A study of wireless technologies for IoT applications in terms of power consumption has been presented in this paper. The study focuses on the importance of using low power wireless techniques and modules in IoT appli...A study of wireless technologies for IoT applications in terms of power consumption has been presented in this paper. The study focuses on the importance of using low power wireless techniques and modules in IoT applications by introducing a comparative between different low power wireless communication techniques such as ZigBee, Low Power Wi-Fi, 6LowPAN, LPWA and their modules to conserve power and longing the life for the IoT network sensors. The approach of the study is in term of protocol used and the particular module that achieve that protocol. The candidate protocols are classified according to the range of connectivity between sensor nodes. For short ranges connectivity the candidate protocols are ZigBee, 6LoWPAN and low power Wi-Fi. For long connectivity the candidate is LoRaWAN protocol. The results of the study demonstrate that the choice of module for each protocol plays a vital role in battery life due to the difference of power consumption for each module/protocol. So, the evaluation of protocols with each other depends on the module used.展开更多
针对现有目标检测模型在无依托供电场景存在检测效果不稳定、小目标大量漏检的问题,基于YOLOv4-tiny提出一种改进模型AMS-YOLOv4-tiny。通过在主干网之后引入更平滑的Mish函数、设计一种浅层特征加固的特征融合网络SCFPN、反复嵌入通道...针对现有目标检测模型在无依托供电场景存在检测效果不稳定、小目标大量漏检的问题,基于YOLOv4-tiny提出一种改进模型AMS-YOLOv4-tiny。通过在主干网之后引入更平滑的Mish函数、设计一种浅层特征加固的特征融合网络SCFPN、反复嵌入通道注意力机制3种策略,大幅提升预测特征层对目标的表达能力。实验结果表明,算法在PASCAL VOC07+12数据集上的mAP(mean of average precision)达到87.19%,相比YOLOv4-tiny提高4.45%,且部署在嵌入式设备上进行可行性验证,满足多种复杂场景下人车检测任务的精度与实时性要求。展开更多
High performance with low power consumption is an essential factor in wireless sensor networks （WSN）. In order to address the issue on the lifetime and the consumption of nodes in WSNs, an improved ad hoc on-demand ...High performance with low power consumption is an essential factor in wireless sensor networks （WSN）. In order to address the issue on the lifetime and the consumption of nodes in WSNs, an improved ad hoc on-demand distance vector routing （IAODV） algorithm is proposed based on AODV and LAR protocols. This algorithm is a modified on-demand routing algorithm that limits data forwarding in the searching domain, and then chooses the route on basis of hop count and power consumption. The simulation results show that the algorithm can effectively reduce power consumption as well as prolong the network lifetime.展开更多
针对低功耗有损网络现有移动性支持路由算法中父节点切换频繁、移动检测机制使用不可靠的瞬时路由度量、未考虑链路质量对能耗影响等问题,提出一种基于剩余通信时间(Remaining Communication Time, RCT)的移动感知节能父节点选择算法。...针对低功耗有损网络现有移动性支持路由算法中父节点切换频繁、移动检测机制使用不可靠的瞬时路由度量、未考虑链路质量对能耗影响等问题,提出一种基于剩余通信时间(Remaining Communication Time, RCT)的移动感知节能父节点选择算法。首先,为了降低父节点切换次数并节省节点能耗,构建了基于RCT和通信成本的父节点选择函数,移动节点基于该函数值选择下一父节点。其次,为了避免在移动场景下由于瞬时路由度量不可靠而导致的不必要的父节点切换,在算法的移动检测阶段提出一种基于接收信号强度指标(Indicator of Received Signal Strength,RSSI)和RCT值的移动检测机制,父节点需要结合RSSI和RCT两个值来共同判断移动节点是否需要进行父节点切换。理论分析和仿真结果表明,算法在控制开销、网络生存时间、能耗等方面的性能均得到了提升。展开更多
文摘A study of wireless technologies for IoT applications in terms of power consumption has been presented in this paper. The study focuses on the importance of using low power wireless techniques and modules in IoT applications by introducing a comparative between different low power wireless communication techniques such as ZigBee, Low Power Wi-Fi, 6LowPAN, LPWA and their modules to conserve power and longing the life for the IoT network sensors. The approach of the study is in term of protocol used and the particular module that achieve that protocol. The candidate protocols are classified according to the range of connectivity between sensor nodes. For short ranges connectivity the candidate protocols are ZigBee, 6LoWPAN and low power Wi-Fi. For long connectivity the candidate is LoRaWAN protocol. The results of the study demonstrate that the choice of module for each protocol plays a vital role in battery life due to the difference of power consumption for each module/protocol. So, the evaluation of protocols with each other depends on the module used.
文摘针对现有目标检测模型在无依托供电场景存在检测效果不稳定、小目标大量漏检的问题,基于YOLOv4-tiny提出一种改进模型AMS-YOLOv4-tiny。通过在主干网之后引入更平滑的Mish函数、设计一种浅层特征加固的特征融合网络SCFPN、反复嵌入通道注意力机制3种策略,大幅提升预测特征层对目标的表达能力。实验结果表明,算法在PASCAL VOC07+12数据集上的mAP(mean of average precision)达到87.19%,相比YOLOv4-tiny提高4.45%,且部署在嵌入式设备上进行可行性验证,满足多种复杂场景下人车检测任务的精度与实时性要求。
基金supported by the National Natural Science Foundation of China under Grant Nos.61373135,60973140,and 61170276Key University Science Research Project of Jiangsu Province under Grant No.12KJA520003+1 种基金Project for Production Study&Research of Jiangsu Province under Grant No.BY2013011The Science and Technology Enterprises Innovation Fund Project of Jiangsu Province under Grant No.BC2013027
文摘High performance with low power consumption is an essential factor in wireless sensor networks （WSN）. In order to address the issue on the lifetime and the consumption of nodes in WSNs, an improved ad hoc on-demand distance vector routing （IAODV） algorithm is proposed based on AODV and LAR protocols. This algorithm is a modified on-demand routing algorithm that limits data forwarding in the searching domain, and then chooses the route on basis of hop count and power consumption. The simulation results show that the algorithm can effectively reduce power consumption as well as prolong the network lifetime.
文摘针对低功耗有损网络现有移动性支持路由算法中父节点切换频繁、移动检测机制使用不可靠的瞬时路由度量、未考虑链路质量对能耗影响等问题,提出一种基于剩余通信时间(Remaining Communication Time, RCT)的移动感知节能父节点选择算法。首先,为了降低父节点切换次数并节省节点能耗,构建了基于RCT和通信成本的父节点选择函数,移动节点基于该函数值选择下一父节点。其次,为了避免在移动场景下由于瞬时路由度量不可靠而导致的不必要的父节点切换,在算法的移动检测阶段提出一种基于接收信号强度指标(Indicator of Received Signal Strength,RSSI)和RCT值的移动检测机制,父节点需要结合RSSI和RCT两个值来共同判断移动节点是否需要进行父节点切换。理论分析和仿真结果表明,算法在控制开销、网络生存时间、能耗等方面的性能均得到了提升。
