The Expanding Internet of Things Presents Major Challenges for Telecom Companies

How does the Internet of Things (IoT) affect the telecommunications industry? It first appeared on Quora, a place to acquire and share knowledge, enabling people to learn from others and better understand the world.

Answer: Shane Greenstein, Quora:

Regarding the impact of IoT on the telecom industry, you must understand what we can do today to comprehend what we need to do tomorrow.

First, we look to 2020. Even in the short term, we should expect new applications to put pressure on scale, reliability, and security. This will enable telecom operators and equipment suppliers to meet these demands.

The vast majority of IoT applications (so far) involve low‑bandwidth, low‑power, widely distributed networks with minimal reliability requirements, typically industrial use cases. For example, logistics companies commonly record the progress of packages along routes. They do this by deploying cheap readers that send simple signals along the path, automatically recorded by a central server, alerting operators to simple hazard warnings. Redundancy is built in so if one fails another can replace it. This is common in freight, rail, warehousing, as well as oil and gas pipelines, and in large logistics businesses of retailers like Walmart, Amazon, and Procter & Gamble. Such applications can be deployed at scale across large geographic distances with many distributed sensors.

This also tells you the challenges you will face and how they will affect the telecom industry. Acquiring and managing a fairly large and geographically dispersed IP network is not beyond human capability today, as long as it has some fault‑tolerance, if any. But…

Reliability: While 100% reliability is unattainable, 99% reliability is acceptable most of the time, but in many applications 99% is terrible. Designing modern wireless IoT devices like pacemakers or blood pressure monitors is easy, but no one expects 99% reception in every location and scenario. This blocks many valuable applications unless companies find solutions.

Security: Loose security may be fine for blades shipped across the country, but unacceptable for autonomous vehicles. If you disagree, I can tell you about rogue programmers waiting to ransom your car. Again, this will hinder many valuable applications.

Privacy protection: Privacy rules are hardly an issue for temperature sensors on oil pipelines, but problematic for devices reading activity at home. For example, a recent vibrator company got into trouble for sending too much data back to the product design company. Interesting, but a symptom of a larger problem.

System design: Real‑time integration of massive data is feasible, but these applications have many potential failure points. After all, the integration of Waze with Google Maps has already merged large amounts of low‑bandwidth signals into many accessible apps. Inaccuracies can be irritating, but if the network misbehaves you can survive. However, there are pitfalls. No one suggests integrating aircraft traffic with digital weather data and having computers route and land everything. We still need humans because systems are flawed. Certainly we can improve some things.

It is clear that these frontier issues—reliability, privacy, security, and system design—require massive engineering effort in networks, devices, and software. However, they are not solved by engineering alone; legal and managerial aspects play important roles. Many frontier applications need inter‑company cooperation—standard protocol agreements, legal norms for fault responsibility, and system designs that allocate liability boundaries to avoid cascading failures.

Back to the main question: how will the future of IoT affect telecom companies? They need teams of lawyers, engineers, and MBAs to design processes, platforms, and protocols that allow applications to cross company boundaries. This is a challenge for organizations of any size and will certainly pose challenges for telecom firms. Large companies will continue to offer turn‑key solutions, and buyers will resist “integrated” solutions, demanding a mix of best‑of‑breed suppliers.

It sounds familiar; technology has lived with these challenges for decades, and we are doing it again.

If you want to learn more, checking specific applications might be very interesting.