The true value of IoT lies in its ability to revolutionize user experiences, whether that is in your home, in the office, within your car or even as you walk down a sidewalk. Embedded sensors are generating the data that is powering a new revolution for business and their consumers — the power of customer experiences, or CX. As various industries face commoditization on a grander scale than ever before, CX has become the prime differentiator in driving differentiation and consumer purchases. Modern businesses must reprioritize around CX to optimize on costs and provide better user experiences, ultimately driving customer and brand loyalty.
Driving these deeper customer engagements will mean using technology to provide the data-driven capabilities that can revolutionize CX. This is where artificial intelligence comes in. Simply put, AI is the simulation of human intelligence processes by machines. This opens a plethora of new possibilities for enterprises, as AI is both scalable and efficient and can enable a business to automate repetitive processes that may be extremely time-consuming for a human employee to perform. By combining the extensive data capabilities of IoT with the data processing capabilities of AI, a business can truly understand the individual customers and craft a predictive customer experience, as opposed to traditionally reactive customer experiences.
These AI-driven experiences take shape in a variety of different ways. For example, AI in CX can come in the form of a chatbot, intelligent voice-operated system or even in AI-driven applications that can provide translation services and pertinent data directly to the customer. Other ways of using AI in CX include the transformation of more back-end services, such as using machine learning to process and customize data traveling to and from an enterprise and its customers.
How AI can transform CX today
AI is not a single technology, but rather a class of different capabilities that can be applied to many different functions and contexts. Here are a few use cases where AI can feed into your CX strategy:
- AI in natural language processing (NLP) based on speech recognition/synthesis. Natural language processing uses AI to “understand” speech requests using a combination of speech recognition and speech synthesis. Specifically, speech recognition technology is used to understand what the person is saying and speech synthesis is used to formulate a response. NLP can transform CX via the use of automated customer service and personal assistants. For example, enterprises can automate the customer service process to help customers book flights or troubleshoot problems by speaking with an AI assistant as if was another human. Apple’s Siri, Amazon’s Alexa and Google Assistant are common examples of personal assistants that use NLP.
- Using AI and machine learning to customize and predict outcomes. Machine learning can be used to train a system to handle requests around a variety of functions. Once trained, a machine learning-enabled system can ultimately be used to understand needs of the customer, customize interactions with customers or predict specific outcomes based on a set of defined events. For example, online music listening patterns and customer-provided data can be processed to curate recommendations of new music that is customized per user, resulting in a more predictive approach to CX. In an IoT scenario, connected device usage patterns and sensor data in manufacturing can be used to predict maintenance before a car or machine breaks down, enabling an enterprise to provide differentiated customer service.
- AI in image recognition. AI can process images to detect specific objects in an environment and enable other interactions. For example, a retailer can use AI via in-store cameras to analyze queues and the number of shoppers in the store, enabling the retailer to reduce checkout times by automatically summoning cashiers to help with checkout. Additionally, retailers can use in-store images to analyze gender, age, body type, style and other attributes in order to make personalized wardrobe recommendations.
Although AI isn’t the Holy Grail in remediating every problem faced by today’s enterprises, the AI technologies available today can prove overwhelmingly effective when used to transform CX and may be easier to implement that you think. Specifically, enterprises can begin infusing AI capabilities in existing applications that power customer service on mobile and IoT devices as well as via web applications.
Adding AI into your applications via APIs
Infusing AI into your existing applications isn’t as difficult as it sounds. By using application programming interfaces (APIs), enterprise and application development professionals can introduce AI capabilities into their apps without needing an AI-dedicated engineer or data scientist to manage the actual machine learning and data training process.
Major enterprise cloud vendors, such as Microsoft, IBM and Google, as well as a number of emerging vendors, are already offering a rich set of AI APIs that can be easily accessed from the cloud, enabling enterprises to easily integrate these APIs into existing applications and add AI functionality in use cases such as vision, speech, language and conversational assistants. For developers of IoT, mobile and cloud applications, all that is needed to do is “call” the API within an application to utilize these functionalities. For example, a natural language processing API can be used to automate actions based on certain written requests or to process and react to insights around historic customer interactions.
APIs provide a simple and easy-to-scale approach to integrating AI in your IoT applications and beyond. As today’s enterprises are facing a flood of new data from sources like mobile and IoT devices, infusing AI APIs in your applications may be key to ensuring unique and customer-centric experiences.
What’s next for AI in CX
Although AI is on its way toward being extensively used in the enterprise, customers of tech-savvy companies are already benefitting from AI use cases through the implementation of developer-friendly tools such as APIs. For example, we are already seeing AI-driven technology via chatbots, which have quickly become a staple of customer service in online shopping.
With the right approach and training, AI will continue to improve its ability to intelligently process data and drive automation. This, in turn, will help improve the state of CX as systems learn and adapt, enabling more predictive, personalized and timely customer service. To keep up with our increasingly digital world, it will be essential to consider how technologies like AI will help drive your CX strategy to differentiate and innovate in a competitive market.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.
A recent slew of natural disasters has brought city resiliency to the forefront of all local government leaders’ minds. Each city faces unique potential threats, from hurricanes to tornados to fires, and is tasked with ensuring their city’s individual protection needs are met. However, all cities can benefit from preparing for expected and unexpected threats in advance to ensure damages are as minimal as possible.
A strong step toward being prepared is to integrate smart technologies into existing infrastructure, which can help cities achieve resiliency as well as instill a feeling of comfort and safety for its occupants. To start, city leaders should pinpoint their unique challenges and goals, and from there determine which intelligent infrastructure or technologies to integrate.
By establishing goals or desired outcomes in advance, cities can better narrow down and focus their smart city plan. Typically, the main goal for most cities is to integrate technologies that will help them thrive and succeed for years to come. A successful strategy is one that can take advantage of connected technologies, which can positively impact public safety, economic development, traffic and aging infrastructure challenges. It can also help city leaders better manage energy and maintenance costs, reduce environmental impact, enhance resident and visitor comfort and safety, and increase building values.
The biggest challenge that many cities historically face is working with existing infrastructure, which plays a role in which connected technologies can be integrated. Building upon that existing infrastructure is often times the best option given costs associated with replacing entire systems or starting from scratch. For this approach, city leaders must identify the structures and technologies that can be repurposed and enhanced to create a smarter network.
Safety and security
Public safety is always a top priority for city leaders as they continually work to provide a safe place for citizens to live, work and play. Security is part of the city’s overall brand as it impacts all aspects of a municipality from airports to water, wastewater and lighting. When the right assets are in place, it can help attract and retain visitors, residents and businesses. Incorporating technology, such as video surveillance, in a smart city infrastructure can help deter crime, improve response time and improve overall city operations before and after an emergency.
Ensuring a city is not shut down entirely during a natural disaster is crucial. Distributed energy storage technology can help a city stay up and running regardless of a power outage as it releases stored energy to buildings and city residents. Having distributed energy storage in place in advance can lessen the impact of a disaster, as well as save the city money. This technology is extremely important when considering future infrastructure investments for a smart city.
While they may seem rudimentary, streetlights can play an integral role in smart city resiliency. While we know they are important to road safety and, of course, lighting the way for pedestrians, today they are also being used across cities for safety and operational efficiency. This existing infrastructure allows for smart improvements, such as networked LED streetlights, to be easily added and more cost-effective. Ordinary streetlights can be transformed into vertical assets through the addition of cameras, sensors and gunshot detection devices. During a hurricane or forest fire, for example, using traffic sensors and climate detection, smart streetlights can light a path for residents that is both uncongested and safe to pass from flooding or poor air quality. This everyday asset with added intelligence can now be one of the main components that helps improve the speed and safety of evacuations during an emergency.
Water leaks during or after a disaster can put residents’ water at risk and waste the city’s time and money. Implementing an advanced meter infrastructure (AMI) can help monitor for electric meter anomalies, allowing city officials to handle any issues swiftly. An AMI network sends real-time alerts the minute it senses an issue, whether that be a power outage or an unusual change in temperature that could be caused, for example, by a short in the meter which could pose a risk of fire. As with lighting and energy, improvements to water infrastructure can be critical to lessen both impact and downtime following a disaster.
Smart cities are the future, making it even more important for local governments to identify what is possible in their own cities. Implementing these smart technologies for a more comprehensive and holistic network is a step towards a more resilient future. Now is the time to identify goals and challenges and make a plan in advance of emergencies. Resiliency is achievable through smart cities, it’s just a matter of taking advantage of the technologies at hand.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.
Much ado has been made about self-driving cars, and the spirited debate as to how soon humans will be banned from driving. As a human who loves to drive, I hope that’s never the case. For now, I try not to worry too much.
That being said, there is a rising concern among many drivers about the vast amount of data that today’s sensor-laden cars can track. This New York Times article from last summer nails it:
Cars have become rolling listening posts. They can track phone calls and texts, log queries to websites, record what radio stations you listen to — even tell you when you are breaking the law by exceeding the speed limit.
Consumers are aware that many insurance providers offer a discount to drivers who agree to have their habits tracked with telematics devices; these are primitive compared to the sophisticated sensors already riding around in production vehicles. However, while privacy concerns dominate the current zeitgeist around automotive IoT, as a data scientist I’m fascinated how much we can learn about making cars (and drivers) safer by diving into the massive amounts of data that onboard sensors generate.
Profiles driven by behavioral analytics
For most non-data scientists, the term “behavioral analytics for the internet of things” might connote analyzing the behavior of specific IoT devices, themselves. In fact, behavioral analytics, from a pure data science point of view, helps us to utilize the IoT device as a sensor and proxy for the entity associated with the device:
- What an individual person or device does, and
- What they don’t do, but might in the future.
The first comparison is of the customer or device in the context of their own history of events, where one can determine changes from historical behaviors. A second comparison is done by grouping customers or devices into similar archetypes of behavior, and then analyzing how much the behaviors of individuals deviate from the archetyped behavior of similar individuals.
In the case of an IoT device like a car, depending on the degree of variance, we can assess how likely the behavior of the vehicle is to be aberrant, which can indicate if the driver is behaving in a way that is safe (at the speed limit and obeying other rules of the road) or unsafe (driving too fast, erratically or under the influence of alcohol or drugs).
The IoT devices can indicate whether a driver who generally drives defensively becomes aggressive, or an aggressive driver who is operating within the norm of their more speedy tendencies. IoT data can also extend to find anomalies not related to the driver, such as telling how us likely that vehicle is to have been compromised by a cyberattacker, or suffer from component failure.
Sorting data and building profiles
To achieve this we utilize the concept of behavior sorted list technology and collaborative profiling. Behavior sorted lists keep track of events that are recurring in a driver’s behavior, and rank orders these events to understand normalcy based on historical data.
For example, if one associated a set of events say related to:
- -10_in_30MPH ZONE
- +15_in_30MPH_ZONE, etc.
We can understand based on past occurrences whether the driver typically speeds or is careful in a pedestrian zone. Although dangerous, consistently being +15 in a 30 mph zone indicates a past behavior of speeding in this zone and not a change of behavior due to, say, being under the influence or late for work. Both of these situations might mean the driver is less adept to responding to the unique dangers of speeding in these zones.
When using collaborative profiling technology, we look to new rare events that might indicate more or less risk based on similar drivers. This technology works by associating words associated with different driving events — such as high-g turns, rapid acceleration, rapid deceleration and high speed alerts –which might learn that a driver doesn’t have a history of these events. Perhaps through use of collaborative filtering, the driver’s profile is more similar to a new, inexperienced driver, and one who might have a high risk of making such moves. Such activities might be indicative of a teenage driver getting her permit and then going for a joy ride, without yet having acquired sufficient driving experience.
Applying mature technology in new ways
Behavioral analytics, which encompasses behavior sorted lists and collaborative profiling, is a mature technology. FICO introduced it in the early 1990s to fight credit card fraud; we currently analyze two-thirds of the world’s payment card transactions, in real time, for fraud.
Behavioral analytics technology allows us to flag potentially fraudulent transactions with pinpoint accuracy to identify, literally, the needles in the haystack. In addition, behavioral analytics are continuously enhanced with the latest in artificial intelligence and machine learning technologies.
In terms of automotive IoT, techniques such as behavior sorted lists and collaborative profiles allow us to both deeply profile and understand a history of drives for a driver, while also anticipating new driving activity as potentially more or less characteristic and risky. Knowing that a driver has a lead foot off the line provides a view of normalcy of this driver and how they use their accelerator. Knowing that a driver doesn’t have much experience with high-speed driving or braking might indicate someone who is soon going to meet the ditch, or worse, in terms of these driving events.
This raises the question of how AI can help in a number of ways, for example:
- Potentially anticipating this risk of driver limitation as an opportunity to intervene based on driver experience and history with the car and past drives.
- Allowing the car more stopping distance or limiting speed when deemed beyond the driver’s experience.
Thus, it’s pretty easy to imagine driving scores that would rank order drivers based on skill and experience levels, to potentially “unlock” the most aggressive behaviors of the automobile when appropriate, or sideline a driver when his behaviors are indicative of a distracted driver reading texts, under the influence of drugs or alcohol, and other unsafe conditions.
Should cars be ‘trained?’
So, let’s say you drive your car to a bar and have a bit too much to drink. Can, and should, cars be “trained” to react to their drivers’ profiles, such as kicking into self-driving mode and pulling over to the shoulder if the car senses you’re driving while intoxicated? That’s a big question. Drinking, driving and safety are already top-of-mind topics for today’s lawmakers. Recently, at the 50th annual meeting of the Governors Highway Safety Association, autonomous vehicle technology was a key topic of discussion.
As we wait for the dust to settle on how autonomous driving and IoT technology will affect laws around drunk driving, seat belts and open alcohol containers, keep up with my latest thoughts on Twitter @ScottZoldi. Vroom!
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.
Last year, Nintendo brought out a new console. My sons and I are fond of gaming, so we had to get this new console — the Switch. If I am not writing blogs, articles, books or doing any other work, I will be gaming.
The great thing about consoles nowadays is that they are crammed full of sensors and actors that enhance the gaming experience. In-game effects are enhanced and gameplay is expanded into the physical realm. Compared with the old consoles I started on in the ’80s, we now have infinitely better controllers; the controllers of yore are nothing compared with the things we now hold in our hands. The controllers of today are almost consoles themselves!
From infinite gaming to infinite IoT
Physical controllers are one thing. But it is now possible to control games with voice recognition, gestures or running around through your neighborhood. There are infinite possibilities of controlling games even within one type of control. Gaming connectivity, collection of data and the overall gaming experience is an IoT solution.
Let’s take a step into the development world of the internet of things. Creating IoT products and testing them is a big challenge. It’s safe to say that with IoT technologies, the amount of possibilities to test is infinite. We just cannot test it all. Separating a functional part of an IoT product helps with testing. This functional part can be described in some form of requirement. It serves as input for a test case. The four elements that make up a test case are:
- Precondition (desired situation you want to start with)
- Steps to follow through the system under test (manipulation of the starting condition)
- Expected result (describe the situation you expect the system to be in after the steps taken)
- Clean up (make sure the system ends up in a known and safe situation)
Test design techniques produce a set of steps and expected results. Test design techniques are selected because of specific risks are involved and a certain coverage is needed.
This all works in the safe and controlled environment of the functional part of the IoT product. Combining all parts of the IoT technology, and thus combining all tests available, gives a broader test coverage of the system. Not necessarily the right coverage of the system is achieved.
The combination of all functional parts creates infinite possibilities for an IoT product. The next step is making choices on dealing with infinite possibilities. We cannot test it all.
IoT layers steer away from functional testing
Break down an IoT product in layers of functionality and you will get:
- Thing layer (physical thing with actors and sensors)
- Bridge layer (the communication part, think Wi-Fi, LoRa, 4G)
- Data platform layer (data storage, cloud and possible business intelligence)
- App layer (an interface that could be a website or even a touchscreen on the thing itself)
Each layer can now be tested in splendid isolation. Coverage can be reached and a good sense of confidence is gained on each individual layer. Since we now know each layer is functioning well, the integrated product can be looked at. The focus can now shift from a functional view to more non-functional characteristics. Think on aspects like performance, user experience, interoperability or installability.
IoT products characterize themselves as solutions with added value. They have to create a business value on top of an existing function. In that sense, functionality is not top priority anymore. This is something that must work anyway. Take a coffeemaker at home. This machine makes coffee. The IoT version of the coffee machine collects data on your coffee-drinking behavior and learns when to brew what kind coffee. It still brews coffee. The added value here is that you think it makes great coffee because the timing is always right.
From this example, we learn that functionality of coffee making still has to work. Testing the IoT product must focus on the added value of unnoticed brewing of the right coffee at the right time. It is more about (in this case) the IoT coffee experience. Testing an experience is something different then testing a functionality. You might start using techniques like storytelling or crowd-testing. IoT testing is less functional testing and more non-functional testing!
Time to execute
With the focus of testing shifted away from functional testing, it is time to execute the tests you have planned. Test execution requires the IoT system and a test environment. I like to picture the IoT test environment as a set of concentric circles (like an onion).
- Circle 1: The inner core; one physical IoT product (the thing itself)
- Circle 2: The thing and the rest of the IoT technology
- Circle 3: A set of the same things working together within the IoT solution
- Circle 4: IoT solutions of the same product family working together (only domestic appliances of one brand)
- Circle 5: All IoT solutions from one brand must coexist correctly
- Circle 6: A defined set of IoT solutions of different brands, but of the same type (for example, all connected cars must travel the roads safely)
- Circle 7: All IoT solutions out there
- Circle 8: Future IoT solutions that could possible connect to your product
Of course, the entire set of circles may be applicable to you. Risk estimation can lead to testing only in test environments that cover circles 1 to 4. The exact definition of these circles may vary because of the architecture of your IoT product. Maybe you think you need more detailed circles for defining test environments. The idea of the concentric circle model for IoT environments is here to help you define your IoT test environment. With the IoT layers, a structured approach to testing IoT solutions takes shape.
Your IoT testing strategy
It is time for action. This article defines the elements needed to build your IoT testing strategy that copes with the infinite test combinations problem. With IoT layers, the right functional tests are executed (not too much!). Focus shifts to the non-functional quality attributes for IoT technologies. Specifically, the IoT experience needs coverage. With a good definition of your test environment, everything is in place to start with IoT test execution.
No need for infinite test time to cover infinite possibilities! As a side effect, the time-to-market for IoT products goes down. This can be a new bottleneck in IoT testing, by the way. I will keep that discussion for another time.
With this knowledge, I think the next generation of game consoles will be great and perfectly tested when they enter my home. I am going to play a game right now. Let me know how you test your IoT product!
Artificial intelligence, the internet of things and digital transformation have been popular subjects over the last year. A quick scan of your favorite tech publication will likely result in multiple stories covering all three of these concepts as companies across the globe embrace them.
Most recently, AI has driven a bulk of the excitement. According to a recent survey conducted by MIT Sloan Management Group, nearly nine in 10 business executives believe AI will serve as a key competitive advantage in the near future, helping them explore new opportunities and potential revenue streams. Yet, a lot of question marks still hang in the air — less than two-fifths of these same respondents even have an AI strategy in place, while only 16% understand the costs of AI development in the first place.
Of course, it’s hard to talk about AI without also talking about machine learning. As a current popular application of AI, many companies across the globe are trying to identify ways of applying machine learning to provide intelligence that will boost their business. But doing so is beyond the reach of most organizations due to the complex and resource-intensive data science lifecycle. So why not apply AI to AI and reduce the number of expensive and scarce human resources needed to teach AI everything it needs to know to be effective? This approach can effectively democratize machine learning, standardize it across the company and make it accessible to more people. In fact, business giants like Uber are already taking this approach.
IoT is a key driver of both the machine learning and AI craze. With the volume of data produced by machines and people on a daily basis becoming unmanageable, it has become increasingly difficult to make use of this information — and the proliferation of connected sensors only serves to further up the ante. Without AI and machine learning, making heads from tails of this data is downright difficult and creates problems for businesses looking to use their data.
With digital transformation at the forefront of many business initiatives, applying AI to IoT can help drive the innovation and business effectiveness that many companies are hoping to achieve. Applied correctly, companies can change the way they operate through the use of these digital technologies to realize key competitive differentiators.
Some industries are more proactive in this regard than others, which is best observed in Constellation Research’s recently released “Business Transformation 150.” This list recognizes some of the global leaders of digital transformation — people who have minds for productive disruption and experimental technologies, and are spearheading digital initiatives at their companies. Nearly one-quarter of the people on the list are from the manufacturing industry, which should come as no surprise when you consider that the sector is one of the pioneers when it comes to using IoT data to fuel machine learning and AI initiatives.
The convergence of today’s digital trends and what it means for you
AI, IoT, digital transformation — these key concepts will be important to the success of businesses in the near future, if they aren’t already. Business leaders already see the potential advantages they can gain by capitalizing on AI, IoT and digital transformation, now they just need to devise the right approach. This starts from the top, and is one of the many reasons why organizations have created centers for excellence or added new job titles, from chief mobility officer to chief data officer. It may be time to broaden this focus that some organizations are doing via a chief digital officer, but the definition and scope depends on the business priorities. Regardless, it makes sense to determine your organization structure and executive leadership based on your digital goals — including the role of IoT.
Digital transformation means a lot of different things to a lot of different businesses. While there is this notion that these efforts are led by the desire to improve the bottom-line customer experience, that doesn’t mean digital transformation can’t impact behind-the-scenes operations either. IoT, machine learning, AI — these technologies all touch the customer experience in one way or another, whether it’s by speeding up manufacturing time, improving product design, streamlining customer service or anything else.
IoT remains a key driver that will impact nearly every industry in the future. Regardless of whether you’re tracking packages as they’re shipped or looking to identify machine performance anomalies, being able to collect and sort all this data and then act on it will be pivotal to success. Data lies at the foundation of AI, so businesses looking to take advantage of these technologies will need to learn how to manage and use their data effectively.
While the manufacturing sector is pioneering advancements in this space, other industries should use them as a case study and learn by example. From a digital standpoint, many business leaders have a tendency to look inwards as they shape their own strategies — they react to and imitate the leaders in their own sectors. However, in a time of such fast-moving digital innovation, the best approach is often looking outside the industry, where you get can find new insights and strategies. Driven by the Industry 4.0 revolution, manufacturing has done a lot of the early digital transformation groundwork. It’s not a bad starting point for other industries — particularly those making heavy use of machines that need to be optimized and managed effectively. Then you can extend those efforts to other predictive use cases.
There’s no doubt about it. 2017 has not been the year of the smart home. But the future of IoT is huge, and the smart home is undoubtedly one segment that will grow fastest in 2018.
Smart home products are making their way into every part of our lives. In 2017, we saw an explosion of smart products for the home at every life stage. From robots that read kids bedtime stories to devices that feed your cat and play with your dog, there’s a smart product out there for everything.
Thanks to IoT, the days when you can leave your home without a worry are not far.
We have heard for many years that truly intelligent homes are coming and that they will transform our daily lives. Industry experts predict the typical intelligent home could contain more than 500 smart devices by 2022.
One of the reasons to acquire smart home appliances or install smart meters in the home is the promise of saving energy. Another key reason is because smart homes will be a key component in the healthcare system; the elderly population will greatly benefit from smart homes.
The technology while important, it is only as an enabler. Nowadays, the complexity of installing devices and software from smart home vendors requires the expertise of super installers.
Artificial intelligence will be critical in the future smart home, a learning ecosystem that will get insight from personal and home devices. This insight will reveal patterns of behavior of family members’ movements. Our daily life will be transformed in a whole new way.
Most of the big technology players are working on their own voice assistants. In 2018, companies like Amazon and Google will be fighting to lock you into one voice ecosystem for your smart home. Before the year is up, you may have to declare your allegiance for Alexa, Siri, Cortana or Google Assistant.
To realize the future of smart home, it will be critical that smart home ecosystem collaborate to adapt to the changing needs and priorities of consumers. The smart home industry must work hard the make devices reliable. At the end of the day, people crave convenience, and the main appeal of smart products is to make lives simpler. If devices cannot consistently achieve this for their owner, they will not be of use to them.
In order for the smart home to reach critical mass, separate devices need to get better at communicating. So in 2017, companies began to understand that their products, no matter how cool, couldn’t exist in a bubble by themselves. As a result, we began to see interconnection in the smart home move forward, and in 2018 this is only expected to grow.
Cyberattacks have become a true headache for the IoT industry, as each connected device is a potential security breach. Smart homes are not immune from attacks, so manufacturers of home automation devices must pay a lot of attention to security issues and to ways of increasing the safety and protection of the user’s data.
Ransomware is getting more sophisticated, and there have already been examples of smart thermostat hacking. Obviously, if a thermostat can be compromised, what can stop the hackers from getting into a smart door lock or window sensor? Only a set of security measures matching the skill and toolset of the attackers, of course.
We already have smart doorbells and smart thermostats, but what about making other parts of our homes smart? With this month’s Consumer Electronics Show (CES), homeowners and tech enthusiasts alike are antsy about what 2018 has in store for smart homes. It’s pretty clear that this year’s CES is going to be smart home-heavy. At CES 2018, many companies will be announcing new fashion devices, integration with voice assistants, robots, smart appliances and face recognition — but importantly, almost everything they will be showcasing involves interoperability and integration. It’s likely that CES will demonstrate how Google and Amazon make Alexa and Assistant work with as many devices as possible. Amazon has already shown off its own home access platform for deliveries, so why not widen that to third-party hardware? And since it’s a neck-and-neck race between these two companies, expect Google to follow suit, letting you open your front door simply by speaking. Not to mention all of the see-through fridges, connected crockpots and smart bathroom facilities that are likely to be on display.
Other predictions for 2018 in the smart home arena include that we’ll see the first IoT-enabled pest control products. We’ll also see the first robotic IoT-enabled vacuums released that have cameras and are connected to identify objects such as Legos and store them to prevent them from being stepped on.
If you’ve been in the smart home market for the past 15 years, you know the business is quirky and dynamic. Until recently, it seemed that the smart home was only available to a few. But now we are beginning to see the horizon clearer and can dream that in the not too distant future we can all have a smart home.
Thanks everyone for your likes, comments and shares.
We are on the verge of several major technology shifts converging to disrupt everything tech as we know it today.
Cloud to edge
Companies have been embracing cloud computing for nearly a decade, but it’s currently being disrupted by the IoT phenomenon. Analysts are predicting that there will be 75 billion internet-connected devices by 2025. The cloud was not designed for massive sensor data uploads, nor was it designed for low-latency, real-time communications. This is the catalyst for all IoT platform vendors racing to release edge computing gateways and appliances to bring more connectivity and computing capabilities to edge networks rather than routing everything through “the cloud.”
Big data to artificial intelligence
The demand for consuming and interacting with live streaming IoT data along with artificial intelligence and machine learning technologies is now driving a massive shift away from the cloud to edge computing. For example, self-driving vehicles do not have time to stream their sensor data to the cloud and wait for machine learning feedback or driving commands. These self-driving cars are essentially mobile data centers. The car’s on-board computers are analyzing sensor data in real time and learning while making control decisions in nanoseconds — no cloud involved.
Centralization to decentralization
The current value of Bitcoin cryptocurrency has trickled its way into our everyday conversations. At the same time, there is a massive blockchain revolution underway along with controversial net neutrality legislation. These topics are driving the shift to decentralized computing, decentralized web and decentralized app (DApp) development.
The convergence of these three drastically different computing shifts may sound overwhelming, but with great challenges come great opportunities.
With over 75 billion internet-connected devices expected by 2025, there’s going to be a ton of idle/wasted CPU resources and an insatiable demand for machine learning computes! We are moving into an era of decentralized and distributed computing where everything computes (together) as if they are peer-to-peer nodes on a global mesh computer. Decentralized web and decentralized apps will run on this new decentralized and distributed mesh computer. IoT devices will be able to self-organize on ad-hoc mesh networks and request and/or perform machine learning computations in real time without requiring cloud or edge computing infrastructures.
In this new economy, everything computes to form a new worldwide peer-to-peer decentralized and distributed mesh computer!
Is your company creating an “innovation lab” to build a new IoT platform for your enterprise? We’ve worked with Fortune 1000 brands across several industries to accelerate their digital transformation journeys, and have seen only a handful of these centralized teams ever successfully deliver revenue-generating IoT systems. These are strong, motivated developers, but the internet of things is different, even more so for industrial solutions. IoT systems aren’t something you connect to your products — IoT systems are your product. The end-to-end experience, and the end-to-end business model, require much more than a central innovation group can readily deliver. Here’s why.
Machine data is useful to, well, machines. Not humans. Or your business operations. When the data from the sensors on your asset reports “Temperature is 100 degrees” and your programmed logic includes a maximum threshold of 95 degrees, machine-only data is sufficient for turning on a flashing light or wailing siren, or to initiate a system shutdown. But what if you want to send a text or email message to a specific person about the problem, or enter a support ticket, or initiate a parts replacement order? That takes data that does not come from machines at all. How does your IoT solution know who to alert and via what method, for which machines and under which conditions? Where is this machine installed, and what parts are inside it? Is it under warranty and what service agreements have been made for this particular customer? For replacement parts, who are the approved purchasers for automatic ordering? In order to do much beyond announcing “SEND HELP,” your IoT technology must be integrated with your enterprise systems and be able to ingest and produce data that your various systems (CRM, ERP, etc.) can understand, and combine these disparate sources of data into actions that provide value to you and your customers.
Centralized enterprise innovation teams building their own IoT platform can’t do this on their own. They may have total control over what information is collected from your devices and how it is analyzed, but without integration with data from your other enterprise systems, their technologies will be less than compelling for the various lines of business, service organizations and product sales teams for whom these the products are supposedly built for. These enterprises often find themselves with an IoT platform that is ignored outside of the innovation group, and each P&L trying to build its own custom systems, each incompatible with the other, and everybody wondering how they’re going to make any money from IoT at all.
Successful adoption and deployment of revenue-generating connected products require your central IoT champions to work directly with your IT teams and business unit leaders. Together they must create a flexible system for integrating data from systems across your enterprise to produce insights that create business value. Customer data, BOM history and organization details are often unstructured and exist in human-readable formats not readily processed across disparate or automated systems. These silos (and the data inside them) are frequently maintained by IT departments for the business units. To transform this information into useful context for enriching machine data (i.e., notifying the technician responsible for a specific machine under warranty for a particular client that it is having issues with a certain part that is available in what storeroom and where that machine is currently installed), your IT team and your business units will have to work together to structure their data appropriately for integrating with any IoT system.
Enterprises should not seek to build a new all-encompassing IoT platform, but instead adopt and support a central, flexible framework for solving the common complexities of IoT. Data security, user management, access control, data cleaning and transformation, and other functionality where the consistent best practices are both critical and non-specific to any particular type of machine or business context. These should be made available through a common enterprise API that each line of business can use to deliver their unique value to their customers. Each business is still responsible for its specific customer offerings (and can proceed at its own pace), without having to worry about the challenges of collecting, processing and securing the data in the first place — or being constrained in their ability to unlock the value in their data by a monolithic centralized system. Consistent IT policies can be maintained across the entire enterprise, while each team remains empowered to move quickly toward creating compelling connected product scenarios for their particular markets without the cost (time, budget and opportunity) nor the risks (security, reliability and performance) of (re)building infrastructure. This approach delivers not only more efficient use of resources, a faster route to production and higher-quality technologies, but is also much more likely to deliver value to your customers who care about your product, not your IoT platform. They just want business outcomes like asset management, workflow integration, predictive maintenance and yield optimization.
Few organizations would expect each line of business in their enterprise to be responsible for building its own cloud infrastructure, storage or machine learning tools. Likewise, your teams shouldn’t have to work out their own ways of integrating these primitives from public cloud providers like Amazon, Microsoft and Google into their architectures.
A centralized team (independent or virtual via representatives of each business unit) should provide consistent, well-tested APIs for individual teams to build applications on top of, rather than require each application team to work with each cloud service directly. When central teams are instead created as “innovation labs” with the goal of building not a common framework of best practices, but a new company-wide IoT platform, the most common business outcome is simply failure.
Though the internet of things is in its early years, it has already started to transform how people live. For instance, cars have GPS to determine the fastest route. There are also communication systems built inside homes to keep track of children. However, these are not all that IoT can do, not by a longshot. IoT technology could help make a more sustainable world.
What’s special about IoT?
IoT technologies are perfect for environmental sustainability as they are communicative and analytical. Ecosystems, both environmental and technological, are too complicated for action and analysis by any one technology. At the very least, it takes pairs of technologies. Soil sensors have to communicate with sprinkler systems for water use regulation. Emissions sensors should communicate with heavy machinery to minimize noxious pollution. Water contaminant sensors should communicate to determine harmful pollutants in rivers and oceans.
Ways businesses can use IoT to save the environment
Natural disasters have risen in the past years. In the wake of such occurrences, it is undeniable that organizations and businesses are responsible for a great deal of the environmental factors that lead to extreme weather situations. For instance, consider the amount of greenhouse gas emissions that are produced in the manufacturing process alone. Fortunately, businesses have the potential, as well as the necessary technology, to contribute to the restoration of at least some damage and lower harmful effects. The secret weapon? The internet of things, with the help of software companies.
A recent survey found that nearly 100% of CEOs and vice presidents believe IoT is already contributing to a more sustainable future and would continue doing so in five years’ time. Nonetheless, despite the huge potential, only half of C-suite leaders reported at present using data and connectivity to support sustainability initiatives, often citing structural limitations and priorities in business competition as major obstacles. With this in mind, there are a few steps that businesses can take to minimize the effects of the barriers and set their organization on the right path to be champions of a more connected and sustainable future.
- Emphasis on digital citizenship and individual responsibility. For sustainable efforts to succeed, business organizations must first understand their duty to help people function via technology as citizens and help employees recognize that their individual efforts and endeavors are worthwhile. It’s necessary to remind employees that by being corporate citizens, each plays a critical role in shaping a future wherein strategic sustainability and growth opportunities are intertwined and could support one another. To help create a culture of community and commitment, business leaders must encourage their employees to take a proactive stance in designing the sustainability initiatives of the organization. In order to do this, companies should instill a sense of individual responsibility. For example, sustainability teams could use technologies to immerse people in scenarios that demonstrate the effect of short-sighted environmental policies and create empathy regarding issues, like desertification or deforestation, that otherwise could feel too far away to matter.
- Collaborate to make guidelines for technology development. Since technology develops faster than legislation, there’s a lack of unified environmental guidelines for emerging technologies. However, there are also limits to how much companies can prioritize environmental advantages over economic costs, since the key objective in most organization is to maximize revenue. By working with government organizations, businesses can help develop appropriate measures to ensure technology is channeled for the greater good at scale. There are times that businesses have difficulty complying with federal environmental goals as the timeframe is limited or if they lack internal resources. To close the gap, businesses and governments could coordinate and together build more ecofriendly guidelines related to IoT devices and open data.
- Share resources and knowledge across departments. Once individuals determine and acknowledge their responsibility, leaders can help foster a collaborative business environment across numerous departments. As it stands, only 23% of survey respondents said there is considerable collaboration between IoT and sustainability experts within their own organization, let alone within an external network. Focusing on cross-functional disciplines could lead to building systematic and disruptive technologies. When employees are actively sharing projects and resources freely, it creates a fluid information exchange among individuals and groups of different areas of expertise that in turn could spark new ideas as well as inspire experimentation. Often, innovation follows.
Businesses play a critical and necessary role in creating a sustainable future. The exchange of ideas, expertise and data among people, organizations and governments could enable the corporate world to create more sustainable products and services.
IoT devices remain in their infancy, with a lot of questions being asked and experimenting going on. Radio frequency, or RF, is one technology that IoT companies and startups are dialing into to advance state-of-the-art IoT and, in doing so, increasing profit margins for their IoT products. In some cases, there are a lot of guesstimations going on, largely based on assumptions gleaned from smartphone technology.
However, with IoT devices and unlike smartphones, RF is a brand new ballgame, something akin to “black magic” as some tech pundits put it.
The prudent thing to do is take a close look at RF and the specific challenges it poses to IoT devices. There are some key design considerations you have to factor in. Those deal with the right RF antenna to use, anticipated RF interference, impedance matching and testing.
But even before those considerations, you first have to look at RF and what you expect from it in your IoT product. Remember what I said earlier: Growing numbers of IoT devices are based on rather small rigid-flex and flex circuits. There’s very little real estate on these tiny boards for the micro devices populating them.
The type of RF antenna and precisely designing it in your IoT device should be at the top of your considerations. Chip, proprietary and wire antennas are typically used for printed circuit board (PCB) designs. For IoT and RF applications, chip antennas are mostly used for ultra-crammed designs, mostly in low frequency ranges. Chip antennas are easy to implement in a design, but they are somewhat expensive.
An antenna is central to the performance of an RF device, especially when you’re doing verification and certification testing and calibration. There are certain devices that operate at different RF frequencies, sometimes called “multiple bands of frequencies.” The antennas for these must be very precise. For example, the antenna must be designed properly and follow the rules of physics to have good access to the multiple frequencies on the RF band. An example is a large antenna, larger than Wi-Fi running at speeds of 2.5-5 GHz.
Also keep in mind that RF is a very sensitive circuit — and even more so in a small IoT device. Noise is the troublemaker in this instance as it creates RF interference, which translates into poor signal integrity. You have to take into account the types of components you’re using on the rigid-flex or flex circuit and where they’re placed. Analog circuits are usually the culprits causing the most noise. The basic rule of thumb is to separate digital and analog circuitry at a safe distance to avoid problematic analog-created noise from disrupting digital circuit operations.
The third major design consideration is impedance matching. It is important because RF signals are extremely noise sensitive. A small noise ripple can alter RF signal performance to a great extent.
Therefore, impedance matching is extremely critical for RF. Digital signals — even if they are very high-speed — have a certain tolerance. But for RF, the higher the frequency, the smaller the tolerance becomes. For example, the PCB designer must keep it at 50 ohms — 50 ohms out from the driver, 50 ohm during transmission and 50 ohms in to the receiver.
Lastly, RF design must be done so that you can test it. Cost associated with testing and certification is significant. You have to consider that space is a limiting factor. You also have to think outside the box when doing an IoT RF design so that it can be satisfactorily tested.
Keep in mind that you don’t have the latitude or flexibility to test an IoT PCB like a rigid conventional PCB. In a conventional PCB, you have plenty of real estate to put test points for the oscilloscopes, multimeters, all of these. You don’t have that luxury with IoT PCBs due to real estate limitations, and testing in an IoT device should be modular, but precise.
Even the IoT PCB testbeds you design are very tight in terms of size and in terms of the tolerance. Sometimes, you have to design a segment of the board where you can test in a batch format.
Here, for example, you check function levels 1, 2 and 3 in one set of tests because you don’t have the luxury of creating a test program for individual parts of the circuitry because space is limited. Without precise modular testing, your IoT device performance would be jeopardized and field failures would result.