Posts Tagged ‘D3R’

Zigbee and MeshScape

Wireless Sensor Networks (WSNs) emerged in the industry around year 2000 in response to limited functionality offered by traditional forms of point-to-point wireless communications such as Bluetooth and Wi-Fi.  WSNs allow wireless devices to participate in self-forming, self-organizing and self-healing networks; that is, devices are able to communicate on multiple paths that blanket an area to enable greater flexibility and robustness.  This breakthrough makes it easier to install and change devices without costly setup and network administration.  While many competing WSNs (for example, Zigbee) operate well in networks of less than 30 nodes or sensors, they lack critical performance characteristics that are required for more challenging applications.  These characteristics include:

• Scalability (i.e. Effective and reliable communication with more than 100s of nodes and many hops in a network)

• Ability to reliably communicate in a dynamic environment (i.e. Efficient and dynamic routing with minimum overhead)

• Low power requirements for operation, including battery operated and power-harvesting routing/gateway nodes

• Full omni-directional multi-hop communication (i.e. Peer-to-Peer communication) with latencies typically measured in milliseconds per hop

Many WSN vendors manufacture Zigbee-based products.  Zigbee was founded in 2003 as the first industry standard for wireless sensor network.  Though Zigbee has had limited commercial success, it may be suitable for small, limited networks (i.e. less than 30 nodes in relatively small area).  Zigbee uses traditional ad-hoc routing protocols such as AODV and DSR.  Although these traditional techniques are good for relatively small sized networks, they have inherent limitations in scalability and reliability.  Zigbee also heavily relies on a flooding (i.e. broadcasting) technique for its path discovery and address assignment process, which consumes a lot of power and makes it very difficult to be adopted for low power applications.  As the result, it is impractical to run Zigbee routing nodes with batteries or power-harvesting technologies.

Despite such inherent technical issues and limited capabilities, Zigbee is perceived as a standard in WSN market today, especially for smart grid and home automation.  Zigbee also claims interoperability as an advantage, although true interoperability among the end-user product level is not yet proven.

MeshScape uses very different approaches to mesh network.  Due to the advanced routing protocol (Decentralized Distributed Dynamic Routing) and low power technology (“Virtually-On” distributed synchronization) that MeshScape is based on, it can cover more challenging application requirements than can Zigbee.  Decentralized Distributed Dynamic Routing (“D3R”) treats the mesh network like ”fluid flow”, and routing paths are determined dynamically at each hop with the most up-to-date information, instead of using pre-specified paths.  Unlike traditional routing techniques, D3R eliminates the need for an expensive repeated path discovery process in dynamic environments, which enables the network to be highly scalable in real-world deployments.  Since the path is determined smartly in each hop, the packet delivery is much more efficient and dramatically reduces unnecessary waste in network capacity.  The packet delivery is much more reliable since the path is dynamically determined, avoiding disturbance or interference on communication links.

MeshScape also uses patented “Virtually-On” technology to reduce power consumption for routing nodes and gateway nodes.  Each node sleeps and wakes up based on a distributed/localized synchronization mechanism which enables it to run on very low power but still makes the network highly scalable and responsive.  The application does not even perceive when the node is sleeping.  From the perspective of the application, the node appears to always be up and running, making it suitable for applications that require event-based data transmission with low latency while running on low power.  The Virtually-On technique is applied to any node in the network, including the gateway (or coordinator/network manager) node.  This means that even the gateway node can practically run with batteries or power-harvesting techniques.

Furthermore, MeshScape provides many additional high-end features such as omni-directional communication – any node can talk to any other node without going through the gateway; dynamic multi-network joining – a node can be part of multiple networks and change the associations dynamically; and dynamic subnet formation – subnets or clusters can be formed, merged and dispersed dynamically.  MeshScape also provides very flexible and easy-to-use API functions for application integration.

In summary, there may be a good place in the market for Zigbee technology for applications with relatively small-sized networks with limited capabilities requiring interoperability .  For other applications that do not necessarily require interoperability with other vendors, but do need high-end features for more challenging requirements such as a large-sized network, higher reliability, low power, and ultimate flexibility, MeshScape is the right choice.

Detailed technical comparison of MeshScape and Zigbee can be found in MeshScape vs. Zigbee Comparison Chart.

For more information about important requirements of wireless sensor networks, visit “Good Wireless Sensor Network?”


Get every new post delivered to your Inbox.

%d bloggers like this: