Abstract:
A wireless mesh network includes heterogeneous types of nodes, including continuously-powered nodes and battery-powered nodes. The battery-powered nodes may reside in a sleeping state most of the time to conserve power. The various nodes in the network may communicate with one another by transmitting and receiving at scheduled times and on scheduled frequencies. The battery-powered nodes may become active during the scheduled transmit and receive times. Network management nodes may facilitate network formation by transmitting information that reflects the scheduled transmit and receive times across the network. Based on this data, the continuously-powered nodes and battery-powered nodes may establish communication links with one another.
Abstract:
A wireless mesh network is configured to manage a power grid. Each node within the wireless mesh network is configured to detect and classify voltage fluctuations in power supplied by an upstream transformer coupled to the power grid. When a given node detects a particular type of fluctuation (i.e., an “event”), the node generates a timestamped event classification that reflects the type of event and a time when the event occurred. A server configured to manage the wireless mesh network receives timestamped event classifications from each node within the wireless mesh network and then performs a time correlation with the received timestamped event classifications to determine which nodes detected similar events. When two or more nodes detected the same event at similar times, the server determines that those nodes are coupled to the same transformer.
Abstract:
Nodes within a wireless mesh network are configured to monitor time series data associated with a utility network, including voltage fluctuations, current levels, temperature data, humidity measurements, and other observable physical quantities. The nodes execute stream functions to process the recorded time series data and generate data streams. The node is configured to transmit generated data streams to neighboring nodes. A neighboring node may execute other stream functions to process the received data stream(s), thereby generating additional data streams. A server coupled to the wireless mesh network collects and processes the data streams to identify events occurring within the network.
Abstract:
A node within a wireless endpoint device may be coupled to multiple heterogeneous networks simultaneously. The node is configured to select between the different networks based on various constraints associated with the endpoint device, applications executing on the endpoint device, traffic routed by the endpoint device, and constraints associated with the multiple networks. Based on these different constraints, and based on the current operating mode of the node, the node rates each network, and then selects the network with the highest rating to be used for routing purposes.
Abstract:
One embodiment of the present disclosure sets forth a technique for convergence and automatic disabling of access points in a wireless mesh network. Specifically, an access point within a wireless mesh network computes one or more network metrics to determine whether the metrics are unfavorable or favorable. If the network metrics are favorable, then the access point disables the access point's network connection. An access point turns the network connection back on based on whether a routing was lost for at least a preset amount of time, utilization of one or more neighboring access points is above a preset value, or one or more network metrics have degraded by a certain percentage value. One advantage of this approach is that cost savings may be achieved when the number of access points dynamically changes to accommodate varying communications conditions.
Abstract:
One embodiment of the present disclosure sets forth a technique for convergence and automatic disabling of access points in a wireless mesh network. Specifically, an access point within a wireless mesh network computes one or more network metrics to determine whether the metrics are unfavorable or favorable. If the network metrics are favorable, then the access point disables the access point's network connection. An access point turns the network connection back on based on whether a routing was lost for at least a preset amount of time, utilization of one or more neighboring access points is above a preset value, or one or more network metrics have degraded by a certain percentage value. One advantage of this approach is that cost savings may be achieved when the number of access points dynamically changes to accommodate varying communications conditions.
Abstract:
Nodes within a wireless mesh network are configured to monitor time series data associated with a utility network (or any other device network). One or more servers coupled to the wireless mesh network configures a data ingestion cloud to receive and process the time series data from the nodes to generate data streams. The server(s) also configure a distributed processing cloud to perform historical analysis on data streams, and a real-time processing cloud to perform real-time analysis on data streams. The distributed processing cloud and the real-time processing cloud may interoperate with one another in response to processing the data streams. Specifically, the real-time processing cloud may trigger a historical analysis on the distributed processing cloud, and the distributed processing cloud may trigger real-time processing on the real-time processing cloud. Any of the processing clouds may encompass edge nodes configured to perform real-time processing and generate data streams.
Abstract:
A node within a wireless mesh network is configured to record a zero crossing of alternating current or alternating voltage drawn by a single-phase power consumer and a precise timestamp when the zero crossing occurred, thereby generating timestamped zero crossing data. The node receives similar zero crossing data from a neighboring node. The node then compares the timestamped zero crossing data with the received zero crossing data to determine whether the phase associated with the node is equivalent to, leads, or lags the phase associated with the neighboring node. The node then acquires a positive phase identification associated with the neighboring node. Based on the phase identification, and based on the phase difference between the two nodes, the node infers the phase associated with the single-phase power consumer. That phase indicates the specific power line within a three-phase power distribution network to which the single-phase power consumer is coupled.
Abstract:
Nodes within a wireless mesh network are configured to monitor time series data associated with a utility network (or any other device network). One or more servers coupled to the wireless mesh network configures a data ingestion cloud to receive and process the time series data from the nodes to generate data streams. The server(s) also configure a distributed processing cloud to perform historical analysis on data streams, and a real-time processing cloud to perform real-time analysis on data streams. The distributed processing cloud and the real-time processing cloud may interoperate with one another in response to processing the data streams. Specifically, the real-time processing cloud may trigger a historical analysis on the distributed processing cloud, and the distributed processing cloud may trigger real-time processing on the real-time processing cloud. Any of the processing clouds may encompass edge nodes configured to perform real-time processing and generate data streams.
Abstract:
Nodes within a wireless mesh network are configured to monitor time series data associated with a utility network, including voltage fluctuations, current levels, temperature data, humidity measurements, and other observable physical quantities. The nodes execute stream functions to process the recorded time series data and generate data streams. The node is configured to transmit generated data streams to neighboring nodes. A neighboring node may execute other stream functions to process the received data stream(s), thereby generating additional data streams. A server coupled to the wireless mesh network collects and processes the data streams to identify events occurring within the network.