Exploring IoT Barriers to Entry

The number of Internet of Things devices is expected to reach more than 50 billion in the next couple of years, with the perceived ability to connect virtually anything to anything else. However, potential network issues will stand in the way for some organizations.

• IPv6 continues to hamper many anything-to-anything deployments; there are just as many ways to make IPv4 and IPv6 play nicely as there are methods to disengage completely. Middleware solutions will play a role, but their effectiveness remains to be seen.
• Bigger is better seems to be the adopted approach of the HDBaseT Alliance and UL when it comes to meeting the demands of an industry that doesn't care about the size of cable bundles. One hundred watts is a lot of power to deliver, as anyone who has worked with large bundles of cabling for any length of time will tell you. The standard UL 4299 publication is expected soon -- but not soon enough for the many organizations that lack design-from-the-ground infrastructures that support the cabling with better upfront plans for reducing pre- and post-installation costs.
• Not enough power is an ongoing issue for IoT and LAN infrastructures in general. Having either access or extending Power over Ethernet to places within a building is challenging. One of my favorite things to do is to find raceways for cabling of any type -- for example, parking garages that need an antenna, point-of-sale device, or sensors, have unique challenges. Expansion joints can be a cable guy's dream come true, provided there's no significant movement and no building code concerns. Architectural plans need to better address future needs, but unfortunately the future comes pretty quickly with job change orders that drive up costs for building owners and profits for contractors reacting to initial needs not met.
• Poor coverage, either in cellular or Wi-Fi service, translates to missed opportunities. Middleware solutions, such as opening up a separate SSID on a customer's router or throwing in single-use devices to extend cellular coverage, may help but they don't really address the core issue.
• Biting the bullet, so to speak, are integrators pushing ahead by adopting Gigabit Passive Optical Network (GPON) solutions for the convergence of premises access control, security cameras, VoIP, Wi-Fi, sound, TV, building automation, and anything IoT on reliable infrastructure. My take is that providing a PON infrastructure can minimize MACD activities associated with physical infrastructure by providing more faceplate locations put in place to eliminate the need to rearrange furniture, fixtures, or anything needing a connection to the network. Why? Because the perception is that managing fiber is more difficult to than using a punchdown tool (and fiber manufacturers do need to get creative and find a way to replace their tools with a punchdown).

Of course there will always be middleware and MACDs will continue regardless of the technology platform. And problems will always be the architect's fault or the client that didn't want to entertain more costs, or someone else in some other department that failed to account for a need. Easy solutions don't always mean easy maintenance, and by that I mean you can deploy 50 billion-plus IoT devices each with their own tiny battery, but sooner or later that battery needs replacing or disconnecting.

There is no "one network," and there will probably be different layers that open up the "network" to what's appropriate and adequate. Now that's a challenge, and as you attempt to layer IoT onto your existing networks, it will be interesting to see the ratio of sensors and AI devices to IT staff, as well as the ratio of IT staff to others in the organization. In other words, work will be focused, direct, and specific.

Follow Matt Brunk on Twitter!
@telecomworx