Originally posted on Data Science Central

What exactly is “IoT”? Internet of Things, yes; but what does that mean?

Internet of Things is a structural definition; it says there are “Things” such as sensors and devices (on machines or people) connected together in a Network. So what? What does a Network of Sensors & Devices allow us to DO? What is the functional description of IoT?

Being able to connect things together is “table stakes” at the intelligence augmentation game. What you are able to do with this network is the real story.

IoT is an enabler of THREE high-level objectives:

(1)    DO MORE: Whether machines producing more (high throughput) because of less breakdowns or a weekend athlete burning more calories because she is able to keep her heart rate in the fat-burning zone, we are accomplishing more.

(2)    HIGHER QUALITY: By monitoring environmental pollution, cities restrict automobile access into city center for better health outcomes over time.

(3)    BETTER USER EXPERIENCE: In the near future, mass customization will allow me to find eyeglasses with perfect fit at low cost based on video inputs at connected additive manufacturing facilities.

All three objectives can be met because Network of Sensors & Devices exist and will grow. But Networking of Sensors & Devices does NOT capture what has to happen so that we can DO MORE at HIGHER QUALITY & BETTER UX! The information that rattles around the network has to be consumed properly, insights generated and decisions made.

A functional description of IoT is a network that “processes” information in the network towards our three objectives. One can see that “processing” is enabled by a multitude of technologies: IP networking, wireless, chipsets, protocols, security, cloud computing, database technologies, analysis software, visualization and so on. Subsuming this under “IT or Information Technology” and keeping it aside for the moment, the “higher-level” processing involved is Applied Data Science!

Applied Data Science is a tautology. Data Science IS the applied aspects of many pure sciences (see “What exactly is Data Science?” for details). Beyond the network of sensors & devices and base IT technologies partially listed in the last paragraph, what is unique and new in IOT is Data Science applications –Data Science applied with the focus on information extraction, insights generation and prescriptive decisions. There is no identifying name for this *applied* aspect of Data Science but I have been referring to it as “Engineering” Data Science. We use “engineering” in the sense of the applied aspect of any science (Engineering is the applied aspect of Physics, for example).

IoT = (Network of Sensors & Devices) + IT + (Engineering Data Science)

Each component is critically important to IoT; major advances in all three in recent years have made IoT and its promise real.  Engineering Data Science (EDS) is the youngest and the least mature of the three. Immediate next steps in EDS evolution seem clear to me  (more in Next Stage in IoT revolution – “Continuous Learning”).

When IoT is defined as “(Network of Sensors & Devices) + IT + (Engineering Data Science)”, it seems to pervade ALL industries from my vantage point! What do I mean by that?

Let us look at the largest 5 (excluding IT) sectors of S&P 500: Consumer Discretionary, Energy, Financials, Health Care & Industrials.

I have mentioned only a few instances in the right hand column but you can add many more. They all require some data generating mechanism, IT connectivity and decision making software. What this shows me is that IoT has already pervaded and will totally engulf all the businesses! As such, I tend to look at IoT as a technology framework that underpins ALL businesses and industries of the 21^st century.

Now, does the exact same IoT serve all these sectors or are there nuanced variations? As I mentioned, EDS is the youngest and fastest evolving portion of IoT today – let us focus where the changes are most rapid.

I have partitioned applied Data Science into three: Industry, Business & Social Data Science. 

As you can see, each application area calls for refinements and adaptations to its verticals. Specialization for each vertical notwithstanding, the three “types” of Data Science are best seen as a unified whole, which we are calling “Engineering Data Science or EDS”. Rapid progress in Engineering Data Science is required to achieve our three goals of “DO MORE at HIGHER QUALITY & BETTER UX” with IoT.

IoT is not JUST what GE, Siemens, ABB or Hitachi do! It is a technology framework for all business and industrial technologies going forward. IoT is just a metaphor for this powerful, all-encompassing technology framework.

SUMMARY:

• IoT = DO MORE at HIGHER QUALITY & BETTER UX.
• IoT = (Network of Sensors & Devices) + IT + (Engineering Data Science).
• IoT = Technology framework that underpins ALL businesses and industries of the 21^st century.

PG Madhavan, Ph.D. - “Data Science Player+Coach with deep & balanced track record in Machine Learning algorithms, products & business”

http://www.linkedin.com/in/pgmad

Follow us @IoTCtrl | Join our Community

• Minimize maintenance costs - Don’t waste money through over-cautious time bound maintenance. Only repair equipment when repairs are actually needed.
• Reduce unplanned downtime - Implement predictive maintenance to predict future equipment malfunctioning and failures and minimize the risk for unplanned disasters putting your business at risk.
• Root cause analysis - Find causes for equipment malfunctions and work with suppliers to switch-off reasons for high failure rates. Increase return on your assets.
• Efficient labor planning — no time wasted replacing/fixing equipment that doesn’t need it
• Avoid warranty cost for failure recovery – thousands of recalls in case of automakers while production loss in assembly line

TrainItalia has invested 50M euros in Internet of Things project which expects to cut maintenance costs by up to 130M euros to increase train availability and customer satisfaction.

Gone are the days when we used to carry big fat wallet filled with cash, coins, multiple credit cards, business cards, travel tickets, movie tickets, personal notes, papers with names, numbers and the list can go on.

• Payments - NFC payments, Biometric payments using finger scans, facial recognition, voice recognition, retina based check. Major players in this space are PayPal, Apple Pay, Android Pay, Samsung Pay etc.
• Virtual or digital currency
• Tsunami of Apps from Google maps, to zomato helping us throughout the day
• Retailers can use targeted mobile campaigns for customer acquisition, retention
• Live streaming apps like Meerkat and Periscope delivering targeted content to site-specific users which benefits both the consumer and the creator.

• With mobile enablement, a merchant can enhance your payment experience and boost operational efficiency
• Real time communicating with the customer, can be greatly enhanced through mobile enablement. Businesses can quickly respond to customer complaints or questions through social media, or the apps
• By analyzing the data generated by mobiles using Big data Analytics, businesses can give personalized experience to consumers
• Digital assistants like Google Now, Siri are helping everyone

Here are some well-known industry examples:

• Starbucks processes over 8 million mobile transactions each week, this data of mobile user behavior to customer preferences, is then analyzed by a team of data scientists for insights.
• Coca Cola is using mobile apps for field sales folks, equipment service teams and knowledge workers and commercials like get free coke on mobile
• The emergence of hyper-local startups like Jugnoo, Zopper, Grofers and PepperTap using mobile first strategy
• Virgin Atlantic, Bank of America, Delta, Chipotle have their industry leader apps for fantastic customer experience

As the penetration of smartphones and internet is increasing with 5G and beyond, along with the changing shopping behaviors, the mobile revolution is here to stay and impact the Digital Transformation further.

In today’s  digital age of customer  hyper-personalization, organizations identify opportunities for real time engagement based on data-driven understanding of customer behavior.

• Identity: name, location, gender, age and other demographic data
• Relationships: their influence, connections, associations with others
• Current activity: orders, complaints, deliveries, returns
• History: contacts, campaigns, processes, cases across all lines of business and channels
• Value: which products or services they are associated with, including history
• Flags: prompts to give context, e.g. churn propensity, up-sell options, fraud risk, mood of last interactions, complaint record, frequency of contact
• Actions: expected, likely or essential steps based on who they are and the fact they are calling now

• All customer touch point data in a single repository for fast queries
• Next best actions or recommendations for customers
• All key metrics in a single location for business users to know and advise customers
• Intuitive and customizable dashboards for quick insights
• Real time hyper personalized customer interaction
• Enhanced customer loyalty

Customer 360º helps achieve Single View of Customer across Channels – online, stores, marketplaces, Devices – wearables, mobile, tablets, laptops & Interactions – purchase, posts, likes, feedback, service.

Guest blog post by Brian Horvath

“Many products and applications exist today that can help us take steps toward healthier living." - Ben Bajarin

Big data is everywhere. Yes it’s true that it has always been there but, we are just recently finding ways to best utilize it. All industries are being disrupted with the collection and analyzation of data and medtech is no different.

The majority of data has historically been used to help study disease and find cures. However, with the implementation of smartphones and wearables, big data can now help people with their health in real-time. Experts such as Ben Bajarin from Time Magazine agree that wearables are the future of smart health technology.

 “Many products and applications exist today that can help us take steps toward healthier living,” writes Bajarin, in an article on the future of smart health. “Hopefully, technology and smart devices will not only help the healthy stay that way but, also educate others about how to live a healthier life.

So how is big data being used in medtech?

Helping Athletes Get Better

Each year we see athletes setting new records. The recent Summer Olympics in Rio is a great example where 60 Olympic and 19 world records were broken. It may seem like some records will never be surpassed but, there are many working behind the scenes to make sure athletes continue to reach new heights.

One such company working to make this happen is Under Armour. The company has invested close to a billion dollars on numerous apps to help track nutrition, fitness, activity, and sleep data. “Technology will play a bigger and bigger role in helping athletes get better,” Mike Lee, SVP of Connected Fitness at Under Armour, told Fortune in a 2016 interview. “And that’s the mission at Under Armour.”

Tracking a Healthier Lifestyle

It’s not only about athletes. Under Armour uses apps to incorporate data to help everyday people get healthier. One such app used by Under Armour is MyFitnessPal which counts calories for the user. Another is MapMyRun which tracks a user’s running activity.

So why are these important?

According to Lee, these apps will eventually work together by suggesting a place to eat after you complete a run. It could take into account how many calories you’ve burned, how many you have already consumed, and then suggest a place where you can get a meal to fit within your calorie goal.

Preventing Drug Interactions

“More prescriptions means an increased risk of harmful mistakes being made by elder Americans,” writes Drug Lawsuit Source, in a blog post about medication mistakes in the elderly. “Each new prescription increases the likelihood of a mistake being made, either by the physician or the patient.”

Other than talking with the pharmacist, is there another way to avoid dangerous interactions with medication?

Two of the top apps for drug interactions are Medscape and PocketPharmacist. Both pull data from drug interaction databases to compare with your entered prescriptions. So, no longer a need to wait in line to speak to the pharmacist as the apps will give you instant interaction information.

Keep in mind that these  apps also collect data about prescriptions you use. Eventually, the information will be tracked by your physician who can adjust your prescriptions, receive notifications about potential interactions, and obtain suggestions for alternative drugs that may be cheaper for the patient. It could also be used to help speed up the process of prior authorizations required by most insurance companies .

Final Thoughts on the Future of Smart Health Technology

Currently, medtech is using big data to help create a healthier future for us all. Soon we will have a one-stop-app that will not only track our health but, send recommendations to us for a healthier lifestyle and suggestions to our physicians to keep us living longer.

What health apps have you used and are any of them helpful to you? Is there a limit to the amount of data you feel should be collected and used with these apps?

Follow us @IoTCtrl | Join our Community

Digital Transformation has become a burning question for all the businesses and the foundation to ride on the wave is being data driven.

• Increased need & desire among businesses to gain greater value from their data
• Over 80% of data/information that businesses generate and collect is unstructured or semi-structured data that need special treatment 

Data Scientists:

• Typically requires mix of skills - mathematics, statistics, computer science, machine learning and most importantly business knowledge
• They need to employ the R or Python programming language to clean and remove irrelevant data
• Create algorithms to solve the business problems
• Finally effectively communicate the findings to management

Any company, in any industry, that crunches large volumes of numbers, possesses lots of operational and customer data, or can benefit from social media streams, credit data, consumer research or third-party data sets can benefit from having a data scientist or a data science team.

• Kirk D Borne of BoozAllen
• D J Patil Chief Data Scientist at White House
• Gregory Piatetsky of kdnuggets
• Vincent Granville of Analyticsbridge
• Jonathan Goldman of LinkedIn
• Ronald Van Loon

Data science will involve all the aspects of statistics, machine leaning, and artificial intelligence, deep learning & cognitive computing with addition of storage from big data.

Guest blog post by Raj Dalal

Everyone’s talking of the Internet of Things (IoT) and the impact it’s started to make even on day to day lives. Yet, more than anything, it’s also becoming increasingly clear that analysis of the IoT data will be the differentiator between those who simply collect data and those who “use” it to drive their businesses.

Enterprises that will use IoT analysis will see themselves implementing faster customer services than their rivals, as well as add new amounts of additional yields.

That said, it’s clear that the IoT analytics will require a well-thought out strategy on part of the Enterprise. Unlike the other, modern-day streams of data analytics, this branch is slightly more complicated. The primary reason  – the humungous amounts of streaming data that is/will be generated and analysed, in real time.

How Does One Collect This Data?

How does one collect this data? In fact, is it necessary to collect all the data? These and many other questions need to be answered by an Enterprise’s decision-makers to tackle the complexities of the IoT data.

Traditional data that is used in a B2B or B2C operation, and its analysis, requires the collection of raw data, locating it in a data hub, scrubbing it and then handing it over to the analysts to draw predictive or other forms of analytic models.

But that same structure cannot be applied for IoT data because the huge volumes that will pour out in real time means centralizing it will be almost impossible. Imagine you running a national cold storage company with your own in-house fleet of trucks, mini-vans and warehouses. You can expect copious amounts of data to be generated on various fronts – from your fleet of “smart” vehicles, your thermostats in each section of the cold storage unit in every town, data that comes out of your supply chain, customer-related data, data from the RFID chips, and so on and so forth. If all of this data starts pouring into one central location, after a short while, your central data servers will start suffering from “data lag”. Not to speak of the almost non-existence of the viability of scalability of such an analytics operation. The decision makers in your outfit will no longer be able to take efficient, business-related decisions. Take an example of a freezer unit malfunctioning on a truck carrying seafood across the county. Even if the temperature drops 5 degrees, the whole payload may need to be discarded for food safety reasons.

For now, one way forward looks like collecting information and analysing the same on the smart device itself. In other terms – Edge Analytics. Utilising the smartness of the device and its low cost computational power will help run analytics on the device itself or close to the source instead of the hub itself. As close to the edge of the system seems to be the answer, for now. The Big Data analysis than comes out of this “mini analysis” can then be done in the Cloud in real time.

So, in such a model, Enterprises will have to tackle many questions, including – how much of information has to be captured by the sensors, initially? How much has to be analysed and how much forwarded to a core location for further analysis? The analytics team will have to develop rule-based models that can determine all of this including how the gateway will handle data.

A note of caution - Edge Analytics may not be for every business and an initial feasibility test may have to be done to understand whether your business needs one or not.

Follow us @IoTCtrl | Join our Community

Guest blog post by Bill Vorhies

Summary:  Deep learning and Big Data are being adopted in law enforcement and criminal justice at an unprecedented rate.  Does this scare you or make you feel safe?

When you read the title, whether your mind immediately went for the upstairs “H” or the downstairs “H” probably says something about whether the new applications of Big Data in law enforcement let you sleep like a baby or keep you up at night. 

You might have thought your choice of “H” related to whether you’ve been on the receiving end of Big Data in law enforcement but the fact is that practically all of us have, and for those who haven’t it won’t take much longer to reach you.

There is an absolute explosion in the use of Big Data and predictive analytics in our legal system today driven by the latest innovations in data science and by some obvious applications.

It hasn’t always been so.  In the middle 90s I was part of the first wave trying to convince law enforcement to adopt what was then cutting edge data science.  At the time that was mostly GIS analysis combined with predictive analytics to create what we called predictive policing.  That is predicting where and at what time of day crime of each type was most likely to occur so that manpower could be effectively allocated.  Seems so quaint now.  It was actually quite successful but the public sector had never been quick to adapt to new technology so there weren’t many takers.

That trend about slow adoption has changed.  So while accelerating the usage of advanced analytics to keep the peace may keep some civil libertarians up at night, it’s coming faster than ever, and it’s our most advanced techniques in deep learning that are driving it.

By now you’ve probably figured out the deep learning is best used for three things: image recognition, speech recognition, and text processing.  Here are two stories illustrating how this is impacting law enforcement.

Police Ramp Up Scrutiny Over On Line Threats

The article by this title appeared in the July 20 WSJ.  Given what’s been happening recently both internationally and at home most of us probably applaud the use of text analytics to monitor for early warning signs of home grown miscreants.  The article states “In the past two weeks at least eight people have been arrested by state and federal authorities for threats against police posted on social media”.  It remains to be seen if these will turn into criminal prosecutions and how this will play out against 1^st Amendment rights but as a society we seem to be OK for trading a little of one for more of the other.

It’s always in the back of our minds whether this is Facebook, Twitter, Apple, Google and the others actively cooperating in undisclosed programs to aid the police, but this article specifically calls out the fact that the police were the ones doing the monitoring.  Whether they’ve built these capabilities in-house or are using contractors isn’t clear.  What is clear is that advanced text analytics and deep learning were the data science elements behind it.

Taser – the Data Science Company

The second example comes from an article in Business Week’s July 18 issue, “Will a Camera on Every Cop Help Save Lives or Just Make a Tech Company Richer”.  

Taser – a tech company?  When I think about Taser, the maker of the ubiquitous electric stun gun, I am much more likely to associate them with Smith & Wesson than with Silicon Valley and apparently I couldn’t be more wrong.

In short the story goes like this.  In the 90s Taser dominated the market for non-lethal police weapons to provide better alternatives for a wide variety of incidents where bullets should not be the answer.  By the 2000s Taser had successfully saturated that market and its next big opportunity came from the unfortunate Ferguson Mo. unrest. 

That opportunity turned out to be wearable cameras.  Although the wearable police cameras date back to about 2008 there really hadn’t been much demand until the public outcry for transparency in policing became overwhelming.

Taser now also dominates the wearable camera market.  Like its namesake stun gun however, sales of Tasers or wearable cameras are basically a one-and-done market.  Once saturated, it offers only replacement sales, not a robust model for corporate expansion.  So far this sounds more like a story about a hardware company than a data science tech company and here’s the transition.

The cameras are producing huge volumes of video images that need to be preserved at the highest levels of chain-of-evidence security for use in criminal justice proceedings.  Taser bought a startup in the photo sharing space and adapted it to their new flagship product Evidence.com, a subscription based software platform now positioned as a ‘secure cloud-based solution’.

According the BW article, “4.6 Petabytes of video have been uploaded to the platform, an amount comparable to Netflix’s entire streaming catalogue”.  Taser is a major customer of MS Azure. And for police departments that have adopted, video is now reported to be presented as evidence in 20% to 25% of cases.

But this story is not just about storing recorded video.  It is about how police and prosecutors have become overwhelmed with the sheer volume of ‘video data’ and the need to simplify and speed access.  The answer is image recognition driven by deep learning.  Taser now earns more than ¾ ths of its revenue from its Evidence.com platform and is rapidly transforming from hardware to app to data science company to answer the need for easier, faster, more accurate identification of relevant images.

The Direction Forward

You already know about real-time license plate scanners mounted on patrol cars that are able to automatically photograph license plates without operator involvement, transmit the scan to a central database, and return information in real time about wants and warrants associated with that vehicle.

What Taser and law enforcement say is quite close is a similar application using full time video from police-wearable cameras combined with facial recognition.  Once again those civil liberties questions will have to be answered but there’s no question that this application of data science will make policing more effective.

About those huge volumes of videos and the need to recognize faces and images.  There are plenty of startups that will benefit from this and many with products already in commercial introduction.  Here’s a sampling.

Take a look at Nervve Technologies whose byline is “Visual search insanely fast”.  Using their visual search technology originally developed for government spy agencies they are analyzing hours of sporting event tape in a few seconds to identify the number of times a sponsor’s logo (on uniforms or on billboards) actually appears in order to value the exposure for advertising.

And beyond simple facial recognition is an emerging field called facial or emotional analytics.  That’s right, from video these companies have developed deep learning models that predict how you are feeling or reacting. 

Stoneware incorporates image processing and emotional analytics in its classroom management products to judge the attentiveness of each student in a classroom.

Emotient and Affectiva have similar products in use by major CPG companies to evaluate audience response to advertisements, and to study how NBA spectators respond to activities such as a dance cam.

Real time facial-emotional scanning of crowds to find individuals most likely to commit misdeeds can’t be far away.

For audio, Beyond Verbal has a database of 1.5 million voices used to analyze vocal intonations to identify more than 300 mood variants in more than 40 languages with a claim of 80% accuracy.

All of these are deep learning based data science being put rapidly to work in our law enforcement and criminal justice systems.

About the author:  Bill Vorhies is Editorial Director for Data Science Central and has practiced as a data scientist and commercial predictive modeler since 2001.  He can be reached at:

Bill@DataScienceCentral.com

Follow us @IoTCtrl | Join our Community

The greatest advantage we have today is our ability to communicate with one another.

• Smart Cars will communicate with traffic lights to improve traffic, find a parking spot, lower insurance rates based on telematics data
• Smart Homes will have connected controls like temperature, electricity, cameras for safety and watch over your kids
• Smart healthcare devices will remind patients to take their medication, tell doctors when a refill is needed & help curb diabetic attacks, monitor symptoms and help disease prevention in real time, including in remote areas
• Smart Cities & Smart Industries are the buzz-words in IT policies of many governments
• With sensors and IoT enabled Robots used in Manufacturing - new products could potentially cost less in the future, which promotes better standards of living up and down all household income levels
• Hyper-Personalization – with Bluetooth, NFC, and Wi-Fi all the connected devices can be used for specifically tailored advertising based on the preferences of the individual
• Real time alerts in daily life - The Egg Minder tray holds 14 eggs in your refrigerator. It also sends a wireless signal to your phone to let you know how many eggs are in it and which ones are going bad.

Now here are the Bad things:

• There are no international standards of compatibility that current exist at the macro level for the Internet of Things
• No cross-industry technology reference architecture that will allow for true interoperability and ease of deployment
• All the mundane work can be transferred to Robots and there is potential to loss of jobs
•  All smart connected devices are expensive – Nest the learning thermostat cost about $250 as against $25 for a standard which gets a job done. Philips wireless controlled light cost $60 so your household will be huge expense to be remotely controlled

And the Ugly part:

• Remember the Fire Sale of Die Hard movie, a Cyber-attack on nation’s computer infrastructure - shutting down transportation systems, disabling financial systems and turning off public utility systems. Cyber-attacks can become common when devices are sold without proper updated software for connectivity
• Your life is open to hackers who can intercept your communications with individual devices and encroach your privacy. Imagine a criminal who can hack your smart metering utility system & identify when usage drops and assume that means nobody is home
• Imagine when you get into your fully connected self-driving car, and with some hacking a stalker’s voice come up from speaker “your have been taken” and you may not find Liam Neeson anywhere nearby, to rescue you.

Guest blog post by Brian Rowe

Another big part of the food supply comes from ranches and farms that raise and slaughter various livestock. While ranching is sometimes bundled with agriculture, I discussed farming in Big Data in Agriculture, so we’ll focus on ranching this time around. Somewhat surprising is that big data usage in ranching appears more limited than in farming. That said, there are a number of novel uses of technology and data in animal husbandry.

Credit: Emilio A. Laca

Land Use Optimization

At a high level, the goals of ranching and farming are the same as any business: increase yields and lower costs. Production maximization has long played a role in large operations. A twist to the optimization problem is land use optimization and how that can affect yields. According to NASA, “Australia’s rangelands provide an opportunity to sustainably produce meat without contributing to deforestation” if properly managed. This sort of optimization is made possible by big data coming from satellites. The same article cites how some West African nations use satellite data “to identify areas with agricultural potential and to estimate the amount of food available.” Growing up in rural Colorado, the most advanced tech I saw at ranches were solar powered fences and artificial insemination. Clearly a lot has changed. From a supply chain perspective, these trends also demonstrate how just-in-time manufacturing can be extended to resource allocation.

From a technical perspective, crop and livestock rotation will become outputs of a multi-objective optimization problem. I imagine that the challenge will be less about the optimization and more about the inelasticity of “bioprocesses”. Aside from slaughter or transfer to somewhere else, there aren’t too many options for reducing “inventory”. Presumably these issues already exist, so any solution is bound to be an improvement. Ultimately, there is a race to avoid the outcome that the U.N. foresees: the majority of humans eating insects as a primary source of protein. Even if that future is unavoidable (not necessarily bad), presumably similar techniques can be used to maximize insect yields.

Sensors and IoT

Technology advancements are driving parralel trends in agriculture and ranching. While satellite imagery offers a big picture overview, sensors provide a micro view of individual plants and animals. RFID tags are a first step enabling real-time tracing of an animal. Equally important is the assignment of a unique identifier to facilitate storing electronic records that can be merged into a centralized dataset. RFID is fundamentally passive, whereas sensors are active. This is where biosensors and Precision Livestock Farming (PLF) come into play. PLF is a comprehensive approach to livestock management and animal welfare. The goal is “continuous, fully automatic monitoring and improvement of animal health and welfare, product yields and environmental impacts” Some of the sensors developed to achieve this are surprisingly simple and surprisingly clever, such as sensors that monitor the vocalizations of livestock to determine stress, illness, etc. These advances can also “raise milk yields, while also increasing cows’ life expectancy and reducing their methane emissions by up to 30%” (CEMA). The Biosensors in Agriculture workshop held in the UK presents even more exciting examples.

Other notable research around PLF include image analysis to monitor animal welfare and
classifying the behavior of cattle and fowl based on GPS movements. According to one paper, a decision tree was used to classify four behaviors: ruminating, foraging, standing, and walking. The features were based on distances and turning angles from the GPS data. Not surprisingly, the confusion matrix was pretty poor in terms of distinguishing between ruminating, foraging, and standing. So there’s lots of opportunities to whip out R and randomForest or party to conduct your own analysis (assuming you have access to the data).

Data and Accessibility

Big data is often synonomous with cloud computing and for ranching it’s no different. As with agriculture there are trends to centralize data to “help ranch managers track livestock, view production statistics, plan grazing rotations and generate reports that can offer insight into the health of a livestock operation.” Unlike in agriculture, it doesn’t appear that the machinery manufacturers are taking a role, although it wouldn’t surprise me if some PLF suppliers have cloud platforms for their customers. GrowSafe Systems is creating their own cloud-based dataset based on their customer data. Their system collects and forecasts “complex animal traits such as efficiency, growth, health, stress and adaptation.”

Europe has taken a different approach focusing on defining a comprehensive classification scheme for agricultural systems. Clearly the goal is data interoperability, so data can be widely shared and applied across farms and ranches. This goal is reflected in the three-level system that encompasses environmental factors and GIS data to site-specific measurements of individual animals that affect yields and animal welfare. Landcover data appears to be the most extensive, while biosensing is likely where the most immediate opportunities are to be found.

As data becomes more focused on individual sites and animals, scarcity is the word that comes to mind. In the USA public datasets don’t come anywhere near the level of detail to make a useful analysis. See data.gov for an example of a disappointing dataset. Of course it isn’t clear whether transparency of this sort is even possible. One rancher believes they have a right to privacy and shouldn’t be compelled to open their books to external scrutiny. This is understandable, but does this belief extend to data? Data privacy is a thorny issue, particularly balancing privacy, ownership, and the need for transparency vis a vis food security/safety. Eventually I think economics will force a change of heart if yields and margins increase significantly with the help of open data. However, this may take the shape of data cartels as opposed to truly open data. As big data and centralized data stores become more wide spread, this debate over data ownership will continue to be visited.

Know of some public datasets available for ranching and animal husbandry? Post links in the comments!

This post first appeared on cartesianfaith.com. Brian Lee Yung Rowe is Founder and Chief Pez Head of Pez.AI // Zato Novo, a conversational AI platform for guided data analysis and Q&A. Learn more at Pez.AI.

Follow us @IoTCtrl | Join our Community

Guest blog post by Dhananjay Yadav

Have you heard of medical radio-telemetry? Up until very recently, it’s been the closest thing that the medical field has had to remote, cloud-based intel on the human body. But now, with the convergence of IoT, cloud, and big data technologies, the health-care industry is primed for a revolutionary influx of new life-saving results. In this article you’re going to see a plausible scenario from the healthcare industry, but keep in mind that such cloud-enabled IoT advancements are happening in every vertical – from utilities and transportation, to finance and retail, and everything in between.

Cloud-Enabled IoT in Action

Going back to the remote telemetry example, it works like this. Imagine your Uncle Jimmy recently had a heart attack and was implanted with a telemetry unit that continuously monitors his cardiac activity, and uses radio waves to stream data alerts back to his doctor when a disturbance occurs. Although this may sound cutting edge, it’s old technology. It’s basically an old school monitoring and alert system.

So, here’s how things have changed with the IoT. Imagine now that Uncle Jimmy has a FitBit (to track his daily level of physical activities) and is also taking sensor-enabled pills, like Helius, to capture and report data on his medication adherence, body temperature, heart rate, and rest patterns. Now imagine that each of these devices; the cardiac data streaming device, the FitBit, and the Helius pills are all connected to one another through a cloud-based network. Also consider that this cloud-based network is equipped with big data processing and analytics applications that work ceaselessly to derive insights from the data that’s collectively streaming in from all of these connected devices.

From this network, Uncle Jimmy gets real-time updates and suggestions about his health and wellness status. As a form of preventative medicine, the network is able to tell Jimmy when he forgot to take his heart meds, or should consider getting in a little exercise, or maybe should sleep in for a few extra hours, to protect his health in light of his underlying cardiac condition. These are predictive and prescriptive data insights based on real-time streaming data that is generated by data producing devices connected across a cloud-based network. Definitely a step up from the monitor and alert systems of yesteryear, that were only able to warn of problems after they had started. Now that you understand what the convergence of these technologies could look like in real-life, let’s take a gander at how it all works from the inside out.

Getting to Know the Lingo

First and foremost, to grasp this technology you need to know the vocabulary that’s used to describe it. An IoT cloud (also called the “fog”) is a network of cloud-based services that are connected to IoT-enabled devices. An IoT cloud supports the big data processing and analytics requirements of a broad IoT network, enabling it to make intelligent, adaptive, and autonomous decisions. The IoT-enabled devices that are connected to, and that sit on an IoT cloud are called edge devices.

Most of these edge devices come paired with their own device-embedded analytic applications, or device-embedded applications that are capable of processing and deriving insights from local data that’s captured by the device. One benefit of these device-embedded analytics applications is that, in many cases, they successfully bypass the need to send data back up into the IoT cloud for processing there.

Many device-embedded analytics applications are built on adaptive machine learning algorithms, called adaptive IoT applications. These adaptive IoT applications enable devices to adjust and adapt autonomously to the local conditions in which the device is operating. IoT cloud application developers are data scientists and engineers who focus exclusively on building adaptive IoT applications for deployment on local devices. The more general IoT developer, on the other hand, is responsible for building products and systems that serve the greater needs of the IoT cloud at-large, including all of its connected IoT devices, data sources, and cloud computing environments.

So once again referring to Uncle Jimmy and his network of connected health monitoring devices, in this example, the heart monitor, the FitBit, and the sensor-enabled pills are edge devices. They’re all connected across the fog, or IoT cloud. Since the FitBit can produce analytics outputs locally, without the need to stream data back to the cloud, it must have a local device-enabled analytics application, certain to have been developed by an IoT cloud application developer. The entire network of connected devices, their data streams, and the cloud-based applications that are used to process, store, and analyze all of this health data is built and maintained by IoT developers. Makes sense now, right?

Where People and Skillsets Fit In

All of this talk of autonomously functioning, adaptive devices may have you worried if there’s even going to be a place for people and their skillsets in the future. Rest assured, just looking at this from the technical perspective, it takes a whole army of knowledge workers to build, deploy, and maintain cloud-based IoT networks. Spark and Hadoop developers are required to design and execute the adaptive algorithms that are embedded on edge devices. Electrical and mechanical engineers are required to design many of the electromechanical devices that will be connected to IoT clouds. Legions of software engineers that code in Python, Java or C will be required to develop and maintain the IoT network. Data engineers will be required to configure and maintain cloud-based big data processing applications, and data scientists are required to build the device-embedded analytics applications.

About the Author

Lillian Pierson, P.E. is a leading expert in the field of big data and data science. She equips working professionals and students with the data skills they need to stay competitive in today's data driven economy.

She is the author of three highly referenced technical books by Wiley & Sons Publishers: Data Science for Dummies (2015), Big Data / Hadoop for Dummies (Dell Special Edition, 2015), and Big Data Automation for Dummies (BMC Special Edition, 2016).

Lillian has spent the last decade training and consulting for large technical organizations in the private sector, such as IBM, Dell, and Intel, as well as government organizations, from the U.S. Navy down to the local government level.

As the Founder of Data-Mania LLC, Lillian offers online and face-to-face training courses as well as workshops, and other educational materials in the area of big data, data science, and data analytics. Follow me here: Twitter | Linkedin

Follow us @IoTCtrl | Join our Community

Guest blog post by Vishal Sharma

A buzz word around us for quite some time now is Internet of Things (IOT). To Simply define it:

“The internet of things (IoT) is the network of physical devices, vehicles, buildings and other items—embedded with electronics, software, sensors, actuators, and network connectivity that enable these objects to collect and exchange data.” (Wikipedia)

It simply means for me all devices that are connected to internet forms part of global network, producing data, that can be utilize for the betterment of services or customer experience or the way one can use it, some examples

• Send real time alert, Smart wear to your doctor or from a machine nearing permissible temperature for over heating
• Real time diagnostic like heart rate, pulse, Temp or SO2 Levels.
• Security breach detection Etc.

How this is done is not what I am focusing around, once it is implemented and if it’s done with integration of all your devices / networks which work as entry point e.g. your mobile, GPS you use etc.

What will be level of privacy remains in a complete IoT world?

All devices are in connection and all are talking to each other with some kind of BIG data tool and Analytics working together. What will happen?

Some scenario

• You are running out of fresh milk in your fridge and now smart fridge will send an order to your grosser for replenishment of the same.
• You have a Smart watch or a fitness device which help your Coach to monitor your activity and hear rate and other vitals, helping him or her to identify the best fit regime for you.

All above are good but let’s go little forward and think a real life scenario that can happen,

You went to your favorite food joint and at point of sale you provided and identifiable information specific to you what will happen if everything is connected in true IOT scenario, Person on the sales counter will have lots of information and POS machine will not take your order if you have any disease and your doctor have said no without giving any details only a small bit information,

E.G. I want fries, Person at POS will say “Sorry Sir can’t, as your doctor has instructed that no high salt/ deep fried items for you, so please pick other item from menu” and then again you go for selecting other things.

Now imagine how much your privacy is at risk, for a total stranger knowing about your health.

Another Scenario of some card company calling you saying you are using at X POS service use of Y gives more incentive for shopping.

Do I really want world to know about it, may be not directly but through different ways.This is just one/two example however there can be many other one can think off.

Questions remains is IoT good and i will say definitely it is as per my view its adaptability will give far more benefits than risk caused.

However level of integration will tell how much personal the use can be called invasion of privacy and how much is actually required.

Note till the time I was writing this post there was no Single protocol that connect different devices and make it part of Global or Actual Term IOT, which I know of; hence integration can take long time.

Follow us @IoTCtrl | Join our Community

By Jean-Jacques Bernard and Ajit Jaokar

This set of blog posts is part of the book/course on Data Science for the Internet of Things. We welcome your comments at jjb at cantab dot net.  Jean-Jacques Bernard  has been a founding member of the Data Science for Internet of Things Course. Please email at ajit.jaokar at futuretext.com if you are interested in joining the course.

Introduction

The rapidly expanding domain of the Internet of Things (IoT) requires new  analytical tools. In a previous post by Ajit Jaokar, we addressed the need for a generic methodology for  IoT Analytics. Here, we expand on those ideas further.

Thus, the aim of this document is to capture the specific elements that make up Data Science in an Internet of Things (IoT) context. Ultimately, we will provide a high level methodology with key phases and activities with links to specific templates and contents for each of those activities.

We believe that one the best methodologies for undertaking Data Science is CRISP-DM. This seems to be the views of a majority of data scientists as the latest KDnuggets poll shows. Therefore, we have loosely based the methodology on CRISP-DM.

We have also linked the methodology to the technical framework proposed  [above] (Data Science for Internet of Things – A Problem Solving Methodology) which aims at providing a technical framework for solving IoT problems with Data Science.

The methodology we propose is divided in the following 4 phases:

1. Problem Definition
2. Preparation
3. Modelling
4. Continuous Improvement

We describe the first 3 phases in this post before describing the last phase and  the detailed activities and deliverables encompassing each phase in upcoming posts.

The relationships between the first three phases are presented in the figure below.

High level description of the methodology

Problem definition

This first phase is concerned with the understanding of the problem. It is important to define the terms here in the context of IoT.

By problem, we really mean something that needs to be solved, addressed or changed, either to suppress or revert a situation or create a new situation. Of course the end situation should be better than the initial situation. In the context of IoT, solving a problem means solving the initial problem and providing incremental feedback.

In any business context, to manage scarce resources, it is necessary to provide a business case for projects. Thus, it is important to define an IoT Analytics business case, which will provide a baseline understanding (i.e. measurement of the initial situation) of the problem through Key Performance Indicators (KPIs). In addition, the business case must provide a way to measure the impact of the project on the defined KPIs as well as a project timeline and project management methodology. The timeline and project management methodology should include deployment, a critical activity for large scale IoT Analytics projects.
The baseline and the measurement of the impact will be used to understand whether the IoT Analytics has reached its goals.

For instance, in the case of a Smart City project aiming at reducing road congestion, KPIs like number of congestion points and average duration of congestion at those points can be used to understand whether the project had a positive impact or not.

However, defining the problem and understanding how to measure it might be harder than it sounds as pointed here and here.

Preparation

The second phase of the methodology is concerned with data collection, preparation (i.e. cleaning) and exploration. However, in the context of IoT, the sources of data are more diverse than in other data science set-ups, and there are also elements of architecture to consider before starting more classic exploratory type of work.

Therefore, we believe there are three types of activities in this phase:

1. Define the data requirements
2. Select and design the IoT architecture
3. Collect, clean, explore and build data

 
And those three activities are to be conducted in an iterative manner, until the data build fits the problem we are trying to solve.

First we need to define the data needed to solve the problem defined previously as well as its characteristics. We need to do so in the context of the IoT vertical we are working in. Examples of IoT verticals include (not exhaustive):

• Smart homes
• Retail
• Healthcare
• Smart cities
• Energy
• Transportation
• Manufacturing (Industrie 4.0 or Industrial Internet)
• Wearables

The selection and design of the IoT architecture focuses on two parts: the network of devices and the processing infrastructure.

The first part is concerned with the set-up of the network of devices that will be used to measure and monitor some parameters of the environment of the problem. The design of this network is outside the scope of this article, but it is nonetheless important (for more information on this topic, you can refer to this article from Deloitte University Press). Some of the key considerations are:

• Availability & security
• Latency & timeliness
• Frequency
• Accuracy & reliability
• Dumb vs. smart devices

Those elements will determine some of the characteristics of the data that will be collected from the network of devices (in essence, this is a kind of meta-data). Those characteristics will be used to establish the processing infrastructure.

For instance, these are those characteristics which will help in choosing whether edge devices need to be used, whether event collectors are best suited, etc. For more information, see our article for an in-depth treatment of what potential processing infrastructures for IoT can be.

Then the final activities are the collection, cleaning and exploration of the data available. This is typical data analytics type of work, where the practitioner clean up the data available, and explore what are the properties of the data, etc. It is also a step where additional data can be build on the basis of the data available. However, this is also a step where it can become clear that the data provided by the IoT architecture is neither enough nor processed in a correct way.

This is why this phase is an iterative one, the learnings from the last step can be used to refine the IoT architecture until the data fits the problem to be solved.

Modelling

The modelling phase is the phase where a models are built and evaluated, both from a statistical standpoint and from a problem solving standpoint. It is composed of three activities:

1. Design model to solve the problem
2. Evaluate the model
3. Deploy the model and the architecture

Like for the preparation phase, those activities are to be conducted in an iterative manner, until the model:

• Is statistically sound;
• Shows potential to solve the problem (i.e. impact on KPIs defined in the problem definition phase).

In this phase, the data scientist will choose among different types of algorithms, depending on the problem to solve and build models using those.

As described in the previous article, many algorithms and techniques are applicable, such as time series analytics, complex event processing (CEP) or deep learning. An important element, linked to the activities from the previous phase, is where will the analytics be applied. While this should be part of the design of the IoT architecture, this while guide the choice of algorithm to apply. Indeed, we can apply analytics at:

• The device (if we use smart devices)
• The edge
• The data lake / Cloud
• Etc.

In addition, the type of analytics will depend on the type of processing we are focusing on: batching vs. streaming.

When the model has been designed, then comes the evaluation activities, which should first evaluate the model using classic statistical and data science techniques: using training, validation and testing datasets, minimizing bias and variance and trading-off precision and recall.

Then, the model should be evaluated from a business point of view: does it improve the KPIs that were set during the problem definition phase? It might not be obvious to measure the improvement until the model is deployed, thus, it is important to keep an improvement loop over this phase and the previous. If the model does not improve the KPIs defined in the problem definition phase, then it is necessary to rework from the preparation phase since some of the assumptions underlying the data may be wrong.

When the model is considered as sound and solves the problem it was designed to solve, it is time to deployed it together with its IoT Architecture. The deployment of the architecture and model is a project in itself and should come with its own project management structure and timeline, as defined in the problem definition phase.

In upcoming posts, we will present the continuous improvement phase and explore the detailed activities and deliverables of each of the phases presented here.

To conclude on this post, we welcome your comments. Please email ajit.jaokar at futuretext.com if you are interested in joining the course.

Follow us @IoTCtrl | Join our Community

Guest blog post by Ajit Jaokar

Many thanks for the retweets and feedback on Part one of this blog 

A methodology for solving problems with DataScience for Internet of Things - Part One

Here is Part Two

Here we extend the discussion and also suggest a practical (and open) way to create a way forward

To recap, lets keep in mind the big picture and after considering Streaming in the previous section, let us consider more techniques like Edge Processing etc

Edge Processing

Many vendors like Cisco and Intel are proponents of Edge Processing  (also  called  Edge  computing).  The  main  idea behind Edge Computing is to push processing away from the core and towards the Edge of the network. For IoT, that means pushing processing towards the sensors or a gateway. This enables data to be initially processed at the Edge device possibly enabling smaller datasets sent to the core. Devices at the Edge may not be continuously connected to the network. Hence, these devices may need a copy of the master data/reference data for processing in an offline format. Edge devices may also include other features like:

•    Apply rules and workflow against that data

•    Take action as needed

•    Filter and cleanse the data

•    Store local data for local use

•    Enhance security

•    Provide governance admin controls

IoT analytics techniques applied at the Data Lake

Data Lakes

The concept of a Data Lake is similar to that of a Data warehouse or a Data Mart. In this context, we see a Data Lake as a repository for data from different IoT sources. A Data Lake is driven by the Hadoop platform. This means, Data in a Data lake is preserved in its raw format. Unlike a Data Warehouse, Data in a Data Lake is not pre-categorised. From an analytics perspective, Data Lakes are relevant in the following ways:

• We could monitor the stream of data arriving in the lake for specific events or could co-relate different streams. Both of these tasks use Complex event processing (CEP). CEP could also apply to Data when it is stored in the lake to extract broad, historical perspectives.
• Similarly, Deep learning and other techniques could be applied to IoT datasets in the Data Lake when the Data  is ‘at rest’. We describe these below.

ETL (Extract Transform and Load)

Companies like Pentaho are applying ETL techniques to IoT data

Deep learning

Some deep learning techniques could apply to IoT datasets. If you consider images and video as sensor data, then we could apply various convolutional neural network techniques to this data.

It gets more interesting when we consider RNNs(Recurrent Neural Networks)  and Reinforcement learning. For example – Reinforcement learning and time series – Brandon Rohrer How to turn your house robot into a robot – Answering the challenge – a new reinforcement learning robot

Over time, we will see far more complex options – for example for Self driving cars  and the use of Recurrent neural networks (mobileeye)

Some more interesting links for Deep Learning and IoT:

• http://stats.stackexchange.com/questions/8000/proper-way-of-using-recurrent-neural-network-for-time-series-analysis
• Noisy Time Series Prediction using a Recurrent Neural Network and Grammatical Inference –https://clgiles.ist.psu.edu/papers/MLJ-finance.pdf
• https://www.quora.com/Can-recurrent-neural-networks-with-LSTM-be-used-for-time-series-prediction
• Time series forecasting with recurrent neural networks http://www.neural-forecasting-competition.com/downloads/NN3/methods/24-NN3_Mahmoud_Abunasr_competition_complete.pdf
• and an article by Sibhanjan Das and me on Deep Learning and IoT with H2O @kdnuggets

Optimization

Systems level optimization and process level optimization for IoT is another complex area where we are doing work. Some links for this

• http://blog.tsia.com/blog/how-iot-process-optimization-can-improve-customer-outcomes
• http://www.intel.co.uk/content/dam/www/public/us/en/documents/white-papers/industrial-optimizing-manufacturing-with-iot-paper.pdf

 Visualization

Visualization is necessary for analytics in general and IoT analytics is no exception

Here are some links

• iotdataviz and also
•  Nick Bearman’s work on R and Spatial analytics

NOSQL databases

NoSQL databases today offer a great way to implement IoT analytics. For instance,

Apache Cassandra for IoT

MongoDB and IoT tutorial

Other  IoT analytic techniques

In this section, I list some IoT  technologies where we could implement analytics

• spatial analytics for IoT  – ex from ESRI
• Implementation of IoT analytics on DSPs
• Augmented reality – ex in conjunction with Pokemon Go
• GPUs – there is some extensive work done by Nvidia in this space especially in relation to Deep learning
• I am also following Apache NiFi with great interest with features like  Birectional flow of data, perimeter of control changes and data provenance

A Methodology to solve Data Science for IoT problems

We started off with the question: Which points could you apply analytics to the IoT ecosystem and what are the implications? But behind this work is a broader question:  Could we formulate a methodology to solve Data Science for IoT problems?  I am exploring this question as part of my teaching both online and at Oxford University along with Jean-Jacques Bernard.

Here is more on our thinking:

• CRISP-DM is a Data mining process methodology used in analytics.  More on CRISP-DM HERE and HERE(pdf documents).
• From a business perspective (top down),we can extend CRISP-DM to incorporate the understanding of the IoT domain i.e. add domain specific features.  This includes understanding the business impact, handling high volumes of IoT data, understanding the nature of Data coming from various IoT devices etc
• From an implementation perspective(bottom up),  once we have an understanding of the Data and the business processes, for each IoT vertical : We first find the analytics (what is being measured, optimized etc). Then find the data needed for those analytics. Then we provide examples of that implementation using code. Extending CRISP-DM to an implementation methodology, we could have Process(workflow), templates,  code, use cases, Data etc
• For implementation in R, we are looking to initially use Open source R and Spark and the  h2o.ai  API
• Make the methodology practical by considering high volume IoT data problems, Project management methodologies for IoT,  IoT analytics best practices etc
• Most importantly, we will be Open  and Open Sourced
• There are some parallels with this thinking with Big Data Business maturity model index and to Systems thinking – my favourite systems thinking text is An Introduction to General Systems Thinking – Gerald M. Weinberg
• When used in teaching for my course, it has parallels to the Concept-context pedagogy (pdf)  where the concepts are tied to the practise in terms of Projects which take center stage

Conclusion

We started off with the question:  At which points could you apply analytics to the IoT ecosystem and what are the implications? And extended this to a broader question:  Could we formulate a methodology to solve Data Science for IoT problems?  The above is comprehensive but not absolute. For example, you can implement deep learning algorithms on mobile devices (Qualcomm snapdragon machine learning development kit for mobile mobile devices).  So, even as I write it, I can think of exceptions!

This article is part of my forthcoming book on Data Science for IoT and also the courses I teach

Welcome your comments.  Please email me at ajit.jaokar at futuretext.com  - Email me also for a pdf version if you are interested. If you want to be a part of my course please see the testimonials at Data Science for Internet of Things Course.  

Finally, I will syndicate more sections of the book on Data Science Central. Stay tuned! 

Follow us @IoTCtrl | Join our Community

Guest blog post by Manoj

It has been estimated that the Internet of Things (IoT) will contain 26 billion devices by 2020 (according to Gartner, Inc.), while Cisco’s CEO puts the commercial opportunity from these devices to reach $19 trillion. But behind the glorious financial opportunities is a new community of data, and a complex testing challenge to help support those devices. Usually, IoT is categorized based on markets like “wearable technology” and "smart home" instead of broken down by how the data itself is handled. I would like to talk about these devices from the IoT data-handling perspective, while highlighting some of the testing challenges of each type.

The simplest IoT devices have a highest IoT function of making one-way service requests while monitoring themselves. These range from home appliances and propane tanks, to commercial vending machines, to urban devices like porta-potties and garbage cans. The data flows only outward, with “Help me!” messages like “I need to be filled”, “I need to be emptied” or “I need to be serviced because of the following diagnostics code”. Deployment of these devices should include a confirmation of valid Internet connection, although a phone signal may also suffice to SMS an SOS. Testing should confirm that the expected message is sent to the right destination when the service condition is met. No message should be sent when no servicing is needed (unless a periodic ping of existence is required).  Since the number of different causal events and corresponding messages is limited, it is relatively easy to achieve full coverage from a testing perspective. We do not expect split-second response, so performance testing is not so important yet.

The next type of device also involves one-way export of monitored data, but functions as a real-time reporter of more complex data in a much more ramped-up fashion. This often includes medical data (like heart rate, glucose level, or blood pressure) and/or GPS/motion tracking. GPS/motion tracking may include: athletic analyzers (often reporting speed or swing data), fitness trackers, and vehicle trackers (which may include OBD-II non-GPS data). Due to mobility, weak or lost connection is a greater concern when testing. The device may store data in case transmission is interrupted, for manual download at a later time. Some devices – especially medical ones – may need to issue service calls too (low battery, low medicine, etc.) which may be duplicated visibly on the IoT device. Testing is concerned with the accuracy and completeness of the data received that was generated from the IoT device, as well as high data tolerances on the detection side. Testing coverage is more complicated, given reception issues for in-the-wild in-transit locations, and the complexity of the data details that can be transmitted, and the greater likelihood of security being needed on the broadcast data. Non-data operational logistics of the devices may also require testing.

The next step up the evolutionary ladder is interactive devices.  We have now achieved two-way communication of data! This allows a level of programmability, although we may no longer need to transmit a constant flood of data to be monitored. These smart devices include pet feeders, DVR’s, home security, watering systems, lighting, smart appliance, toys, game systems and of course tablets and cellphones.  With two-way data flow, there may now be greater security concerns over unauthorized access to the device and its data. Also, now that there can be input data, we need to test for and respond to potentially incomplete or nonsensical input data. We still have the same concerns we had for one-way real-time monitors, but now we have an added focus on response time performance, including stress-testing heavy 2-way flow of data possibly while specific tasks are fighting for processor time on the IoT device.

Lastly, we have devices that do everything listed above plus possess a high degree of artificial intelligence, an intelligence that must be tested, especially where safety is concerned. This includes drones, smart vehicles, and smart factories. Real-time incoming data monitoring via sensors is used to make subtle adjustments or abrupt alteration, safely reacting and maneuvering to avoid potential hazards while monitoring the success or failure of the device’s actions. In other words, interpretation of current data directly impacts future data. In addition to the usual connectivity concerns, forms of transportation include weather conditions as an important testing variable, as slipperiness affects vehicular braking, fog hampers visibility, and heavy precipitation limits movement for lightweight drones. Smart factories symbolize what big data can achieve, and involve many machines interacting and transferring data in progressively more wireless ways, and require interoperability testing to ensure that data is recognized correctly at all levels.

As you can see, the Internet of Things has introduced us to different ways of managing and communicating data of different complexities.  The safeguards and testing become more complex as the device abilities become more complex, which is tied to what that data needs to accomplish.

Follow us @IoTCtrl | Join our Community