IdO de type industriel
7 min readGetting IoT-ready – So, you want to get connected? Overcoming IoT connectivity & [...]
Connecting to an analytics engine, such as the SmartObjects platform, requires a cloud/server and a web portal to manage the connected products and users of these products as well as send the events to analytics. More importantly, the communication between the objects and server as well as the communication between the cloud/server and analytics platform need to be secure.
The overall solution has to have high-availability and scalability. Businesses are accustomed to managing networks of dozens, hundreds or thousands of devices. With IoT, there is a potential to increase these numbers exponentially.
As billions of objects get connected, it is evident that IPv4 addresses are running out. Businesses will need to consider IPv6 in their connectivity solution. Initiatives such as 5G are underway, causing a massive increase in bandwidth, capacity and reach, so businesses need to be prepared to change their IT infrastructure as new connectivity needs from IoT surface.
Protocols for monitoring, storing and managing connected objects will need to be re-assessed. Also, there is a big difference in bandwidth usage between deployments for consumer vs industrial applications. The consumer is paying for the bandwidth and is accountable for its usage. However, industrial IoT applications involve much larger bandwidth usage. It is in the company’s best interest to optimize the connectivity of industrial IoT devices to reduce costs. This is especially important if video and audio analytics are involved.
Ultimately, the business will need to be equipped and trained to use the analytics data to make better decisions and improve the product and services.
The number of interfaces between connected devices, databases, and networks is growing exponentially. The industry lacks standards and regulations for all IoT device manufacturers. To mitigate the exponential growth of IoT devices, there is a growing need to centralize control of IoT devices at various levels to increase convenience/interoperability and reduce development costs.
Integration is where a cause-and-effect relationship can be established between devices. Having connected devices doesn’t necessarily translate into meaningful, usable and useful interactions. Most people are not techies or hackers. The entire IoT philosophy revolves around using connected objects to increase efficiency and ‘make life easier’. If the solution is inaccessible to professionals or consumers in non-technical fields, the IoT dream falls apart. A company getting into IoT will need to consider this to remain competitive and differentiate itself. A number of connectivity platforms are now available to partially address this issue. The following are some examples from the consumer space: Openhab, InitialState, DeviceHub, Logitech Harmony Hub/Elite, AWS, IFTTT, Google Weave/Brillo, NodeRed, and Octoblu. In the industrial space, there is the PI System from OSISoft.
In short, interoperability, collaboration and integration between systems and devices within the IoT ecosystem will be a key factor in the rise of data-driven industries. There is a need to collapse the “closed” environments into a single unified global IoT ecosystem with a uniform data model. In the meantime, we will need to resort to “hub/bridge” devices and platforms to solve the “Tower of Babel” of platforms and protocols.
Privacy and security are growing concerns for consumers and corporations. Connected objects allow businesses to collect massive amount of data, which can be used for their own internal purposes or can potentially be sold to outside parties.
As stories of hacking and cyberattacks pop up in the media, people are becoming more sensitive to these issues. For some, IoT represents a level of connectedness they are not yet comfortable embracing. Highly sensitive data will be subject to more strict privacy and security requirements. The aggregation and correlation of multiple data points can quickly reveal rich private/group information that can be accessible to hackers. As more and more devices become connected to the Internet every day, the security threats will expand to new areas and industries. Security is only as good as the weakest link!
A significant emphasis on the ownership and use of data will be required to guarantee privacy. It needs to be balanced with policies that will allow users to determine how their personal data can be used and shared. Without these managed controls, the user could potentially lose trust in IoT.
A company embracing IoT will likely consider the associated risks and impact on their customer base due to perceived security issues. The risk for smaller businesses will be to rush a connectivity solution onto the market based on costs and other constraints to be ahead of competitors without carefully implementing the necessary security measures. These connected devices will potentially run unprotected for many years as businesses focus on newer products. Advanced analytics enable real-time threat protection and intelligence (real-time behavioural analytics pinpointing suspicious activities at the users, devices and events levels through anomaly detection, analytics of known security issues, reducing false positives through aggregation/correlation, etc.) .
Also, the collection of sensor data exposes legal and regulatory challenges related to data protection and privacy. Legal issues can include cross-border IoT data flow, legal liability for security breaches, and data retention/destruction. IoT businesses will need to equip their legal department to face these challenges.
There are a number of considerations when implementing IoT:
Before a functioning IoT ecosystem can develop, the price of connectivity modules will need to decrease considerably in consumer and industrial spaces. In the consumer space, less than $2 per module will become a necessity to make connected devices affordable for large-scale deployments. The average IoT device often costs five to ten times more than its non-connected counterpart due in part to the connectivity stack.
Cost reduction is also required for sensors, actuators, and battery systems so that IoT can truly proliferate. Manufacturers of sensors are making an effort in this direction with high levels of integration and miniaturization (e.g. SoC, MEMS).
Many businesses have yet to establish what benefits IoT can deliver. They haven’t devoted the time to develop ideas on how to capitalize on IoT for their own business. It is a tricky exercise to clearly predict ROI and margins as businesses look at varying factors that change over time in this highly disruptive IoT world (varying cloud vendor pricing in terms of the object count, messaging rate, etc.). Just because an object can be connected doesn’t mean it needs to be part of the IoT ecosystem. It will have to ultimately make practical sense to the business and consumers. In many cases, it will be: “we can live without it.” Both technical and business/marketing innovation will be required to promote an IoT product.
But price does not dictate everything! Businesses should consider the tangible and intangible benefits of the entire end-to-end solution. The hardware is often a smaller consideration of the overall solution. One should consider how an “ecosystem” can be built around the product (think of Apple’s appstore which increases the perceived value of Apple’s phones and tablets). Businesses need to realize the “opportunity” in the sensor data.
Without an actual Internet connection, IoT cannot exist. There are still many places in the world that have no Internet connection. Even countries that do have high connectivity to the Internet will often have dead spots. Worldwide Internet coverage needs to happen for IoT to become a fully effective reality.
It is a lot cheaper and easier to introduce sensors in a new consumer product. It’s a lot more expensive to add sensors to places/structures that are already widely deployed throughout the world (roads, traffic lights, utility grids, etc.). It’s a large undertaking that makes the ROI not very obvious. The IoT connectivity infrastructure in these cases will need to be planned carefully.
To find out how mnubo can help with your IoT data strategy, request a free consultation today!