Choosing the most suitable network layout is critical to the efficient operation of any system. For wireless networks, this is highly relevant, as it is closely tied to factors such as latency, power, speed, and redundancy.
A modern network is made up of many different types of devices, such as routers, smartphones, Bluetooth headsets, and smart light bulbs, and every device on the network is called a “node.”
Network topology describes how the different nodes in a network connect to and communicate with one another.
Simple Network Topologies
The industry has many kinds of network topologies, each with its own advantages and disadvantages.
The simplest network is point-to-point. This is a single network connection between two nodes. In practice, apart from certain critical or specialized applications, there is rarely a need today to connect just two nodes. A modern example is an ad-hoc connection over Bluetooth, used for Apple’s AirDrop file-sharing service or for remote control of things like model aircraft. The simplest way to add more nodes to a network is a daisy chain, which comes in two forms: linear and ring. A linear daisy chain connects a third node to one of the existing nodes, a fourth node to the third, and so on. It is a simple approach, but it quickly becomes impractical as more nodes are added. In a ring network, the third node connects to both existing nodes at once, and each additional node is inserted between two existing nodes to create a loop. Every node connects to exactly two other nodes. Data travels around the ring in one direction or both, and each node examines the data and acts on it, or retransmits it until it reaches its destination. Daisy chains can be used to create networks of connected devices (such as smart light bulbs), but for most use cases there are better solutions.
Solutions for Scaling Up
Once we add a large number of nodes to a network, point-to-point and daisy chains become inefficient. If two nodes on opposite sides of a 1,000-node ring want to communicate, their data must first pass through 500 other nodes. Daisy chains are also prone to errors, since a single failed node can cause widespread disruption and, in the worst case, prevent messages from getting through at all. In a bus topology, we connect all devices to a central backbone (called the bus), which is similar to a linear topology but has a single line with multiple individual branches. This is a common topology in wired networks, but there is no real wireless equivalent. While understanding these topologies is important, none of them are common in modern wireless ecosystems. We will now look at the more common architectures.
Star Topology in Wireless Networks
The most useful topology for wireless networks today is the star topology, in which every node connects to a central hub that distributes data wherever it needs to go. The most obvious example is a home network, where all nodes (phones, printers, tablets, and so on) connect to a wireless access point (the hub), which is usually both the router for the local network and the bridge to the internet. Star networks are well suited to seamlessly connecting wired and wireless nodes at the same time. It is a simple way to implement a wireless network, but it has two main drawbacks. The hub at the center of the system is a single point of failure. If the hub fails, the entire network ceases to exist, leaving devices unable to communicate at all. In addition, every device in a star topology must be within a certain distance of the hub. To add resilience, and communication range, we need smarter ideas.
Mesh Networks
There are two types of mesh networks, full and partial. In a full mesh network, every node connects directly to every other node, which helps with resilience but does not increase communication range. These networks are implemented only in applications such as military networks, where 100% redundancy is essential. On the other hand, in a partial mesh network, each node connects to one or more other nodes. Connecting to multiple other nodes increases resilience, since there is no single point of transmission failure. At the same time, it also improves communication range, because node A may not be within direct communication range of node C, yet messages can still be relayed through node B. Mesh networks are the biggest area of growth in the Internet of Things, and in theory they allow an unlimited number of devices to connect across unlimited distances. With mesh networks, resilient smart homes, smart businesses, and smart cities become a reality.
Hybrid Networks
Today, examples of networks that use a single topology are in fact increasingly rare. Most networks are hybrids that combine one or more different topologies. A tree network connects multiple star networks together over a single bus network. This approach is commonly used when there are multiple clusters of nodes, such as in WANs. In this case, each node connects to a router to form a star network, and the routers are then connected to one another, creating a bus connection between them. This can also be done with a snowflake network, which connects multiple star networks to a single central node, like a star of stars. The internet is the ultimate hybrid network: it comprises many individual networks of every type, with bus, star, ring, and mesh networks all combined. Standardization — The key to increasing mesh network adoption is standardization. Because vendors in the mesh space are eager to profit from industry demand for mesh networking, multiple standards typically emerge. While we may see a de facto standard appear in the market, many different standards will coexist in this segment. A smart home today might have mesh light bulbs from one vendor and a mesh heating control network from another, with both networks connected to a conventional star network that lets users control them from a smartphone. Clearly, this approach is inefficient and, in the long run, benefits no one. The industry’s main goal is to establish a standard mesh network that all nodes — whether smartphones, laptops, bulbs, or sensors — can join. There is still some way to go before that goal is achieved, but once it is realized, the Internet of Things will be able to reach its true potential.
