The deflection routing protocol is an effective contention resolution in Optical Burst Switching network, However, it can worsen loss performance of non-deflected burst on the deflection route. To improve the burst lo...The deflection routing protocol is an effective contention resolution in Optical Burst Switching network, However, it can worsen loss performance of non-deflected burst on the deflection route. To improve the burst loss performance, a limited deflection routing scheme based on burst loss threshold is proposed to prevent injudicious deflection routing. By using threshold check function, it restrainedly allows the deflected burst to preemptive network resource, consequently, improve the QoS performance of non-deflected burst. Simulation results show that the scheme can efficiently prevent deflected burst contending with non-deflected burst on deflection route, and effectively improve the burst loss performance of entire networks.展开更多
In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is...In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is proposed to characterize the packet loss threshold. Our results show that the packet loss threshold can be optimized with a relative small buffer size. Meanwhile, a large buffer size will increase the travel time. Furthermore, we propose a Buffered-Shortest-Path-First(BSPF) queuing strategy. Compared to the traditional First-In-First-Out(FIFO) strategy, BSPF can not only increase the packet loss threshold but can also significantly decrease the travel length and travel time in both identical and heterogeneous node capacity cases. Our study will help to improve the traffic performance in finite buffer networks.展开更多
Objective: In order to provide a theoretical basis for the revision of the current diagnostic criteria for occupational noise-induced deafness (ONID), we evaluated the degree of ONID by analyzing different high-freque...Objective: In order to provide a theoretical basis for the revision of the current diagnostic criteria for occupational noise-induced deafness (ONID), we evaluated the degree of ONID by analyzing different high-frequency-hearing- threshold-weighted values (HFTWVs). Methods: A retrospective study was conducted to evaluate the diagnosis of patients with ONID from January 2016 to January 2017 in Guangdong province, China. Based on 3 hearing tests (each interval between the tests was greater than 3 days), the minimum threshold value of each frequency was obtained using the 2007 edition’s diagnostic criteria for ONID. The speech frequency and the HFTWVs were analyzed based on age, noise exposure, and diagnostic classi-fication using SPSS21.0. Results: 168 patients in total were involved in this study, 154 males and 14 females, and the average age was 41.18 ± 6.07. The diagnosis rate was increased by the weighted value of the high frequencies and was more than the mean value of the pure speech frequency (MVPSF). The diagnosis rate for the weighted 4 kHz frequency level increased by 13.69% (χ2 = 9.880, P = 0.002), the weighted 6 kHz level increased by 15.47% (χ2 = 9.985, P = 0.002), and the weighted 4 kHz + 6 kHz level increased by 15.47% (χ2 = 9.985, P = 0.002). The differences were all statistically significant. The diagnostic rate of the different thresholds showed no obvious difference between the genders. The age groups were divided into less than or equal to 40 years old (group A) and 40 - 50 years old (group B). There were several groups with a high frequency: high frequency weighted 4 kHz ( group A χ2 = 3.380, P = 0.050;group B χ2 = 4.054, P = 0.032), high frequency weighted 6 kHz (group A χ2 = 6.362, P = 0.012;group B χ2 = 4.054, P = 0.032), weighted 4 kHz + 6 kHz (group A χ2 = 6.362 P = 0.012;B χ2 = 4.054, P = 0.032) than those of MVPSF in the same group on ONID diagnosis rate. The differences between the groups were statistically significant. There was no significant difference between the age groups (χ2 = 2.265, P = 0.944). The better ear’s (the smaller hearing threshold weighted value) MVPSF and the weighted values for the different high frequencies were examined in light of the number of working years;the group that was exposed to noise for more than 10 years had significantly higher values than those of the average thresholds of each frequency band in the groups with 3 - 5 years of exposure (F = 2.271, P = 0.001) and 6 - 10 years of exposure (F = 1.563, P = 0.046). The differences were statistically significant. The different HFTWVs were higher than those of the MVPSF values, and the high frequency weighted 4 kHz + 6 kHz level showed the greatest difference, with an average increase of 2.83 dB. The diagnostic rate that included the weighted high frequency values was higher for the mild, moderate, and severe cases than those patients who were only screened with the pure frequency tests. The results of the comparisons of the diagnosis rates for mild ONID were as follows: the weighted 3 kHz high frequency level (χ2 = 3.117, P = 0.077) had no significant difference, but the weighted 4 kHz level (χ2 = 10.835, P = 0.001), 6 kHz level (χ2 = 9.985, P = 0.002), 3 kHz + 4 kHz level (χ2 = 6.315, P = 0.012), 3 kHz + 6 kHz level (χ2 = 6.315, P = 0.012), 4 kHz + 6 kHz level (χ2 = 9.985, P = 0.002), and 3 kHz + 4 kHz + 6 kHz level (χ2 = 7.667, P = 0.002) were significantly higher than the diagnosis rate of the mean value of the PSF. There was no significant difference between the 2 groups for the moderate and severe grades (P > 0.05). Conclusion: Different HFTWVs increase the diagnostic rate of ONID. The weighted 4 kHz, 6 kHz, and 4 kHz + 6 kHz high frequency values greatly affected the diagnostic results, and the weighted 4 kHz + 6 kHz high frequency hearing threshold value has the maximum the effect on the ONID diagnosis results.展开更多
不同织机由于生产情况和影响参数各异,实际的织布效率和了机时间也存在着很大的差别。针对利用预先设定好的计划生产静态参数对织机了机时间进行计算时,存在理论计算值与实际织机了机时间偏差过大的问题,提出了一种基于长短时记忆(Long ...不同织机由于生产情况和影响参数各异,实际的织布效率和了机时间也存在着很大的差别。针对利用预先设定好的计划生产静态参数对织机了机时间进行计算时,存在理论计算值与实际织机了机时间偏差过大的问题,提出了一种基于长短时记忆(Long short term memory,LSTM)循环神经网络的织机了机预测方法。从织机经纬向停车情况、人员工作效率、加工布匹品种3个方面出发,分析影响织机了机时间的各类因素,构建了具有时间序列特性的织机生产情况数据集。通过设置时间进度系数动态调整模型在织轴整个生命周期内的预测情况,并从损失程度和训练耗时两方面考虑对模型性能进行优化。最后,利用8组实验数据对模型的可靠性进行验证。结果表明:模型在了机预测截止时间的前30 h至前6 h,模型的预测结果值与实际值之间的平均误差范围为0.84 h至1.52 h,满足对实际生产时的所需指标要求。展开更多
基金This workis supported by project of National Science Foundation of China (No.90304004) , National"863"Project Plan (No.2005AA122310) , Keyproject of Science and Technology of National Education Department (No.204125) , project of Science Foundation of Chongqing Science and Technology Commission(No.2005BB2062) , Youth Science and Technology Grant of CQUPT(No.A2005-28) .
文摘The deflection routing protocol is an effective contention resolution in Optical Burst Switching network, However, it can worsen loss performance of non-deflected burst on the deflection route. To improve the burst loss performance, a limited deflection routing scheme based on burst loss threshold is proposed to prevent injudicious deflection routing. By using threshold check function, it restrainedly allows the deflected burst to preemptive network resource, consequently, improve the QoS performance of non-deflected burst. Simulation results show that the scheme can efficiently prevent deflected burst contending with non-deflected burst on deflection route, and effectively improve the burst loss performance of entire networks.
基金Project supported by the National Key Research and Development Program of China(Grant No.2016YFC0802508)the National Natural Science Foundation of China(Grant Nos.11672289 and 61503355)the support from the Chinese Scholarship Council
文摘In real complex systems, the limited storage capacity of physical devices often results in the loss of data. We study the effect of buffer size on packet loss threshold in scale-free networks. A new order parameter is proposed to characterize the packet loss threshold. Our results show that the packet loss threshold can be optimized with a relative small buffer size. Meanwhile, a large buffer size will increase the travel time. Furthermore, we propose a Buffered-Shortest-Path-First(BSPF) queuing strategy. Compared to the traditional First-In-First-Out(FIFO) strategy, BSPF can not only increase the packet loss threshold but can also significantly decrease the travel length and travel time in both identical and heterogeneous node capacity cases. Our study will help to improve the traffic performance in finite buffer networks.
文摘Objective: In order to provide a theoretical basis for the revision of the current diagnostic criteria for occupational noise-induced deafness (ONID), we evaluated the degree of ONID by analyzing different high-frequency-hearing- threshold-weighted values (HFTWVs). Methods: A retrospective study was conducted to evaluate the diagnosis of patients with ONID from January 2016 to January 2017 in Guangdong province, China. Based on 3 hearing tests (each interval between the tests was greater than 3 days), the minimum threshold value of each frequency was obtained using the 2007 edition’s diagnostic criteria for ONID. The speech frequency and the HFTWVs were analyzed based on age, noise exposure, and diagnostic classi-fication using SPSS21.0. Results: 168 patients in total were involved in this study, 154 males and 14 females, and the average age was 41.18 ± 6.07. The diagnosis rate was increased by the weighted value of the high frequencies and was more than the mean value of the pure speech frequency (MVPSF). The diagnosis rate for the weighted 4 kHz frequency level increased by 13.69% (χ2 = 9.880, P = 0.002), the weighted 6 kHz level increased by 15.47% (χ2 = 9.985, P = 0.002), and the weighted 4 kHz + 6 kHz level increased by 15.47% (χ2 = 9.985, P = 0.002). The differences were all statistically significant. The diagnostic rate of the different thresholds showed no obvious difference between the genders. The age groups were divided into less than or equal to 40 years old (group A) and 40 - 50 years old (group B). There were several groups with a high frequency: high frequency weighted 4 kHz ( group A χ2 = 3.380, P = 0.050;group B χ2 = 4.054, P = 0.032), high frequency weighted 6 kHz (group A χ2 = 6.362, P = 0.012;group B χ2 = 4.054, P = 0.032), weighted 4 kHz + 6 kHz (group A χ2 = 6.362 P = 0.012;B χ2 = 4.054, P = 0.032) than those of MVPSF in the same group on ONID diagnosis rate. The differences between the groups were statistically significant. There was no significant difference between the age groups (χ2 = 2.265, P = 0.944). The better ear’s (the smaller hearing threshold weighted value) MVPSF and the weighted values for the different high frequencies were examined in light of the number of working years;the group that was exposed to noise for more than 10 years had significantly higher values than those of the average thresholds of each frequency band in the groups with 3 - 5 years of exposure (F = 2.271, P = 0.001) and 6 - 10 years of exposure (F = 1.563, P = 0.046). The differences were statistically significant. The different HFTWVs were higher than those of the MVPSF values, and the high frequency weighted 4 kHz + 6 kHz level showed the greatest difference, with an average increase of 2.83 dB. The diagnostic rate that included the weighted high frequency values was higher for the mild, moderate, and severe cases than those patients who were only screened with the pure frequency tests. The results of the comparisons of the diagnosis rates for mild ONID were as follows: the weighted 3 kHz high frequency level (χ2 = 3.117, P = 0.077) had no significant difference, but the weighted 4 kHz level (χ2 = 10.835, P = 0.001), 6 kHz level (χ2 = 9.985, P = 0.002), 3 kHz + 4 kHz level (χ2 = 6.315, P = 0.012), 3 kHz + 6 kHz level (χ2 = 6.315, P = 0.012), 4 kHz + 6 kHz level (χ2 = 9.985, P = 0.002), and 3 kHz + 4 kHz + 6 kHz level (χ2 = 7.667, P = 0.002) were significantly higher than the diagnosis rate of the mean value of the PSF. There was no significant difference between the 2 groups for the moderate and severe grades (P > 0.05). Conclusion: Different HFTWVs increase the diagnostic rate of ONID. The weighted 4 kHz, 6 kHz, and 4 kHz + 6 kHz high frequency values greatly affected the diagnostic results, and the weighted 4 kHz + 6 kHz high frequency hearing threshold value has the maximum the effect on the ONID diagnosis results.
文摘目的探讨短纯音刺激皮层听觉诱发电位(cortical auditory evoked potentials,CAEP)对听力评估的价值。方法受试者为听力正常成人19例(36耳),感音神经性聋成人患者21例(21耳),以500、1000、2000和4000 Hz短纯音作为刺激声,在醒觉状态下颅顶电极记录CAEP的N1-P2复合波,分析N1-P2阈值与纯音听阈的相关性,以及两者差值在不同人群中的差异。结果所有受试者四个测试频率的N1-P2阈值与对应频率纯音听阈值的相关系数依次为0.971、0.967、0.958、0.955,呈显著的线性相关;听力正常组36耳500、1000、2000和4000 Hz的N1-P2阈值上限(95%置信区间)依次为32、27、32、39 dB SPL,即正常人阈值上限为32.5 dB SPL;正常人N1-P2阈值与纯音听阈的差值依次为18.9±6.1、11.6±5.9、13.3±7.0、18.7±8.3 dB;感音神经性聋患者两者差值依次为10.3±6.0、6.8±8.9、6.5±9.5、12.9±11.4 dB,四个测试频率的差值在两组人群中的差异均有统计学意义(P<0.05)。结论听力正常人短纯音CAEP的N1-P2平均阈值小于32.5 dB SPL;N1-P2阈值与对应频率纯音听阈相关性非常好,听力评估中使用N1-P2阈值预测听力时应注意刺激声的校准及校正值的使用。
文摘不同织机由于生产情况和影响参数各异,实际的织布效率和了机时间也存在着很大的差别。针对利用预先设定好的计划生产静态参数对织机了机时间进行计算时,存在理论计算值与实际织机了机时间偏差过大的问题,提出了一种基于长短时记忆(Long short term memory,LSTM)循环神经网络的织机了机预测方法。从织机经纬向停车情况、人员工作效率、加工布匹品种3个方面出发,分析影响织机了机时间的各类因素,构建了具有时间序列特性的织机生产情况数据集。通过设置时间进度系数动态调整模型在织轴整个生命周期内的预测情况,并从损失程度和训练耗时两方面考虑对模型性能进行优化。最后,利用8组实验数据对模型的可靠性进行验证。结果表明:模型在了机预测截止时间的前30 h至前6 h,模型的预测结果值与实际值之间的平均误差范围为0.84 h至1.52 h,满足对实际生产时的所需指标要求。
