The Internet of Things(IoT)has orchestrated various domains in numerous applications,contributing significantly to the growth of the smart world,even in regions with low literacy rates,boosting socio-economic developm...The Internet of Things(IoT)has orchestrated various domains in numerous applications,contributing significantly to the growth of the smart world,even in regions with low literacy rates,boosting socio-economic development.This study provides valuable insights into optimizing wireless communication,paving the way for a more connected and productive future in the mining industry.The IoT revolution is advancing across industries,but harsh geometric environments,including open-pit mines,pose unique challenges for reliable communication.The advent of IoT in the mining industry has significantly improved communication for critical operations through the use of Radio Frequency(RF)protocols such as Bluetooth,Wi-Fi,GSM/GPRS,Narrow Band(NB)-IoT,SigFox,ZigBee,and Long Range Wireless Area Network(LoRaWAN).This study addresses the optimization of network implementations by comparing two leading free-spreading IoT-based RF protocols such as ZigBee and LoRaWAN.Intensive field tests are conducted in various opencast mines to investigate coverage potential and signal attenuation.ZigBee is tested in the Tadicherla open-cast coal mine in India.Similarly,LoRaWAN field tests are conducted at one of the associated cement companies(ACC)in the limestone mine in Bargarh,India,covering both Indoor-toOutdoor(I2O)and Outdoor-to-Outdoor(O2O)environments.A robust framework of path-loss models,referred to as Free space,Egli,Okumura-Hata,Cost231-Hata and Ericsson models,combined with key performance metrics,is employed to evaluate the patterns of signal attenuation.Extensive field testing and careful data analysis revealed that the Egli model is the most consistent path-loss model for the ZigBee protocol in an I2O environment,with a coefficient of determination(R^(2))of 0.907,balanced error metrics such as Normalized Root Mean Square Error(NRMSE)of 0.030,Mean Square Error(MSE)of 4.950,Mean Absolute Percentage Error(MAPE)of 0.249 and Scatter Index(SI)of 2.723.In the O2O scenario,the Ericsson model showed superior performance,with the highest R^(2)value of 0.959,supported by strong correlation metrics:NRMSE of 0.026,MSE of 8.685,MAPE of 0.685,Mean Absolute Deviation(MAD)of 20.839 and SI of 2.194.For the LoRaWAN protocol,the Cost-231 model achieved the highest R^(2)value of 0.921 in the I2O scenario,complemented by the lowest metrics:NRMSE of 0.018,MSE of 1.324,MAPE of 0.217,MAD of 9.218 and SI of 1.238.In the O2O environment,the Okumura-Hata model achieved the highest R^(2)value of 0.978,indicating a strong fit with metrics NRMSE of 0.047,MSE of 27.807,MAPE of 27.494,MAD of 37.287 and SI of 3.927.This advancement in reliable communication networks promises to transform the opencast landscape into networked signal attenuation.These results support decision-making for mining needs and ensure reliable communications even in the face of formidable obstacles.展开更多
In order to deploy short-range wireless communication devices in the primary transformer substation, a Maximum Inner Product(MIP) Method is proposed to extract the path-loss parameters in 110 kV and 220 kV primary tra...In order to deploy short-range wireless communication devices in the primary transformer substation, a Maximum Inner Product(MIP) Method is proposed to extract the path-loss parameters in 110 kV and 220 kV primary transformer substations. The maximum inner product of the testing data is calculated to find out the loss index n and the standard deviation σ, and then the pathloss models can be set up. By comparing the MIP with Minimum Mean Square estimation(MMSE) and Cumulative Sum(CUSUM), MIP can match the measured values best. In order to apply the MIP path-loss model, under the initial signal to noise ratio(SNR) at 5 dB and 10 dB, a ZigBee simulation system is constructed to validate the situation that bit error rate(BER) varies with distance. And the ZigBee devices with 5 units are tested in a 220 kV primary transformer substation. The result of the test proves that the path-loss model is accurate.展开更多
6G加速实现由万物互联到万物智联的跃迁,工业互联网(Industrial Internet of Things,IIoT)是6G通信系统重要应用场景之一。建立通用的路径损耗模型是构建IIoT的关键。在IIoT场景中,由于密集散射体和多点移动导致的传播状态变换、路径损...6G加速实现由万物互联到万物智联的跃迁,工业互联网(Industrial Internet of Things,IIoT)是6G通信系统重要应用场景之一。建立通用的路径损耗模型是构建IIoT的关键。在IIoT场景中,由于密集散射体和多点移动导致的传播状态变换、路径损耗快速变化,给建立精确的路径损耗模型带来巨大挑战。为此,提出了一种新的IIoT场景下的路径损耗模型,建立了工业场景下适用的视距(Line of Sight,LOS)概率模型;基于马尔可夫理论对传播状态转移概率建模,利用LOS概率和状态转移模型建立新的路径损耗模型。与3GPP模型进行对比以验证模型准确性,仿真结果表明,所提出的模型与射线追踪结果吻合度更高,可以准确描述IIoT场景下无线信道的路径损耗,为可靠的无线链路构建提供依据。展开更多
无线环境知识(WEK,Wireless Environment Knowledge)旨在描述环境与信道之间的映射关系,对于动态信道特征预测及优化通信资源配置具有重要意义。工业互联网(IIoT,Industrial Internet of Things)将互联网技术与传统工业相结合,具有场景...无线环境知识(WEK,Wireless Environment Knowledge)旨在描述环境与信道之间的映射关系,对于动态信道特征预测及优化通信资源配置具有重要意义。工业互联网(IIoT,Industrial Internet of Things)将互联网技术与传统工业相结合,具有场景空间大、设备分布密集、金属设备多、设备材质多样、传输信号的频率依赖性强等特点,通信环境受多种因素影响。对环境特性的准确描述有助于在IIoT等复杂电磁环境中实现更精确的路径损耗(PL,Pass Loss)预测,从而提升通信质量和可靠性。针对上述IIoT场景的多维特性,首先分析了场景中设备材质对于信号传输的影响,基于不同材质的电磁参数和频率特性构建了相应的知识系数;然后,提出了一个面向IIoT场景的WEK表示方法,基于位置信息和知识系数表示无线传播过程中散射体的反射、绕射和遮挡对于接收信号功率的贡献,并构建了一个基于WEK和神经网络(NN,Neural Network)的PL预测框架;最后,搭建了一个简单的室内IIoT场景进行仿真,验证了所提出的知识系数和WEK的有效性。展开更多
针对煤矿井下复杂环境中第五代移动通信技术(5th-Generation Mobile Communication Technology,5G)网络覆盖性能优化的问题,提出一种基于深度强化学习的矿井5G优化方案。面向10 km主运巷道场景,综合考虑巷道截面尺寸、壁面粗糙度、设备...针对煤矿井下复杂环境中第五代移动通信技术(5th-Generation Mobile Communication Technology,5G)网络覆盖性能优化的问题,提出一种基于深度强化学习的矿井5G优化方案。面向10 km主运巷道场景,综合考虑巷道截面尺寸、壁面粗糙度、设备遮挡等多重传输损耗因素,建立融合视距/非视距路径损耗模型与粗糙度衰减因子的信号传播数学模型。将深度Q网络作为价值函数近似器的强化学习智能体,并通过基站部署与发射功率将在线优化转化为多目标决策问题,以最小基站数量实现覆盖率最大化。采用动态功率调整机制,以实时优化基站发射功率,从而适应局部信号衰减的突发变化。实验结果表明,该方案可以实现95%以上的覆盖率,相较于传统静态方案减少了28%基站部署,显著提升了井下5G网络覆盖性能,并能够降低部署成本与运行功耗。展开更多
文摘The Internet of Things(IoT)has orchestrated various domains in numerous applications,contributing significantly to the growth of the smart world,even in regions with low literacy rates,boosting socio-economic development.This study provides valuable insights into optimizing wireless communication,paving the way for a more connected and productive future in the mining industry.The IoT revolution is advancing across industries,but harsh geometric environments,including open-pit mines,pose unique challenges for reliable communication.The advent of IoT in the mining industry has significantly improved communication for critical operations through the use of Radio Frequency(RF)protocols such as Bluetooth,Wi-Fi,GSM/GPRS,Narrow Band(NB)-IoT,SigFox,ZigBee,and Long Range Wireless Area Network(LoRaWAN).This study addresses the optimization of network implementations by comparing two leading free-spreading IoT-based RF protocols such as ZigBee and LoRaWAN.Intensive field tests are conducted in various opencast mines to investigate coverage potential and signal attenuation.ZigBee is tested in the Tadicherla open-cast coal mine in India.Similarly,LoRaWAN field tests are conducted at one of the associated cement companies(ACC)in the limestone mine in Bargarh,India,covering both Indoor-toOutdoor(I2O)and Outdoor-to-Outdoor(O2O)environments.A robust framework of path-loss models,referred to as Free space,Egli,Okumura-Hata,Cost231-Hata and Ericsson models,combined with key performance metrics,is employed to evaluate the patterns of signal attenuation.Extensive field testing and careful data analysis revealed that the Egli model is the most consistent path-loss model for the ZigBee protocol in an I2O environment,with a coefficient of determination(R^(2))of 0.907,balanced error metrics such as Normalized Root Mean Square Error(NRMSE)of 0.030,Mean Square Error(MSE)of 4.950,Mean Absolute Percentage Error(MAPE)of 0.249 and Scatter Index(SI)of 2.723.In the O2O scenario,the Ericsson model showed superior performance,with the highest R^(2)value of 0.959,supported by strong correlation metrics:NRMSE of 0.026,MSE of 8.685,MAPE of 0.685,Mean Absolute Deviation(MAD)of 20.839 and SI of 2.194.For the LoRaWAN protocol,the Cost-231 model achieved the highest R^(2)value of 0.921 in the I2O scenario,complemented by the lowest metrics:NRMSE of 0.018,MSE of 1.324,MAPE of 0.217,MAD of 9.218 and SI of 1.238.In the O2O environment,the Okumura-Hata model achieved the highest R^(2)value of 0.978,indicating a strong fit with metrics NRMSE of 0.047,MSE of 27.807,MAPE of 27.494,MAD of 37.287 and SI of 3.927.This advancement in reliable communication networks promises to transform the opencast landscape into networked signal attenuation.These results support decision-making for mining needs and ensure reliable communications even in the face of formidable obstacles.
基金the scientific project supported by the National Natural Science Foundation of China (No. 61571063)supported by the Beijing Municipal Natural Science Foundation (No. 3182028)
文摘In order to deploy short-range wireless communication devices in the primary transformer substation, a Maximum Inner Product(MIP) Method is proposed to extract the path-loss parameters in 110 kV and 220 kV primary transformer substations. The maximum inner product of the testing data is calculated to find out the loss index n and the standard deviation σ, and then the pathloss models can be set up. By comparing the MIP with Minimum Mean Square estimation(MMSE) and Cumulative Sum(CUSUM), MIP can match the measured values best. In order to apply the MIP path-loss model, under the initial signal to noise ratio(SNR) at 5 dB and 10 dB, a ZigBee simulation system is constructed to validate the situation that bit error rate(BER) varies with distance. And the ZigBee devices with 5 units are tested in a 220 kV primary transformer substation. The result of the test proves that the path-loss model is accurate.
文摘6G加速实现由万物互联到万物智联的跃迁,工业互联网(Industrial Internet of Things,IIoT)是6G通信系统重要应用场景之一。建立通用的路径损耗模型是构建IIoT的关键。在IIoT场景中,由于密集散射体和多点移动导致的传播状态变换、路径损耗快速变化,给建立精确的路径损耗模型带来巨大挑战。为此,提出了一种新的IIoT场景下的路径损耗模型,建立了工业场景下适用的视距(Line of Sight,LOS)概率模型;基于马尔可夫理论对传播状态转移概率建模,利用LOS概率和状态转移模型建立新的路径损耗模型。与3GPP模型进行对比以验证模型准确性,仿真结果表明,所提出的模型与射线追踪结果吻合度更高,可以准确描述IIoT场景下无线信道的路径损耗,为可靠的无线链路构建提供依据。
文摘无线环境知识(WEK,Wireless Environment Knowledge)旨在描述环境与信道之间的映射关系,对于动态信道特征预测及优化通信资源配置具有重要意义。工业互联网(IIoT,Industrial Internet of Things)将互联网技术与传统工业相结合,具有场景空间大、设备分布密集、金属设备多、设备材质多样、传输信号的频率依赖性强等特点,通信环境受多种因素影响。对环境特性的准确描述有助于在IIoT等复杂电磁环境中实现更精确的路径损耗(PL,Pass Loss)预测,从而提升通信质量和可靠性。针对上述IIoT场景的多维特性,首先分析了场景中设备材质对于信号传输的影响,基于不同材质的电磁参数和频率特性构建了相应的知识系数;然后,提出了一个面向IIoT场景的WEK表示方法,基于位置信息和知识系数表示无线传播过程中散射体的反射、绕射和遮挡对于接收信号功率的贡献,并构建了一个基于WEK和神经网络(NN,Neural Network)的PL预测框架;最后,搭建了一个简单的室内IIoT场景进行仿真,验证了所提出的知识系数和WEK的有效性。
文摘针对煤矿井下复杂环境中第五代移动通信技术(5th-Generation Mobile Communication Technology,5G)网络覆盖性能优化的问题,提出一种基于深度强化学习的矿井5G优化方案。面向10 km主运巷道场景,综合考虑巷道截面尺寸、壁面粗糙度、设备遮挡等多重传输损耗因素,建立融合视距/非视距路径损耗模型与粗糙度衰减因子的信号传播数学模型。将深度Q网络作为价值函数近似器的强化学习智能体,并通过基站部署与发射功率将在线优化转化为多目标决策问题,以最小基站数量实现覆盖率最大化。采用动态功率调整机制,以实时优化基站发射功率,从而适应局部信号衰减的突发变化。实验结果表明,该方案可以实现95%以上的覆盖率,相较于传统静态方案减少了28%基站部署,显著提升了井下5G网络覆盖性能,并能够降低部署成本与运行功耗。
