Resource Allocation for Cognitive Network Slicing in PD-SCMA System Based on Two-Way Deep Reinforcement Learning

In this paper,we propose the Two-way Deep Reinforcement Learning(DRL)-Based resource allocation algorithm,which solves the problem of resource allocation in the cognitive downlink network based on the underlay mode.Secondary users(SUs)in the cognitive network are multiplexed by a new Power Domain Sparse Code Multiple Access(PD-SCMA)scheme,and the physical resources of the cognitive base station are virtualized into two types of slices:enhanced mobile broadband(eMBB)slice and ultrareliable low latency communication(URLLC)slice.We design the Double Deep Q Network(DDQN)network output the optimal codebook assignment scheme and simultaneously use the Deep Deterministic Policy Gradient(DDPG)network output the optimal power allocation scheme.The objective is to jointly optimize the spectral efficiency of the system and the Quality of Service(QoS)of SUs.Simulation results show that the proposed algorithm outperforms the CNDDQN algorithm and modified JEERA algorithm in terms of spectral efficiency and QoS satisfaction.Additionally,compared with the Power Domain Non-orthogonal Multiple Access(PD-NOMA)slices and the Sparse Code Multiple Access(SCMA)slices,the PD-SCMA slices can dramatically enhance spectral efficiency and increase the number of accessible users.

Deep learning algorithm featuring continuous learning for modulation classifications in wireless networks

Although modulation classification based on deep neural network can achieve high Modulation Classification(MC)accuracies,catastrophic forgetting will occur when the neural network model continues to learn new tasks.In this paper,we simulate the dynamic wireless communication environment and focus on breaking the learning paradigm of isolated automatic MC.We innovate a research algorithm for continuous automatic MC.Firstly,a memory for storing representative old task modulation signals is built,which is employed to limit the gradient update direction of new tasks in the continuous learning stage to ensure that the loss of old tasks is also in a downward trend.Secondly,in order to better simulate the dynamic wireless communication environment,we employ the mini-batch gradient algorithm which is more suitable for continuous learning.Finally,the signal in the memory can be replayed to further strengthen the characteristics of the old task signal in the model.Simulation results verify the effectiveness of the method.

Spectrum Sensing Based on Deep Learning Classification for Cognitive Radios

Spectrum sensing is a key technology for cognitive radios.We present spectrum sensing as a classification problem and propose a sensing method based on deep learning classification.We normalize the received signal power to overcome the effects of noise power uncertainty.We train the model with as many types of signals as possible as well as noise data to enable the trained network model to adapt to untrained new signals.We also use transfer learning strategies to improve the performance for real-world signals.Extensive experiments are conducted to evaluate the performance of this method.The simulation results show that the proposed method performs better than two traditional spectrum sensing methods,i.e.,maximum-minimum eigenvalue ratio-based method and frequency domain entropy-based method.In addition,the experimental results of the new untrained signal types show that our method can adapt to the detection of these new signals.Furthermore,the real-world signal detection experiment results show that the detection performance can be further improved by transfer learning.Finally,experiments under colored noise show that our proposed method has superior detection performance under colored noise,while the traditional methods have a significant performance degradation,which further validate the superiority of our method.

Individual Differences in Cognitive Performance Regulated by Deep-Brain Activity during Mild Passive Hyperthermia and Neck Cooling

Hyperthermia-induced decline in cognitive performance is a moderate complication that poses challenges to the maintenance of safety. Although the underlying mechanism can be attributed to the disruption of brain networks, the propensity remains unclear. This study aimed to test the hypothesis that the extent of the alterations in cognitive performance is governed by the activity of deep brain structures, including monoaminergic neural systems. A decline in cognitive performance during mild hyperthermia and the beneficial effects of neck cooling were demonstrated using the Continuous Performance Test as a battery of cognitive tasks. Aspects of cognitive performance were characterized using the deep-brain activity (DBA) index as a neural activity parameter and the State-Trait Anxiety Inventory to assess the extent of alterations in cognitive performance as an individual measure. It was found that a higher average DBA index during tasks is essential for high cognitive performance in the heat. This beneficial effect of DBA is governed by the upper brainstem. This DBA benefit is more significant for individuals with higher average DBA indices at rest in a normal environment. Individual differences in cognitive performance in the heat are governed by differences in DBA. In addition, the beneficial effect of DBA on cognitive performance in heat only applies under conditions including neck cooling. This limited neck-cooling effect is attributed to anti-homeostatic thermoregulatory responses to cognitive tasks regulated by DBA.

Downlink Resource Allocation for NOMA-Based Hybrid Spectrum Access in Cognitive Network

To solve the contradiction between limited spectrum resources and increasing communication demand,this paper proposes a wireless resource allocation scheme based on the Deep Q Network(DQN)to allocate radio resources in a downlink multi-user cognitive radio(CR)network with slicing.Secondary users(SUs)are multiplexed using non-orthogonal multiple access(NOMA).The SUs use the hybrid spectrum access mode to improve the spectral efficiency(SE).Considering the demand for multiple services,the enhanced mobile broadband(eMBB)slice and ultrareliable low-latency communication(URLLC)slice were established.The proposed scheme can maximize the SE while ensuring Quality of Service(QoS)for the users.This study established a mapping relationship between resource allocation and the DQN algorithm in the CR-NOMA network.According to the signal-to-interference-plusnoise ratio(SINR)of the primary users(PUs),the proposed scheme can output the optimal channel selection and power allocation.The simulation results reveal that the proposed scheme can converge faster and obtain higher rewards compared with the Q-Learning scheme.Additionally,the proposed scheme has better SE than both the overlay and underlay only modes.

Cognitive Control and Brain Network Dynamics during Word Generation Tasks Predicted Using a Novel Event-Related Deep Brain Activity Method

There is a growing interest in the diagnosis and treatment of patients with dementia and cognitive impairment at an early stage. Recent imaging studies have explored neural mechanisms underlying cognitive dysfunction based on brain network architecture and functioning. The dorsal anterior cingulate cortex (dACC) is thought to regulate large-scale intrinsic brain networks, and plays a primary role in cognitive processing with the anterior insular cortex (aIC), thus providing salience functions. Although neural mechanisms have been elucidated at the connectivity level by imaging studies, their understanding at the activity level still remains unclear because of limited time-based resolution of conventional imaging techniques. In this study, we investigated temporal activity of the dACC during word (verb) generation tasks based on our newly developed event-related deep brain activity (ER-DBA) method using occipital electroencephalogram (EEG) alpha-2 powers with a time resolution of a few hundred milliseconds. The dACC exhibited dip-like temporal waveforms indicating deactivation in an initial stage of each trial when appropriate verbs were successfully generated. By contrast, monotonous increase was observed for incorrect responses and a decrease was detected for no responses. The dip depth was correlated with the percentage of success. Additionally, the dip depth linearly increased with increasing slow component of the DBA index at rest across all subjects. These findings suggest that dACC deactivation is essential for cognitive processing, whereas its activation is required for goal-oriented behavioral outputs, such as cued speech. Such dACC functioning, represented by the dip depth, is supported by the activity of the upper brainstem region including monoaminergic neural systems.

Spectrum Sensing Using Optimized Deep Learning Techniquesin Reconfigurable Embedded Systems

The exponential growth of Internet of Things(IoT)and 5G networks has resulted in maximum users,and the role of cognitive radio has become pivotal in handling the crowded users.In this scenario,cognitive radio techniques such as spectrum sensing,spectrum sharing and dynamic spectrum access will become essential components in Wireless IoT communication.IoT devices must learn adaptively to the environment and extract the spectrum knowledge and inferred spectrum knowledge by appropriately changing communication parameters such as modulation index,frequency bands,coding rate etc.,to accommodate the above characteristics.Implementing the above learning methods on the embedded chip leads to high latency,high power consumption and more chip area utilisation.To overcome the problems mentioned above,we present DEEP HOLE Radio sys-tems,the intelligent system enabling the spectrum knowledge extraction from the unprocessed samples by the optimized deep learning models directly from the Radio Frequency(RF)environment.DEEP HOLE Radio provides(i)an opti-mized deep learning framework with a good trade-off between latency,power and utilization.(ii)Complete Hardware-Software architecture where the SoC’s coupled with radio transceivers for maximum performance.The experimentation has been carried out using GNURADIO software interfaced with Zynq-7000 devices mounting on ESP8266 radio transceivers with inbuilt Omni direc-tional antennas.The whole spectrum of knowledge has been extracted using GNU radio.These extracted features are used to train the proposed optimized deep learning models,which run parallel on Zynq-SoC 7000,consuming less area,power,latency and less utilization area.The proposed framework has been evaluated and compared with the existing frameworks such as RFLearn,Long Term Short Memory(LSTM),Convolutional Neural Networks(CNN)and Deep Neural Networks(DNN).The outcome shows that the proposed framework has outperformed the existing framework regarding the area,power and time.More-over,the experimental results show that the proposed framework decreases the delay,power and area by 15%,20%25%concerning the existing RFlearn and other hardware constraint frameworks.

Alzheimer’s Disease Stage Classification Using a Deep Transfer Learning and Sparse Auto Encoder Method

Alzheimer’s Disease(AD)is a progressive neurological disease.Early diagnosis of this illness using conventional methods is very challenging.Deep Learning(DL)is one of the finest solutions for improving diagnostic procedures’performance and forecast accuracy.The disease’s widespread distribution and elevated mortality rate demonstrate its significance in the older-onset and younger-onset age groups.In light of research investigations,it is vital to consider age as one of the key criteria when choosing the subjects.The younger subjects are more susceptible to the perishable side than the older onset.The proposed investigation concentrated on the younger onset.The research used deep learning models and neuroimages to diagnose and categorize the disease at its early stages automatically.The proposed work is executed in three steps.The 3D input images must first undergo image pre-processing using Weiner filtering and Contrast Limited Adaptive Histogram Equalization(CLAHE)methods.The Transfer Learning(TL)models extract features,which are subsequently compressed using cascaded Auto Encoders(AE).The final phase entails using a Deep Neural Network(DNN)to classify the phases of AD.The model was trained and tested to classify the five stages of AD.The ensemble ResNet-18 and sparse autoencoder with DNN model achieved an accuracy of 98.54%.The method is compared to state-of-the-art approaches to validate its efficacy and performance.

老年遗忘型轻度认知障碍执行功能的神经机制及数字干预

阿尔茨海默病具有极高的发病率和致死率。遗忘型轻度认知障碍(Amnestic Mild Cognitive Impairment,aMCI)作为临床前驱期,探究其形成和发展机制有助于预防阿尔茨海默病的发生。现有研究显示,多个执行域缺陷与aMCI记忆衰退密切相关,但尚未回答何种执行域是关键致病因子、关键干预治疗靶标等科学问题。为突破以往研究将执行功能视作整体抑或割裂元素的局限性,本研究拟从执行功能结构全貌着眼,在提出aMCI执行功能与记忆损害关系假说的基础上,利用脑电技术系统考察aMCI抑制、刷新和转换三种执行功能子成分的时域、时频和动态脑网络特征;并结合三维卷积神经网络筛选、识别执行功能缺陷的特异性神经靶标,探索将抑制域相关神经标记物加入aMCI早期识别的可能性;最后,通过纵向因果设计分析不同靶向数字干预对aMCI患者的训练效果及神经基础,以揭示抑制域相关额顶控制网络在干预中的重要作用。本研究有望从计算认知神经视角阐明抑制是aMCI执行功能缺损和干预的认知新靶点,进而为aMCI早期识别和制定精准化诊疗方案提供循证依据。

具身认知理论视域下语文大单元教学策略研究

大单元教学是《义务教育语文课程标准(2022年版)》和学习任务群的必然要求,但在教学开展过程中存在着困境。具身认知理论主张认知不仅仅依赖于大脑,更是身体和环境互动的产物,强调教学过程中“身心一体”与“主客融合”的教学范式。基于具身认知理论的大单元教学是在新课标背景下,以具身认知为指导,以大单元教学为依托,促进学生问题解决与迁移,思维提升与创造。为此,从学习目标、学习情境、学习任务、学习评价四个方面进行设计,发挥身体在学习中的作用,以更好地实现学生核心素养发展。

基于深度学习的认知物联网频谱感知算法研究

针对认知物联网(Internet of Things, IoT)对低信噪比(Signal to Noise Ratio, SNR)的频谱感知性能低下以及传统卷积神经网络(Convolutional Neural Network, CNN)频谱感知方法提取数据特征不充分导致感知性能差等问题,提出了一种改进残差网络——ResNeXt的单节点频谱感知算法,ResNeXt只需要设置少量超参数且高度模块化,将该网络在图像处理上的优势应用在频谱感知问题上,先将接收信号转成二维矩阵并归一灰度化处理,得到灰度图像作为网络的输入。通过训练ResNeXt来提取灰度图像特征,将在线数据输入完成频谱感知。将各个次用户(Secondary User, SU)得到的评分向量矩阵直接用融合中心SoftCombinationNet(SCN)融合获得协作频谱感知结果,有效解决了传统硬融合方法检测性能低、软融合处理复杂等问题。实验结果表明,所提方法在低SNR仍能实现低虚警率、高检测概率,优于传统频谱感知方法。

丘脑底核脑深部电刺激术治疗帕金森病患者的疗效及术后认知功能分析

目的 分析帕金森病(PD)患者实施丘脑底核脑深部电刺激术(STN-DBS)的临床效果。方法选取2018-10—2022-12河南大学第一附属医院治疗的PD患者50例,采用随机数字表法分为2组各25例,对照组使用普拉克索治疗,观察组使用普拉克索联合STN-DBS治疗,对比2组患者的焦虑及抑郁程度、认知功能、帕金森核心症状、睡眠质量、生活质量。结果 治疗后2组汉密尔顿焦虑量表(HAMA)、汉密尔顿抑郁量表(HAMD)评分均呈降低趋势,观察组改变更明显(P<0.05)。治疗后2组蒙特利尔认知功能评估量表(MoCA)、简易精神状态检查量表(MMSE)、日常生活活动能力量表(ADL)、统一PD评定量表第Ⅲ部分(UPDRS-Ⅲ)评分均不同程度改善,观察组改变更明显(P<0.05)。治疗后2组帕金森睡眠评分量表(PDSS)评分呈升高趋势,39项帕金森患者生活质量问卷(PDQ-39)评分呈降低趋势,观察组改变更明显(P<0.05)。结论 PD患者在普拉克索治疗基础上联合STN-DBS可改善症状及生活质量,一定程度上提高认知功能。

基于数字孪生和强化学习的低空智联网协同认知干扰

针对低空智联网协同认知干扰决策过程中,多架电子干扰无人机对抗多部多功能雷达的干扰资源分配问题,提出了一种基于数字孪生和深度强化学习的认知干扰决策方法。首先,将协同电子干扰问题建模为马尔可夫决策问题,建立认知干扰决策系统模型,综合考虑干扰对象、干扰功率和干扰样式选择约束,构建智能体动作空间、状态空间和奖励函数。其次,在近端策略优化(Proximal policy optimization,PPO)深度强化学习算法的基础上,提出了自适应学习率近端策略优化(Adaptive learning rate proximal policy optimization,APPO)算法。同时,为了以高保真的方式提高深度强化学习算法的训练速度,提出了一种基于数字孪生的协同电子干扰决策模型训练方法。仿真结果表明,与已有的深度强化学习算法相比,APPO算法干扰效能提升30%以上,所提训练方法能够提高50%以上的模型训练速度。

基于自主认知深度时间聚类表示的隔离开关故障诊断方法

为准确识别隔离开关发生的故障,并确定故障类型,保证电网的稳定运行,提出一种基于自主认知的深度时序聚类表示模型(Autonomous-cognition deep temporal clustering representation model,AC-DTCR)对隔离开关的故障进行诊断。在数据量少且类别标签信息不可用的情况下,时间序列聚类是非常好的无监督学习技术,而AC-DTCR模型集成了时间重建和K-means目标,为提高编码器的能力,提出一种假样本生成策略和辅助分类任务,改进集群结构,获得特定于集群的时间表示。根据高压隔离开关故障模拟试验得到的电机电流数据,使用AC-DTCR模型分成四个部分对试验数据进行训练。结果表明,该模型具有良好的分类性能,与传统的分类模型和时间序列聚类模型相比,有更高的准确率,可应用于电力设备故障诊断领域中。

近期皮质下小梗死患者深髓静脉可见性与血管周围间隙扩大及认知功能的相关性研究

目的探讨近期皮质下小梗死(RSSI)患者颅内深髓静脉(DMV)可见性与不同区域血管周围间隙扩大(EPVS)及认知功能的相关性。方法回顾性连续纳入南京医科大学附属常州市第二人民医院神经内科自2022年10月至2023年10月收治的RSSI患者,所有患者入院后3 d内完成MR的常规及磁敏感加权成像(SWI)序列扫描。所有RSSI患者发病7 d内进行蒙特利尔认知评估(MoCA)量表评分。对所有患者基底节区(BG)和半卵圆中心区的EPVS进行分级评估和体积测量,使用DMV视觉评分对患者SWI幅度图或最小强度投影图上的DMV可见性进行评估,并将患者分为可见性较高的DMV低-中分组(评分0~12分,104例)及可见性较低的DMV高分组(评分13~18分,47例),采用单因素分析比较两组患者的临床和影像学资料,采用多因素Logistic回归和Spearman相关分析方法分析DMV可见性与BG-EPVS分级及体积的关系以及其与患者认知功能的关系。结果共纳入RSSI患者151例,平均年龄(69±10)岁,其中男92例(60.9%),女59例(39.1%)。DMV高分组RSSI患者的年龄[(76±5)岁比(65±10)岁,t=-10.875]、高血压病患者比例[78.7%(37/47)比54.8%(57/104),χ^(2)=7.879]、BG-EPVS分级、BG-EPVS体积[5.67(5.30,5.81)ln mm 3比4.61(3.66,5.30)ln mm 3,Z=-6.772]、脑白质高信号体积[7.67(6.23,8.43)ln mm 3比4.31(3.53,5.89)ln mm 3,Z=-6.501]均明显高于DMV低-中分组,差异均有统计学意义(均P<0.05)。DMV高分组RSSI患者的总胆固醇[3.74(3.20,4.39)mmol/L比4.09(3.47,4.96)mmol/L,Z=-2.082]、三酰甘油[1.20(0.78,1.86)mmol/L比1.53(1.05,1.99)mmol/L,Z=-2.343]、MoCA量表评分[21.0(20.0,22.0)分比24.0(22.0,25.0)分,Z=-9.862]均低于DMV低-中分组(均P<0.05)。其余基线资料差异均无统计学意义(均P>0.05)。多因素Logistic回归分析结果显示,较高的年龄(OR=1.181,95%CI:1.070~1.304,P=0.001)、中重度BG-EPVS(OR=2.441,95%CI:1.186~5.024,P=0.015)、较高的BG-EPVS体积(OR=4.987,95%CI:1.218~19.350,P=0.020)和较高的WMH体积(OR=1.285,95%CI:1.044~1.582,P=0.018)与较高的DMV评分相关。Spearman相关性分析结果显示,DMV评分与RSSI患者的BG-EPVS分级呈正相关(r=0.613,P<0.01),与BG-EPVS体积呈正相关(r=0.549,P<0.01),与RSSI患者的MoCA量表评分呈负相关(r=-0.449,P<0.01)。结论年龄、BG-EPVS分级、BG-EPVS体积和WMH体积与RSSI患者的DMV可见性相关;RSSI患者DMV的可见性越差,认知功能损伤越严重。

全频谱深度认知理论与方法

深入研究了全频谱认知与智能利用领域,强调了现有认知无线电理论和无线电监测技术在特定环境中的优势,指出在面对宽带、瞬变、未知、复杂的电磁环境时存在“认不深”的问题。为解决这一挑战,提出了多维度的深度频谱认知作为未来的发展方向,从信号多维特征与个体本征属性双重维度对辐射源目标进行精准认知。通过探索辐射源目标信号的多域特征,构建辐射源完备表征数据库,实现通信终端的本征属性认知,可以更全面地理解和利用频谱资源。分析了调制域、协议域中信号的调制识别以及协议解析等研究现状;高速目标信号识别和空域定向认知的发展瓶颈及主要研究方向。提供了较为全面的理论和方法,为未来多维度的电磁环境感知和频谱利用奠定了基础。

深度学习认知架构的反表征主义转向

当代认知研究发展出了符号主义和联结主义两种不同的范式。符号主义的计算-表征是以思想语言假说为基础的“句法图像”,具有内容与载体相分离、符号语境无关性等表征特征。深度学习是对联结主义技术的创新和深化,其认知架构是具有分布式加工和叠加存储、语境敏感和原型提取学习等特点的亚符号计算,表现出一系列的反表征特征,反映在深度网络中并不以明确的概念表征为对象的操作,推动了认知哲学中反表征主义的兴起。在充分理解符号主义和深度学习认知架构表征方式的基础上,探索二者在某种程度上的统一,也许是值得努力的目标。

语音重复任务在轻度认知功能障碍检测中的应用

轻度认知功能障碍(mild cognitive impairment,MCI)通常被视为痴呆的前驱阶段,其主要特征为认知功能轻度下降。研究表明,MCI患者中语言变化可能先于其他认知功能症状,这为早期识别和干预提供了机会。MCI患者语言特点包括语速、发音和语调等异常。五个单词测验、数字延迟匹配测试和句子重复测试等语音重复任务,是评估MCI患者语言特点的有效方法,这些任务要求患者重复特定内容,分析重复准确性,从而评估其语言功能。机器学习和深度学习技术的应用,能自动提取语音重复任务数据中的MCI相关特征,提高诊断准确性。这些技术的结合应用有助于早期发现MCI,为及时干预提供依据。

基于深度学习的动态主用户频谱感知算法

实际的频谱感知场景中主用户可能随机到达或者离开,当主用户状态在实时频谱感知期间动态变化时,现有的静态频谱感知算法性能急剧恶化。针对该现状,研究提出基于残差收缩注意力机制的动态主用户频谱感知算法。频谱感知间隔内,主用户随机到达或者随机离开的时间服从均匀分布。采用深度残差收缩网络(DRSN)提取动态主用户特征,并且滤除冗余的噪声特征;利用协调注意力模块(CAM)增强每个通道不同方向的特征信息,提高模型对动态主用户特征的表达能力。仿真结果表明,所提算法性能优于对比算法ResNet、CBAM_IQ和CBAM_Energy,所提算法对主用户随机到达或者离开服从不同分布的主用户都可以保持较高的检测概率。

基于多检查结果融合的MCI进展预测方法

为提高轻度认知障碍(MCI)患者向阿尔茨海默症(AD)阶段病情进展的预测性能,提出一种融合病人多项检查数据进行学习的半监督神经网络新型模型MVIDG。通过mRMR算法对高维特征进行降维,对病人单项检查数据使用Dual-GCN进行基础模型训练,利用改进后的MVCDN网络对各项检查数据训练出的模型进行融合,以对未来一年内病人从MCI阶段向AD阶段的病情进展进行预测。实验结果表明,所提模型可有效整合病人多项检查结果以提高预测性能,效果优于其它数据融合方法。