粒子群的Hopfield神经网络方法。通过实验将其与文献[5,8]以及“PSO＋HNN”策略比较,验证了该文算法不仅能够以更大概率收敛到全局最优,而且耗时更少。     

基于Hopfield神经网络和禁忌搜索的软/硬件功耗划分

在目前全球倡导“低碳经济”的背景下,随着嵌入式系统大量而广泛的使用,嵌入式软件功耗已成为嵌入式系统设计的一个关键因素,而软/硬件划分是嵌入式软件功耗优化的一种重要方法。首先在性能约束条件下,建立以嵌入式软件功耗为目标的软/硬件双路划分模型;然后,提出了一种基于离散Hopfield神经网络(HNN)和禁忌搜索(TS)融合的求解算法,采用离散Hopfield算法作为主算法能较快地获得可行解,使用禁忌搜索算法“禁忌”当前解而转移到目标函数的其他极小点,从而可跳出局部最优解而快速趋于全局最优解;最后,仿真实验表明,与同类算法相比,该算法不但具有搜索速度上的优势,而且求得全局最优解的概率更高。