Wideband spectrum sensing with a high-speed analog-digital converter(ADC) presents a challenge for practical systems.The Nyquist folding receiver(NYFR) is a promising scheme for achieving cost-effective real-time spec...Wideband spectrum sensing with a high-speed analog-digital converter(ADC) presents a challenge for practical systems.The Nyquist folding receiver(NYFR) is a promising scheme for achieving cost-effective real-time spectrum sensing,which is subject to the complexity of processing the modulated outputs.In this case,a multipath NYFR architecture with a step-sampling rate for the different paths is proposed.The different numbers of digital channels for each path are designed based on the Chinese remainder theorem(CRT).Then,the detectable frequency range is divided into multiple frequency grids,and the Nyquist zone(NZ) of the input can be obtained by sensing these grids.Thus,high-precision parameter estimation is performed by utilizing the NYFR characteristics.Compared with the existing methods,the scheme proposed in this paper overcomes the challenge of NZ estimation,information damage,many computations,low accuracy,and high false alarm probability.Comparative simulation experiments verify the effectiveness of the proposed architecture in this paper.展开更多
目的:探讨强直性脊柱炎(ankylosing spondylitis,AS)胸腹部折叠畸形的CT影像学分型及评估方法。方法:回顾性分析2017年7月~2024年1月31例行胸腰椎CT检查的AS胸腰椎后凸畸形患者资料,男28例,女3例,平均年龄45.0±8.9岁。在胸腰椎CT...目的:探讨强直性脊柱炎(ankylosing spondylitis,AS)胸腹部折叠畸形的CT影像学分型及评估方法。方法:回顾性分析2017年7月~2024年1月31例行胸腰椎CT检查的AS胸腰椎后凸畸形患者资料,男28例,女3例,平均年龄45.0±8.9岁。在胸腰椎CT正中矢状面上测量胸腹折叠角(thoracoabdominal folded angle,TAFA)及剑突-耻骨联合距离(the distances between xiphoid process and superior edge of the pubis,XP),同时在脊柱全长侧位片上测量全脊椎后凸Cobb角(global kyphosis,GK)、胸椎后凸Cobb角(thoracic kyphosis,TK)、腰椎前凸Cobb角(lumbar lordosis,LL)及矢状面躯干偏移(sagittal vertical axis,SVA)。根据CT矢状面腰椎生理曲度对腹腔容积变化的影响创新性提出AS胸腹部折叠畸形的CT影像学分型,腰椎存在生理前凸时为Ⅰ型,腰椎生理曲度变直时为Ⅱ型,腰椎后凸畸形时为Ⅲ型。根据TAFA将Ⅲ型患者分为两个亚型,TAFA>90°为A亚型,TAFA≤90°为B亚型。由5名经过培训的脊柱外科医师先后对患者的临床资料进行独立评估与分型(间隔10d),采用Kendall′s W检验分析多组观察结果的一致性。采用单因素方差分析检验比较各型间上述测量参数的差异性。结果:31例患者中,胸腹部折叠畸形Ⅰ型5例、Ⅱ型8例、ⅢA型12例、ⅢB型6例。观察者间分型Kendall′s W一致性系数为0.954(P<0.001)。患者平均GK、TK、LL、SVA、TAFA及XP分别为83.7°±29.9°、48.7°±21.3°、-13.9°±25.3°、22.8±14.9cm、128.1°±50.5°及16.8±8.9cm;各组TAFA、XP测量数值间Kendall′s W一致性系数分别为0.946(P<0.001)和0.979(P<0.001);各分型间TAFA及XP两两比较均具有显著性差异(P<0.001)。结论:CT影像学分型可以客观评价AS胸腹部折叠畸形情况,剑突-耻骨联合距离及胸腹折叠角是评估AS胸腹部折叠畸形的重要指标。展开更多
传统CNN算法在花生荚果外观识别任务中存在内存密集型和计算密集型问题,以及其在资源受限的边缘终端上部署困难,基于此,该研究提出了一种高效的花生荚果识别模型——PPINET(peanut pod identification network),以适应嵌入式设备的资源...传统CNN算法在花生荚果外观识别任务中存在内存密集型和计算密集型问题,以及其在资源受限的边缘终端上部署困难,基于此,该研究提出了一种高效的花生荚果识别模型——PPINET(peanut pod identification network),以适应嵌入式设备的资源限制需求。该模型通过结合深度可分离卷积和倒残差结构显著降低参数量和计算量,同时保留特征提取能力,并引入MQA(multi-query attention)模块增强关键特征提取,并利用TuNAS(easy-to-tune and scalable implementation of efficient neural architecture search with weight sharing)策略优化模型结构,使其在资源受限设备上表现优异。此外,采用ResNet(residual neural network)进行知识蒸馏配合三折交叉验证训练提升精度,最终量化为RKNN格式并在瑞芯微RK3588上实现NPU加速部署。PPINET模型尺寸仅为1.85 MB,参数量为0.49 M,浮点运算数为0.30G。PPINET在花生荚果分类中表现优异,准确率达98.65%,在RK3588上推理速度达321 fps。该模型具备较高的识别准确率和快速的识别速度,能够实现花生荚果的实时精准检测。展开更多
基金supported by the Key Projects of the 2022 National Defense Science and Technology Foundation Strengthening Plan 173 (Grant No.2022-173ZD-010)the Equipment PreResearch Foundation of The State Key Laboratory (Grant No.6142101200204)。
文摘Wideband spectrum sensing with a high-speed analog-digital converter(ADC) presents a challenge for practical systems.The Nyquist folding receiver(NYFR) is a promising scheme for achieving cost-effective real-time spectrum sensing,which is subject to the complexity of processing the modulated outputs.In this case,a multipath NYFR architecture with a step-sampling rate for the different paths is proposed.The different numbers of digital channels for each path are designed based on the Chinese remainder theorem(CRT).Then,the detectable frequency range is divided into multiple frequency grids,and the Nyquist zone(NZ) of the input can be obtained by sensing these grids.Thus,high-precision parameter estimation is performed by utilizing the NYFR characteristics.Compared with the existing methods,the scheme proposed in this paper overcomes the challenge of NZ estimation,information damage,many computations,low accuracy,and high false alarm probability.Comparative simulation experiments verify the effectiveness of the proposed architecture in this paper.
文摘目的:探讨强直性脊柱炎(ankylosing spondylitis,AS)胸腹部折叠畸形的CT影像学分型及评估方法。方法:回顾性分析2017年7月~2024年1月31例行胸腰椎CT检查的AS胸腰椎后凸畸形患者资料,男28例,女3例,平均年龄45.0±8.9岁。在胸腰椎CT正中矢状面上测量胸腹折叠角(thoracoabdominal folded angle,TAFA)及剑突-耻骨联合距离(the distances between xiphoid process and superior edge of the pubis,XP),同时在脊柱全长侧位片上测量全脊椎后凸Cobb角(global kyphosis,GK)、胸椎后凸Cobb角(thoracic kyphosis,TK)、腰椎前凸Cobb角(lumbar lordosis,LL)及矢状面躯干偏移(sagittal vertical axis,SVA)。根据CT矢状面腰椎生理曲度对腹腔容积变化的影响创新性提出AS胸腹部折叠畸形的CT影像学分型,腰椎存在生理前凸时为Ⅰ型,腰椎生理曲度变直时为Ⅱ型,腰椎后凸畸形时为Ⅲ型。根据TAFA将Ⅲ型患者分为两个亚型,TAFA>90°为A亚型,TAFA≤90°为B亚型。由5名经过培训的脊柱外科医师先后对患者的临床资料进行独立评估与分型(间隔10d),采用Kendall′s W检验分析多组观察结果的一致性。采用单因素方差分析检验比较各型间上述测量参数的差异性。结果:31例患者中,胸腹部折叠畸形Ⅰ型5例、Ⅱ型8例、ⅢA型12例、ⅢB型6例。观察者间分型Kendall′s W一致性系数为0.954(P<0.001)。患者平均GK、TK、LL、SVA、TAFA及XP分别为83.7°±29.9°、48.7°±21.3°、-13.9°±25.3°、22.8±14.9cm、128.1°±50.5°及16.8±8.9cm;各组TAFA、XP测量数值间Kendall′s W一致性系数分别为0.946(P<0.001)和0.979(P<0.001);各分型间TAFA及XP两两比较均具有显著性差异(P<0.001)。结论:CT影像学分型可以客观评价AS胸腹部折叠畸形情况,剑突-耻骨联合距离及胸腹折叠角是评估AS胸腹部折叠畸形的重要指标。
