Introduction to the Internet of Things
Introduction
For a very long time, there has been a necessity for people to have the possibility to monitor places of value, like a store or at home. For this case, in the old days, a system of cameras and a recording system where turn on manually when necessary. This will record until someone turns it off, the power is cut off or the recording media it’s full.
This kind of set up its very limited, because if you are outside of the place in question, you cannot see in real time what it’s happening, you are recording everything, whether its useful or not and if you are not careful you won’t record something of importance because you run out of storage.
We can summarized the main issues of this approach in the following:
- It lacks certain intelligence in determined when to record, that’s because there isn’t a sensor that detects movements.
- The system it’s not very efficient. In order to been able to record, you have to manually start the system and will record everything until its turn off.
- The set up offers a corrective solution. You always find out after the fact that something wrong happens, there it’s no way for the user to know in real time what is happening.
The capabilities and range of sensors to make equipment more intelligent and the advent of the internet bring the possibility of being able to connect remotely to the device and also to being able to use different applications to control them. This brings us to the era of the internet of things (IoT)
IoT is about connecting all those devices that previously were isolated, being able to collect and send data to make them better and more useful for people and enterprises.
Two cases of IoT to consider
- Factories
Like in the case of surveillance of home and offices, there is a great need for people in all levels to see things at any given moment in real time, otherwise they are just flying blind about the real state of the company, whether is about info for supply change, production lines, efficiency, etc.
The case of factories¹ its particular in its complexity, as it involves different areas that are not connected in real time with their process and between them. A few of this areas in a common set up are as follows:
· Maintenance: This area is in charge of keeping the all the machines and production lines in proper condition with minimum downtime. Normally if there is a problem somewhere, the workers of the affected area need to find the maintenance people for help, set up the revision and start to look up what the problem is. This takes time to solve and organize, and also there are other areas that would not be immediately aware of the problem.
· Production: In charge of the production lines, people on the ground need to visit constantly to the control room to check if something it’s not right, and they never know it in real time.
· Purchases: How better way to handle the supply chain than to know in real time the needs of materials and spare parts. The reality it’s that there exits layers of the communication process that delay this process to be completed.
· Management: What are the efficiency at any given time at the production lines, what are the downtimes because of materials or maintenance, how the quality of the product deliver, what is the client satisfaction? All those things can be discuss during meetings, but are not constantly check up in the present time.
How can internet of things help in some of this cases? If the production lines had sensor to check product quality and machine performance and availability, they can inform all the interested parties in real time what is happening. If the data send by the machines is send to the net, it can be available via mobile devices, so that people looking at stuff at the ground floor don’t need to check in the control room what the current status is.
Maintenance could be made aware immediately in the case of issues and can understand better beforehand what the problem could be. The supply chain can have more time to make adjustments on different requirements and Management is better inform to make decisions based on the present data.
Some of the steps taken in the development of automatization leading to IoT in industries it’s illustrated in the following image¹.
- Automatic driving vehicules²
With the success of tesla and their introduction of self-driving technology, other traditional automaker had been push to develop equivalent systems. But why exactly self-driving cars? Accidents mostly happened due to human error, like looking at your smartphone, dangerous driving, lack of driving skills or falling sleep on the wheel.
Ideally, self-driving cars would eliminate this dangerous scenarios by taking control of the car using and array of sensor s that feeds the internal car computer about the road, vehicles and other obstacles. But what are the current situation and what dangers are present in this technology?
Tesla autopilot, the most known system and despite the label, it’s far from perfect, as there are many news concerning people being killed by abusing this technology, because they are assuming that they actually have an autopilot feature³. In the present there isn’t any available technology in production cars that are 100% “self-drive” and yes, the autopilot feature could be better that falling asleep while you are driving, but thinking that that is the case is very dangerous as the experience has shown, also, there is the possibility of different kinds of driving in the future. Like in the movie I Robot (2004), what if there are a significant difference between the driving capabilities between self-driving cars and human drivers, because many people won’t want to or can’t use cars with an autopilot like option.
The autopilot is not the only option that exemplifies the concept of IoT to make cars better in theory, others devices and capabilities are shown in the next picture¹:
What about security? With a car like the tesla that is constantly connected to the internet and so many features to open and move the car remotely, are they vulnerable to hacking? Yes they are, as there were even an instance where someone was able to take command of an entire fleet of teslas⁴, so the possibility of this happening again in the future its always possible, as some of the challenges of IoT shows in the following section.
Challenges of IoT¹
- Security: The same possibilities that the internet brings in terms of functionality also brings dangers, such as hacking. Before gadgets work in close systems, so there were very difficult if not impossible to manipulate remotely. Now the connection to the cloud made this possibility very real.
The jeep hack⁵ is particular notorious and probably the most known. In 2015 two security researchers demonstrated that they can take control remotely of a jeep because of security problems with Chrysler’s Uconnect system, and cause the calling of over 1 million vehicles. The researches found that a port was accessible and happened to be used to communicated between processes. On this port a service call NavTrailService had an execute method and allow to run code on the device. The code used to open a remote root shell on the head unit it shown next⁶:
This code allow to control the steering, the brakes, heading lights and other functions of the car.
- Privacy: If a device is constantly sending data to the cloud, how the user know for sure what is being send. Much has been talk about how Amazon Alexa may be sending stuff its nor ordered to do by error or by hacking. There is also the possibility that companies used this connection for their own advantage, like the possibility that amazon uses Alexa to determined preferences to of users to adjusts their marketing.
- Storage: to make things intelligent, you need to use sensors to tell them what to do, and that means a lot of data to handle. Yes, you can use cloud storage and tools for data analysis, which is a superior option than a VHS cassette or a CD, but if the idea it’s to have smart devices in every home, that translate in massive amounts of data that has to be manage.
- Interconnection: Luke Millanta demonstrated that you can control a tesla using an Amazon Echo. This brings an interesting point. It would be ideal if you can connect devices with different functionality depending of what the user may want. The problem it’s the different standard that each manufacturer have. Some may be open but others are not, so there isn’t a seamless way to do it at the present time. For this to work it would be necessary the creation of standards that can be follow, but considering the investments in IA and sensor technology, it could be difficult for cooperation to exist between rival companies.
- Politics⁷: The IoT concept rest on the idea of an open internet, in other words, access its independent where the device and the user are located. Unfortunately, cyberattacks to government infrastructure and operations and the black listing of technology companies because of ties to their home country have made the possibility in the future of fragmented access around the world, where countries create regional walls to prevent access from other regions. This would create problems in the availability and development of IoT.
[1] Hanes, D., IoT Fundamentals: Networking Technologies, Protocols, and Use cases for the internet of Things. Edited By Mark Taub. Cisco Press. 2017
[2] Impacs.io https://impacx.io/blog/autonomous-cars/
[3] bbc.com https://www.bbc.com/news/technology-51645566
[4]electrek.co https://electrek.co/2020/08/27/tesla-hack-control-over-entire-fleet/
[5]Gupta, A., The IoT hacker’s handbook: A practical guide to hacking the internet of things. Apress. 2019
[6]Miller, C., Valasek, C., Remote Exploitation of an unaltered passenger vehicle. 2015. http://illmatics.com/Remote%20Car%20Hacking.pdf
[7] Roe, D., 7 Big problems with the internet of things. cmswire, 2018. https://www.cmswire.com/cms/internet-of-things/7-big-problems-with-the-internet-of-things-024571.php
