Allocating resources effectively is an important consideration when building IoT networks because often, we will face constraints in the resources that we have to build them. In this subsection, we will explore some of the constraint types that are often encountered within IoT networks.
Constrained nodes
Constrained nodes refer to nodes that do not have features that can communicate with the internet. This is often caused by constraints such as size, weight, costs, and power to help the nodes function. Furthermore, memory may be another constraint that prevents the nodes from functioning as we would like them to. With these considerations, such nodes require a low-level understanding of how to architect networks based on them to ensure that they can still function properly within the network without any issues.
Constrained networks
Constrained networks are networks that may be limited in terms of processing power, data rates, and throughput. This may be caused by several factors, including restrictions that are imposed based on supported packet sizes. It may also lead to other limitations, including a lack of advanced Layer 3 functions, including multicasting and broadcasting, limited reachability from outside the network due to power management configurations, and packet losses that may come as part of intermittent transmissions.
Constrained devices
Constrained devices are devices that have limited resources in terms of processing power, connectivity, memory, and more. They are usually small and inexpensive and consume a low amount of power, though they come with their own set of challenges. They can be split into three different classes based on the device’s functionalities, as follows:
Class 0: The most constrained devices, these devices usually have very limited processing power, memory, and connectivity options. Even if connectivity was possible, the device usually has poor security capabilities due to its reduced capabilities. Usually, to make up for this, it must communicate with the internet through a gateway or proxy, where further encryption can be ensured. Some examples of devices within this class include sensors, actuators, and simple control devices.
Class 1: The devices in this class have slightly more capabilities compared to the devices in Class 0, with more options in terms of processing power, memory, and connectivity. They can run more complex tasks and are designed to support more advanced protocols and applications. They cannot work with a full protocol stack such as HTTP, though some specially designed protocols such as CoAP can work with it. These devices normally do not need a gateway to access the internet. Some examples of devices in this class include smart appliances, smart meters, and devices that are within the home automation network.
Class 2: Devices in this class have the same capabilities as regular portable computers such as personal digital assistants (PDAs) and laptops. They can work with full protocol stacks such as TLS, HTTP, and others. They can also have comparatively higher standards employed for security and also require a much higher power budget, as can be expected.
Now, we can take a look at the different infrastructure protocols that are based on IoT.
