Skip to main content

Evaluating IoT Device Constraints

Published Date
Author Cadence PCB Solutions

Key Takeaways

  • Power consumption and its interplay with portability in IoT design.

  • Additional factors that influence the performance of an IoT system.

  • How to evaluate IoT along a simple three-branch approach.

IoT devices forming the shape of a cloud.

As IoT grows, designers must be cognizant of IoT device constraints to optimize their designs.

Perhaps one of the most frustrating experiences in the modern world is taking an electronic device home only to spend fruitless hours attempting to work it into an existing network. The sheer functionality of many devices unfortunately means that there are many possible failure points within a design. Of course, we all want the latest-and-greatest products and performance, so it is an unavoidable catch-22 in many ways.

Taking this idea a step further, IoT device constraints need to be properly accounted for before system integration. There are simply too many variables that a team can choose to emphasize or prioritize, and in doing so, can squeeze the best performance out of an IoT system.

Many Factors Influence IoT Device Constraints

In most IoT applications, the largest constraint in design is power consumption. Sensors need to constantly monitor system states and update the network, with the added wrinkle that devices may be distributed over a wide area with limited access to a continuous power supply. On-board power has the challenge of balancing portability against the total capacity of the supply, and network transactions that supply power between devices require extreme proximity. Conversely, a network may be distributed such that devices have access to regular power, but at a greater distance, increasing the response time between devices on the network.

There‘s an additional host of factors to consider shaping an IoT network, and much like board design, there is no such thing as a free lunch when it comes to performance. Some of the major influences on the network include:

  • Networking - Connections need to be established before transmission, which due to the often trivial length of the latter, means considerable energy expenditures spent on setup versus the actual content of the data. Standard TCP congestion control and buffering are also similarly inefficient; though in fairness, these are not issues exclusive to TCP, and solutions geared towards the specific needs of IoT have been developed. 
  • Congestion  - Fail safes need to be in place to prevent overwhelming systems, especially at points that route a large number of signals. Protocols with increased bandwidth can be an option, generally with higher associated power consumption, but systems can also prevent network congestion by designing built-in throttling during periods of high traffic.
  • Hardware - The major cost driver is components alongside labor, and designers will want to ensure that cost-per-unit is kept to a minimum due to the high number of IoT devices that may be present throughout a system. RFID can be a successful implementation of IoT-aligned technology due to a few key aspects such as low power consumption, large broadcast area, and low manufacturing costs.
  • Interoperability - Navigating through multiple different devices and protocols becomes a challenge, especially when attempting to optimize data send/receive for the lowest energy cost per transmission. Specifically, the focus for IoT is a shift to machine-to-machine communication (M2M), which attempts to facilitate communication by either including higher levels of control abstraction or defining the actionable elements of the circuit.
  • Security and privacy - Security is an increasing concern within the realm of IoT: as connectedness continues to climb, so too does the possibility of a data breach. Not only do designers need to be aware of how a system prevents unintentional access to the network, but long-term storage of data gathered on users is increasingly being stored indefinitely as the value of the information supersedes that of the long-term storage cost. To that end, system designers need to be mindful of how their strategy for data storage (especially sensitive information) provides confidence to users.

Analyzing Systems By Constraints

With the multitude of design constraints to weigh, designers need to be able to focus on some key metrics that provide a tidy wrap-up of the system. While configurations and architectures can come in a dizzying amount of forms, all can be evaluated at the same fundamental level:

  • Throughput - Networks can be set up for continuous connectivity or discrete events based on a timer or some other controlling variable. The more data flowing into the system (and likewise flowing out to the periphery), the greater the bandwidth needs, at least at the system chokepoints. Depending on the architecture, this can ultimately affect everything from the total range of the system to its ability to be built on existing mobile network infrastructure.
  • Efficiency - Sleep periods influence the polling rate and can be pushed to extreme ends when efficiency is paramount. In doing so, networks become less flexible and responsive, but not every IoT application will require updates at a high frequency. 
  • Delay - For transmissions, the larger the amount of data sent in a single shot, the more “data-rich” each transmission is due to the necessary protocol features like three-way handshakes for authentication and the like. In general, the more that can be reasonably fit into every transmission, the less delay for total data content. However, the sampling needs of the system may override this, depending on the exact usage of the IoT network.

Dovetailing IoT and PCB Design

IoT device constraints need not impede the overall functionality of a circuit once properly accounted for in the design. In fact, by understanding some of the drawbacks of different IoT architectures and protocols when building out a system, designers can better leverage the positives that IoT offers and adapt to optimize network performance. Whether that’s speed, reliability, power efficiency of the network, or more, system designers have the option to tweak the system to focus its efficiencies in the areas deemed most important for its particular function.

By the same token, device constraint extends to the designer just as well. Small product sizes mean less board space for designers and a direct challenge to some major board characteristics like signal integrity and good heat dissipation. While most aspects of IoT board design adhere to basic best practices, having to balance overall performance against efficiency means making some strategic tradeoffs at points.

Luckily, Cadence offers PCB design and analysis software meant to simplify the complexity of simulation without sacrificing robustness. Similarly, designers can use the OrCAD PCB Designer as part of their integrated hub to help reduce turnaround times while maximizing error-checking prior to entering production.

Leading electronics providers rely on Cadence products to optimize power, space, and energy needs for a wide variety of market applications. To learn more about our innovative solutions, talk to our team of experts or subscribe to our YouTube channel.