Short-Term Forecasting of Emergency Inpatient Flow

Hospital managers have to manage resources effectively, while maintaining a high quality of care. For hospitals where admissions from the emergency department to the wards represent a large proportion of admissions, the ability to forecast these admissions and the resultant ward occupancy is especially useful for resource planning purposes. Since emergency admissions often compete with planned elective admissions, modeling emergency demand may result in improved elective planning as well. We compare several models for forecasting daily emergency inpatient admissions and occupancy. The models are applied to three years of daily data. By measuring their mean square error in a cross-validation framework, we find that emergency admissions are largely random, and hence, unpredictable, whereas emergency occupancy can be forecasted using a model combining regression and autoregressive integrated moving average (ARIMA) model, or a seasonal ARIMA model, for up to one week ahead. Faced with variable admissions and occupancy, hospitals must prepare a reserve capacity of beds and staff. Our approach allows estimation of the required reserve capacity.     

多尺度高斯模型的短期负荷预测

为了准确地进行电力负荷的短期预测,借鉴小波分析中对函数进行多尺度表示的思想,文中在高斯过程模型的基础上提出了多尺度高斯过程模型.通过选择合适的尺度参数,采用计算预测均方误差值大小的策略获取最佳延迟时间和最优嵌入维数对,然后对西北某地区电力系统进行短期负荷预测.与传统的支持向量机、径向基函数网负荷预测方法相比,基于多尺度高斯过程模型的短期负荷预测方法预测精度与支持向量机方法相当,性能优于径向基函数.     

一种短时突发PSK信号的自适应解调方法

短时突发相移键控（PSK）通信具有隐蔽性好、信道利用率高的特点,得到广泛应用。将已调信号非线性放大、限幅处理,使其幅度稳定在一定范围,基于锁相原理得到与已调信号同步的本地载波,是PSK信号解调相干载波恢复的常用方法。然而接收信号的非线性处理,会使PSK信号信噪比急剧下降,影响载波相位跟踪精度,给帧同步信号生成和码元抽取带来不确定性,导致误码率升高。针对常规方法存在的问题,提出了基于Hilbert变换的短时PSK信号自适应解调算法,通过信号归一化,在不降低解调信号信噪比情况下,确保载波恢复运算不受数据信号幅度的影响,具有恢复速度快、跟踪精度高、帧同步信号解调可靠、码元采样时间准确的特点。     

Techniques for Short-Term Predictions of Atmospheric Noise Levels

Short-term predictions of operational characteristics of ionospheric radio communication such as MUF, atmospheric noise, etc., are likely to lead to more efficient use of the HF spectrum. There are no techniques available for the short-term predictions of atmospheric noise so far. The paper describes two techniques for the purpose viz., autocorrelation technique and source location technique. The former predicts the noise level at a specified hour from the measurement at an earlier hour since atmospheric noise level variations constitute a Markov process. The latter predicts the noise level by predicting the type of the day from the characteristics observed at 16 hrs. Prediction accuracies are discussed, and the possibility of refinements is investigated. Possible system applications are discussed, and a typical calculation is carried out which shows that economies in the radiated power of the order of 10 dB are possible.     

WNN在短期电力负荷预测中的应用

通过对小波神经网络和BP神经网络的结构和算法进行理论分析,并对实际电力负荷预测算例进行对比研究,结论证明小波神经网络本身适合对波动性的信号进行预测,而且在神经网络节点数目相同的情况下,小波神经网络比BP神经网络具有更高的预测精度,因此采用小波神经网络有利于减少隐节点数目。     

中国电信发展的近期目标及主要措施

本文简述了我国电信发展的环境与条件，近期电信工作的任务和目标。最后指出实现电信发展目标的主要措施。     

一种面向短时突发信号的高精度同步方案设计

针对短时突发通信存在捕获时间长、算法复杂度高、资源开销大等问题,文中设计了一种高精度同步方案。该方案自定义收发双方帧结构,采用独特码差分匹配算法实现信号检测和帧同步,并采用基于导频累积量的前向估计算法实现频偏估计和校正。为兼顾频偏估计精度和算法复杂度,对独特码采用滑动相关算法实现初始相位捕获,使系统结构紧凑,更加适合FPGA硬件实现。理论分析和软件仿真表明,该方案在低信噪比和高动态范围条件下具有较好的接收性能,频率估计精度能够达到符号速率的万分之一。文中同时进行了硬件协同仿真与验证,并给出了相应的仿真波形。     

Short-Term Time Characteristics of Atmospheric Radio Noise Above Different Thresholds

Short-term time characteristics of burst and quasi- continuous forms of atmospheric radio noise were investigated at Bangalore at a number of frequencies in the range 100 kHz-9 MHz and over a 300-Hz-16-kHz range of bandwidth. It was found that for all frequencies, bandwidths, and threshold levels the duration and spacing follow log-normal distributions. The duration of a pulse above a particular threshold is statistically independent of its spacing from an adjacent pulse. Further, the average duration varies approximately linearly with the threshold level. These results are expected to be of use in planning digital communications where atmospheric noise is the limiting factor in reception capability.     

基于VMD和改进TCN的短期光伏发电功率预测

光伏发电功率存在波动性,且光伏出力易受各种气象特征影响,传统TCN网络容易过度强化空间特性而弱化个体特性。针对上述问题,文中提出一种基于VMD和改进TCN的短期光伏发电功率预测模型。通过VMD将原始光伏发电功率时间序列分解为若干不同频率的模态分量,将各个模态分量以及相对应的气象数据输入至改进TCN网络进行建模学习。利用中心频率法确定VMD的最优分解模态分解个数。在传统TCN预测模型的基础上,使用DropBlock正则化取代Dropout正则化以达到抑制卷积层中信息协同的效果,并引入注意力机制自主挖掘并突出关键气象输入特征的影响,量化各气象因素对光伏发电的影响,从而提高预测精度。以江苏省某光伏电站真实数据为例进行仿真实验,结果表明所提预测方法的RMSE为0.62 MW,MAPE为2.03%。     

基于灰色系统理论的雨衰减短期预报方法研究

雨衰减对10GHz以上频段的无线通信有很大的影响,在使用卫星通信系统时必须采取合理的策略以补偿降雨造成的衰减。为了给星地链路的自适应功率控制技术提供保障,首次将灰色系统理论应用在雨衰减短期预报中,分析给出了选择灰色系统模型和参数的原则。最后选取海口站Ka 频段地空链路的实测数据进行了验证,利用获得的雨衰减数据对未来间隔1s 和5s 的时刻进行预测,结果表明,基于灰色系统的雨衰减预测和实测数据吻合较好。     

考虑概率区间的微电网短期负荷多目标预测方法

微电网负荷随机性强、波动大,负荷单点预测已经难以满足微电网稳定运行需要.提出一种考虑概率区间的微电网短期负荷多目标预测方法,以循环神经网络为预测模型,以逼近理想解排序策略、网格筛选策略对基本多目标人工蜂群算法进行改进,优化循环神经网络的权值和阈值,避免单目标区间预测中惩罚系数难以选择的问题,对历史负荷数据进行记忆并修正预测结果,有效提高微电网短期负荷区间预测准确性与可靠性.仿真结果表明,本文所构建的考虑概率区间的微电网短期负荷多目标预测方法,预测性能优越、结果准确,可为微电网安全经济调度提供决策依据.     

Malicious Traffic classification Using Long Short-Term Memory (LSTM) Model

Malicious traffic classification is the initial and primary step for any network-based security systems. This traffic classification systems include behavior-based anomaly detection system and Intrusion Detection System. Existing methods always relies on the conventional techniques and process the data in the fixed sequence, which may leads to performance issues. Furthermore, conventional techniques require proper annotation to process the volumetric data. Relying on the data annotation for efficient traffic classification may leads to network loops and bandwidth issues within the network. To address the above-mentioned issues, this paper presents a novel solution based on artificial intelligence perspective. The key idea of this paper is to propose a novel malicious classification system using Long Short-Term Memory (LSTM) model. To validate the efficiency of the proposed model, an experimental setup along with experimental validation is carried out. From the experimental results, it is proven that the proposed model is better in terms of accuracy, throughput when compared to the state-of-the-art models. Further, the accuracy of the proposed model outperforms the existing state of the art models with increase in 5% and overall 99.5% in accuracy.

     

高动态短时突发扩频信号的快速捕获

针对低信噪比和大多普勒频移环境下,短帧突发直接序列扩频通信系统要求在短时间内以很低虚警和漏警概率捕获信号这一难题,设计了仅使用独特码的短时突发扩频信号帧结构,在接收端将扩频码与独特码作为一个复合扩频码看待,提出了基于FFT的高动态突发信号捕获方法,可同时快速完成突发信号检测、扩频码捕获、频偏估计、帧同步检测和位定时捕获等任务。仿真结果表明:该算法能有效消除多普勒频移对短时突发扩频信号捕获性能的影响。     

基于波束形成的长短时记忆网络语音分离算法研究

在利用深度学习方式进行语音分离的领域,常用卷积神经网络(RNN)循环神经网络进行语音分离,但是该网络模型在分离过程中存在梯度下降问题,分离结果不理想。针对该问题,该文利用长短时记忆网络(LSTM)进行信号分离探索,弥补了RNN网络的不足。多路人声信号分离较为复杂,现阶段所使用的分离方式多是基于频谱映射方式,没有有效利用语音信号空间信息。针对此问题,该文结合波束形成算法和LSTM网络提出了一种波束形成LSTM算法,在TIMIT语音库中随机选取3个说话人的声音文件,利用超指向波束形成算法得到3个不同方向上的波束,提取每一波束中频谱幅度特征,并构建神经网络预测掩蔽值,得到待分离语音信号频谱并重构时域信号,进而实现语音分离。该算法充分利用了语音信号空间特征和信号频域特征。通过实验验证了不同方向语音分离效果,在60°方向该算法与IBM-LSTM网络相比,客观语音质量评估(PESQ)提高了0.59,短时客观可懂(STOI)指标提高了0.06,信噪比(SNR)提高了1.13 dB,另外两个方向上,实验结果同样证明了该算法较IBM-LSTM算法和RNN算法具有更好的分离性能。     

Deep Graph-Long Short-Term Memory: A Deep Learning Based Approach for Text Classification

Wireless Personal Communications - Multi-label text classification is a challenging task in many real applications. Mostly, in all the traditional techniques, word2vec is used to show the...     

Implicit and Explicit Model-Based Signal Processing for the Analysis of Short-Term Cardiovascular Interactions

The paper stresses the importance of model-based signal processing in the analysis of the cardiovascular regulation mechanisms. It is remarked that even traditional signal processing implicitly assumes a model and interprets data according to it. Therefore, traditional signal processing is here referred to as implicit model-based signal processing in contrast with explicit model-based signal processing directly stemming from modeling considerations. The paper points out the advantages that can be achieved by rendering explicit the underlying model structure when it is implicit and by jointly applying implicit and explicit model-based signal processing approaches when dealing with complex, interacting,closed-loop biomedical systems. These advantages are rendered practical over several examples derived from the study of the short-term cardiovascular regulation performed by the autonomic nervous system.     

A Time-Domain Based Approach for Short-Term Fading Depth Evaluation in Wideband Mobile Communication Systems

In this paper a time-domain technique for wideband fading depth evaluation, is proposed, which is applicable to both Line-of-Sight and Non-Line-of-Sight cases, and the probability functions of the wideband received power are derived, contributing for filling in the gap of short-term fading characterisation in wideband systems. Application examples for different systems, namely GSM, UMTS and HIPERLAN/2, working in different standard reference environments, are shown. As expected, it is observed that Rayleigh and Rice distributions are appropriate for evaluating the fading margins for GSM in some environments; nevertheless, for UMTS and HIPERLAN/2, fading margins are usually well below the ones obtained from considering these narrowband distributions, the differences reaching up to around 10 and 13 dB, respectively. A simple relationship between the physical and the geometrical environment properties, and the rms delay spread of the propagation channel is also proposed, establishing a relationship between two models independently derived by different authors. Using the proposed relationship, fading depth results from the proposed time-domain technique are compared with the ones for an environment-geometry based one, and a good agreement is verified. The difference in fading depth between both approaches is roughly below 2 dB. Filipe D. Cardoso received the Licenciado, M.Sc. and Ph.D. degrees in Electrical and Computer Engineering from IST/Technical University of Lisbon in 1991, 1996 and 2004, respectively. Since 1994 he has been with the Department of Electrical Engineering of EST/Polytechnic Institute of Setúbal, Setúbal, Portugal, where he is an Adjunct Professor in the Telecommunications area. He is also a researcher at the Telecommunications Institute, Lisbon. He has been a member of COST 273 and IST/FLOWS European projects and is involved in the Network of Excellence in Wireless Communications (NEWCOM). He has authored papers in national and international conferences and journals, for which he has also served as a reviewer. His research interests include wireless/mobile channel characterisation and modelling and future Mobile Broadband Systems. Luis M. Correia received the Ph.D. in electrical and computer engineering from IST/Technical University of Lisbon in 1991, where he is currently a professor in telecommunications, with a focus on wireless/mobile communications in the areas of channel characterisation, traffic, services, and cellular planning. He as acted as a consultant for Portuguese GSM operators and for the telecommunications regulator. Besides being responsible for research projects at the national level, he was or is part of various ones within European frameworks (RACE, ACTS, and IST). He participated in and was co-editor of the Final Report for COST 231, Chairman and editor of the Final Report for COST 259, and is currently Chairman of COST 273. He was and is responsible for the supervision of students at both the M.Sc. and Ph.D. levels. He has authored papers and communications in international journals and conferences, for which he has also served as a reviewer, has been a co-editor of a special issue of a journal on wireless broadband systems, and is currently the editor for Wireless Systems in the European Transactions on Telecommunications. He as served as evaluator and auditor in ACTS, ESPRIT and IST frameworks. He was Chairman of the Technical Program Committee of PIMRC 2002.     

高动态短时突发扩频信号的快速捕获技术研究

针对传统伪码捕获结构在处理高动态突发信号时存在的局限性,提出了一种基于载波信号功率谱特性的快速捕获电路结构,通过计算解扩信号功率谱最大值与次大值的比值,并将该比值与判决门限进行比较,从而判断捕获成功与否.采用Welch算法改进信号功率谱的方差特性,计算机仿真结果验证了该算法能够完成高动态环境下短时突发扩频信号伪码的快速捕获.     

雷暴云团的识别追踪技术在短时临近预报中的应用

应用K均值聚类算法识别雷暴云团,对其进行追踪,通过线性外推的方法进行预报,并采用雷达反射率因子Z和降水强度R的统计关系式Z=ARb,对雷达实测的Z值与相应地面雨量计实测的雨强数据进行比较,经过回归统计分析,分别得到不同降水量级下的A、b系数,获得相应降水估算方程。选取2007年9月18日降水过程的南京雷达资料,对这种雷暴云团的识别追踪方法进行了详细说明,并对预报结果进行了验证。结果表明：该方法能够较好地识别追踪和外推预报雷暴云团,对短时降水能做出比较准确的估算。     

Device-Free Through-the-Wall Activity Recognition Using Bi-Directional Long Short-Term Memory and WiFi Channel State Information

Activity recognition plays a key role in health management and security. Traditional approaches are based on vision or wearables, which only work under the line of sight (LOS) or require the targets to carry dedicated devices. As human bodies and their movements have influences on WiFi propagation, this paper proposes the recognition of human activities by analyzing the channel state information (CSI) from the WiFi physical layer. The method requires only the commodity: WiFi transmitters and receivers that can operate through a wall, under LOS and non-line of sight (NLOS), while the targets are not required to carry dedicated devices. After collecting CSI, the discrete wavelet transform is applied to reduce the noise, followed by outlier detection based on the local outlier factor to extract the activity segment. Activity recognition is fulfilled by using the bi-directional long short-term memory that takes the sequential features into consideration. Experiments in through-the-wall environments achieve recognition accuracy >95% for six common activities, such as standing up, squatting down, walking, running, jumping, and falling, outperforming existing work in this field.