SpiderMesh

/SpiderMesh
SpiderMesh2019-10-25T15:31:15+00:00

The Wireless Mesh Network Explained

SpiderMesh is a cooperative wireless mesh network technology.

It was designed with scalability in mind. As a result, this wireless technology is perfectly tailored to tackle the very difficult task of connecting millions of devices on the same network, seamlessly.

Let’s compare SpiderMesh to standard technologies. But first, let’s go through some basics on wireless networks.

Defining an array of sensors

A sensor network can be viewed as a network of autonomous microsystems (battery operated) scattered in a given space and communicating with each other via a wireless link. The space where the sensors act is called a network field.

In such a network, the user can contact the sensors through a gateway, which bridges the sensor field and the main computer.
dot white on blue background

The Sensors

dot red on blue background

The Gateway

Network topology

The network topology is a schematic representation of the different nodes of a network (sensors, gateways, relays, etc..) and their physical links. Networks usually have a bus, star or mesh topology.

wireless network topology mesh tree star point to point

Comparing wired and wireless technologies

Two main technological choices are available for connecting a sensor field to a gateway node: either

1) wired, or 2) wireless

Wired network topology

A communication strategy with wires usually implies the use of a bus network topology in which each node (sensor) is connected to only two nearest neighbors in order to minimize the length of the electrical wiring.

Typically, a wired network topology is a bus topology: there is only one communication pathway. Consequently, if there is a node failure, the entire network downstream of it is disabled. For example, if the system is struck by lightning, the entire network is affected by the surge.

Wireless network topology

Today, wireless networks are ubiquitous everywhere, in many fields of application. However, they are currently not widely used for outdoor long-term deployments for applications requiring a large area coverage when grid power is unavailable.

Why?

Because most installations use a wireless star network topology in which sensors communicate directly to the master node, gateway, and therefore do not interact in any way with their neighbors.

Except…

When there are obstructions in the line of sight between sensors and their gateway, communication is no longer possible. As a result, there are three common patches available in this case.

wireless network tree topology 3d representation
wireless point to point network topology 3d representation

Solution 1 : Increase the transmission power

Increasing the transmission power can only eliminate communication problems for partial occlusions and/or over relatively short distances.

But
– The transmission power is limited by standards to 4W (for the ISM band)
– The cost of the transmitter increases substantially when increasing transmit power
– Ineffective for higher levels of occlusion.

Solution 2 : Increase the gateway height

Increasing the gateway height can potentially mitigate many types of occlusions.

But
– Installation of a tower is a very expensive investment
– Installation of a tower involves a lot of time and effort
wireless point to point network topology with antenna
wireless point-to-point or star network topology with antenna and cellular

Solution 3 : Use external infrastructure

The use of external infrastructure can solve all aforementioned communication issues.

But

– Availability of such networks varies from place to place

– Subscription to a public or private external network adds an extra recurring cost

– The equipment necessary for the use of said network increases the cost of the wireless system

Then what?

Distributed wireless networks

Distributed wireless networks use repeater nodes to solve occlusion problems. Through this message routing strategy, the distributed network can be completely independent of external public networks (such as cellular and satellite networks) and minimizes power, and therefore costs.

Typically, distributed wireless networks are based on two types of network topologies: tree topology and mesh topology.

wireless mesh network topology

Tree Networks

A wireless network is a tree if its nodes have a predefined hierarchy.

However, the main disadvantage of such a topology is the presence of multiple bottlenecks which can affect large portions of the network in case of failure.

A wireless tree topology implies a predefined hierarchy of nodes in the network. As a result, at bottleneck nodes, failure may be critical to the overall health of the network. Thus, a localized failure may affect the reliability of the system significantly.

wireless tree topology network with bottleneck

Mesh Networks

A wireless mesh has no strict hierarchy. Therefore, data is carried over large distances by splitting the distance into a series of short hops. Intermediate nodes not only boost the signal but cooperatively pass data from point A to point B by making forwarding decisions based on their knowledge of the network. In such a wireless network, traffic flows between arbitrary pairs of nodes, but in an asynchronous way: any nodes can transmit data at any time (first come-first served). Unlike tree topology, a node that fails does not cause an entire sub-section of the network to fail.

When a node goes out of service, the data takes an alternative route since the messages are relayed by flooding, without a predefined route. This architecture reproduces, in a way, the Internet model while optimizing for wireless. This solution allows fast and simplified deployments, supports scalable coverage, ensures high fault and interference tolerance, and significantly reduces network installation and operating costs. While very efficient with a limited number of peers, the system breaks down rapidly with the network size.

wireless mesh network topology
self-healing wireless mesh network topology node failure

Network Contention

When two nodes attempt to transmit data at the same time, data packets collide. As a result, typical mesh technologies will handle this eventuality by having the nodes either

– listening to the network and wait for it to quiet down before transmitting, or,

– re-transmit collided data until the transmission is successful.

smartrek technologies spidermesh cooperative mesh network logo white

SpiderMesh

Cooperative Mesh Network

Resolving Network Contention

In a wireless cooperative mesh topology, network traffic is synchronized. As a result, there is no data collision in such an architecture since every node has a precise communication slot allocated.

Consequently, a wireless sensor network (WSN) using this topology can scale up in size without the performance degradation shown in standard mesh networking strategies. In return, that makes the technology exceptionally low-power which eliminates the use of solar-panels in most applications. For this reason, SpiderMesh allows to:

  • reduce energy consumption (and eliminate the need for energy capture),

  • increase the reliability and robustness of the network,

  • self-repair communication paths in the event of network failures or breakdowns,

  • deploy a virtually unlimited number of sensors on the same coordinator (with high collision attenuation),

  • facilitate the installation and future maintenance of the network

Standard mesh vs. SpiderMesh

Mesh synchronization makes it possible to counter the well-known effect of transmission degradation with the increase in traffic on the network, and therefore with the increase in the number of nodes. Unlike standard protocols, which retransmits in the event of a packet collision, the SpiderMesh technology makes it possible to avoid these collisions, hence the achievement of unequaled performance in terms of power consumption and network size.

spidermesh comparison with other wireless network technologies and topology

SpiderMesh general specifications

Type of wireless technology FSK wireless cooperative mesh
Max. polling speed (RP-AirEVM mode) 200 ms per node
Max. polling speed (serial port mode) 100 ms per node
Min. polling speed 16+ s per node
Number of nodes on a gateway Virtually unlimited
Max hop count 32
Encryption AES-128

Further Readings

Still haven’t found what you were looking for?

Contact Us