Bronco Webinar 800x100 1
Semiconductor Expert Forum

Posted on

NTSB Entry Raises the Stakes of Tesla Probe

NTSB Entry Raises the Stakes of Tesla Probe
by Roger C. Lanctot on 07-10-2016 at 12:00 pm
Categories: Automotive

The National Transportation Safety Board’s entry into the investigation of the first fatal crash of a Tesla Model S is a monumental turning point in the autonomous driving movement. While long-time observers of the NTSB note that it only gets involved in investigations where broader implications exist, the agency’s interest also reflects the fact that the National Highway Traffic Safety Administration lacks the technical ability to properly investigate a crash cause that is likely tied to a software failure.

As in the case of Toyota’s unintended acceleration fatalities, recalls and penalties, software is chiefly implicated in the fatal Tesla crash in Florida. In the Toyota case, NHTSA turned to the National Aeronautic and Space Administration for help and NASA ultimately turned to outside experts who criticized what they described as Toyota’s “spaghetti code.”

The source of the unintended acceleration in the Toyota Prius remains unresolved, but the primary learning from the experience was the realization of the investigatory limitations of the automotive industry’s primary regulatory agency. Those limitations, a legacy of the agency’s reduction in size going back to the Reagan Administration, remain uncorrected.

As a result, NHTSA lacks the fundamental expertise necessary not only to investigate the crashes of autonomous vehicles but also to evaluate the performance of these vehicles or even to properly set guidelines. This is a big problem for the industry and for the motoring public leaving individual state authorities in the awkward position of blindly cobbling together rules and guidelines of their own for the operation of self-driving vehicles on local and interstate highways.

Tesla and Google have more or less been left in the position of regulating and policing themselves. This is less of a problem for Google since its vehicles are not made available to the general public and generally operate at low speeds on local roads. It’s a different matter for Tesla Motors.

Ironically, Google lobbied the California Department of Motor Vehicles to leave autonomous vehicle regulatory oversight and guidance to Federal authorities in the form of NHTSA. During last year’s California DMV hearings on the subject, Google lobbyists and executives (including at least one former NHTSA executive, Ron Medford) implored the California DMV to relinquish its authority over Google’s local on-road testing activities.

Google’s argument before the California DMV was that the state agency was incapable of comprehending let alone evaluating the self-driving software Google was developing and deploying. It is clear that Google’s pleas were a cynical play to shift control to an agency with which it felt it had greater influence – knowing all along that NHTSA, too, lacked the necessary expertise and resources.

But the cynicism of Tesla’s CEO, Elon Musk, puts Google’s cynicism to shame. Musk has opened up his own traffic court serving as judge, jury, witness and prosecution. With each new Model S crash, Musk is quick to provide his assessment of fault – nearly universally lying with the driver – absolving himself and his company of responsibility.

The fatal crash in Florida is the first instance of Tesla acknowledging a potential flaw in its software and sensing architecture. Still, Musk fell back on the various caveats for use of the autopilot system including hands on the wheel etc. intended to release Tesla from any responsibility.

States such as California have begun insisting on full disclosure of self-driving car crash data – especially in the case of Google. Tesla is technically not offering a self-driving car, but state and Federal authorities may soon begin insisting on this same kind of sharing of crash data.

Transportation network companies (TNCs) such as Lyft and Uber, operating like Google and Tesla, outside the normal regulatory bounds are also being asked to disclose data about their drivers and crashes and other incidents. It seems that the battle for the next generation of transportation technology is evolving into a battle for data.

The manner in which the Tesla fatal crash has exposed the software blindspot of NHTSA has wider implications for the government’s role in redefining transportation safety. Safety in transportation is increasingly being determined by software systems. NTSB’s decision to enter NHTSA’s investigation suggests that NHTSA itself may not be up to the very task it has given itself – of promoting collision avoidance and autonomous driving.

Ultimately, this calls into question its plans to mandate the implementation of vehicle-to-vehicle wireless communications for the purpose of crash avoidance as well as its ability to comprehend, provide guidelines for and regulate the process of self-driving software development and sensor fusion. The arrival of the NTSB on the Tesla crash scene is an acknowledgement within the regulatory community that NHTSA is out of its depth, unequal to the task.

Until such time as the NTSB, NHTSA or NASA can sort out which agency has the scope or expertise to oversee autonomous vehicle development and deployment we are likely to see an ongoing and expanding free-for-all on U.S. highways. Such a free-for-all may lead to more fatalities, technological advancement or simple chaos – maybe all of the above.

Someone in Washington needs to sort out government’s role and properly fund the relevant agencies such that progress is successfully and safely achieved. The alternative will be a widespread call from safety advocates that all autonomous driving testing, development and deployment cease. Since autonomous driving technology is intended to save us all from ourselves, that can’t be the outcome we want to see.

Roger C. Lanctot is Associate Director in the Global Automotive Practice at Strategy Analytics. More details about Strategy Analytics can be found here: https://www.strategyanalytics.com/access-services/automotive#.VuGdXfkrKUk

Posted on

Hacking Your Way Across the Chasm

Hacking Your Way Across the Chasm
by Michael Tanner on 07-10-2016 at 7:00 am
Categories: General

Geoffrey Moore’s “Crossing the Chasm” remains one of the most useful and widely read business books within the high-technology business. I say this not only because I spent a decade of my life working with Geoffrey and the rest of The Chasm Group as a consultant, but also because to this day, I have yet to find another book that is so well written and that communicates the basics of market behavior in such a simple and straightforward way. And, while Crossing the Chasm was originally born in 1991 from what were mostly B2B marketing challenges at the time, the market adoption and market development ideas within remain fundamental, proving themselves applicable across all forms of new innovations.

“Growth-hacking” is a more recent concept. Coined by Sean Ellis1 in 2010, growth hacking involves using technical approaches and analytics to test and then optimize marketing activities, getting around traditional approaches at a substantially lower cost. ( Note: Neil Patel and Bronson Taylor wrote a great on-line reference to growth hacking — click here). Twenty years ago “hackers” were thought of as thieves who broke into networks and computers to steal stuff. Today, the term “hacker” is a more general piece of jargon that’s used to describe someone who uses innovative technical or analytical techniques to overcome barriers.

The two ideas: growth-hacking, and chasm-crossing, feel at odds at fist glance. Where growth hacking involves a series of tactical approaches that broadly test what works, chasm-crossing involves making strategic and more methodical choices about where, what, and how to sell, and then going after a single market beach head with a vengeance. But, growth-hacking can also have more than just tactical objectives. With a little foresight, these two concepts can intersect in some interesting ways that were not possible when Crossing the Chasm was originally authored.

In my experience, one of the big hurdles that teams face when making decisions related to chasm-crossing is the lack of comfort that comes from having insufficient facts. Growth hacking involves a process of trial and error to quickly identify facts about what works and what doesn’t. The challenge is to organize the growth-hacking tactics in a way that produces the types of information you need.

For example, the criteria for identifying potential chasm-crossing segments typically involves quantifying target segment attractiveness based upon a few things:

[LIST=1]

  • The availability and access to a well-funded set of buyers
  • The urgency (rather than importance) of the need
  • The degree to which the required ‘whole product’ is complete, and
  • The referral leverage that one set of customers might have into another segment

    Modern marketing and sales automation tools can give you some real insight here. For example: by simply identifying visitors in an intelligent way through your website and through social media you can help identify hot spots through the purchase process, starting with visits. By reverse IP-lookup, you can learn the names of companies visiting your website. If you have a general idea of the titles that might be good prospective buyers, including the line of business buyers and the infrastructure buyers, you can provide them to an outside service who will in-turn deliver a direct marketing list to you based upon your visitors, whether or not they’ve actually given you their names and email addresses. Over time you can then develop specific content sets that can test different messages in order to see which value propositions return the highest open rates, click through rates, trials, and eventual purchases by segment.

    Using targeted content marketing or feeds from twitter and blog posts, you might also hypothesize prospective solutions to see which resonate the best across different segments, which titles in your list spend the most time looking through content, and which stimulate actual content downloads or product evaluation. Most lead scoring systems within marketing automation tools can be setup to automate this effort. You can then use the analytics generated along with actual sales data to help answer the key questions above. Moreover, if you manage to create some sort of referral mechanism, either built into the process, or built into the actual product, you can methodically track where these referral references are coming from and going to.

    There are challenges to this approach too. First, if you are an early stage business there is a small investment into marketing automation tools and a learning curve. You’ll need to have and work with your outbound direct marketing team to setup analytics that help answer the key questions. Second, you’ll need to make sure that you differentiate between the early adopter tire-kickers who spend lots of time evaluating without consummating a real deal, and the pragmatic buyer types who have the ability to drive lifetime value. A careful study of titles, initial vs. follow-on purchases, and the up-sells achieved by segment can help differentiate these two types of buyers.

    Third, while you can’t understand solution gaps completely with such a low-touch approach, you certainly can glean insight from targeted content marketing that features prospective solution elements. The key is to have some call to action that can be measured. Product managers and engineers who decide solution priorities may not be used to working directly with those who execute tactical marketing programs in this way, so here’s where you may need a real “growth hacker” skill-set on the team who can span and integrate both cultures.

    Fourth (and finally), you must be very sure that you do not confuse correlation withcausation. But, all challenges aside, by putting just a few processes and tools in place early on you can bring far more factual insights to bear for the chasm-crossing discussion than could have been done cost-effectively in the past. Today, facts can actually be quite plentiful with a little planning. The challenge is now more about sifting through what is relevant to arrive at real insight.

    1. [Ellis, Sean (June 26, 2010). “Find a Growth Hacker for Your Startup” Startup-marketing.com ]↩

    • Posted on

    Are Smart things making us smarter?

    Are Smart things making us smarter?
    by Prakash Mohapatra on 07-08-2016 at 12:00 pm
    Categories: Automotive, IoT

    Nowadays, we don’t have to learn how to drive a car well because there are systems (automated braking, monitoring, etc.) in the car that is taking care of many things without our knowledge. We don’t have to remember whether we have switched off the lights before leaving the house. The smart home automation system will switch off the lights after detecting no sound or activity for some time.

    Self-driving cars are the way to go in future. In future, people don’t have to hire a chauffeur. You may just board on the car and tell it where to go. The car shall use GPS to find the optimal route and take you there, cruising through the traffic. You can get down from the car, and then the car will find an empty parking spot for itself. While you are leaving, call the car from your smartphone to be in the entrance in 5 mins. When you reach out, the car will be waiting for you with the rear door open and playing your favourite music, setting the appropriate temperature for your comfort. Awesome!!

    Not only that, come next the smart fridges. You can check whether there are beer cans in the fridge by sending a message to the fridge. On detecting there are are only few beer cans left, the fridge can order beer cans from an online retailer. The beers shall be delivered in your home without your involvement. Maybe we can term this as “Self-Replenishing Fridges”. The advertising gimmick for this product would be “The fridge that never becomes empty”.

    I may sound like a guy who loath technology and who doesn’t want technology to enhance our lifestyle. However, I believe that I am looking at things with a more conservative viewpoint. In my view, most companies are struck in a red ocean, in which the only focus of companies is to create competitive advantage. The companies’ pursuit of doing things better than the competitors tend to expand the chasm between the technology and customer utility.

    I agree that future of technology is all about convergence and integration, i.e, the seamless migration from one environment to another for the end user. In a typical day, people spend majority of their time in three environments: home, office and travel. I believe technology is about integrating all three environments together, so that when a user moves from one environment to another, the devices are aware of the movements (contextual awareness) and take appropriate actions.

    For e.g, when I move from my home network to my office network, my smartphone shall hide my personal profile and show my official profile. I believe consumer electronics giants such as Apple, Samsung have intentions of dominating in each of these environments. Apple has already penetrated our home with iPod, iPhone and iPad. With the tremendous success in the home segment, Apple extended its offerings to penetrate the other two environments with enterprise (BYOD offerings) and automotive (IoS in car).

    Once a firm has dominant position in one environment, it is easy to offer complementary offerings to penetrate other environments. Similarly, Samsung is also attempting to penetrate the enterprise segment with its Samsung Knox offering. I may have missed to notice any tangible drive by Samsung in the automotive segment. However, the eco-system of Android shall work out in favour of Samsung as most of its smartphones are based on Android.

    With the growing prominence of automotive apps, Samsung is in a favorable position to also penetrate into vehicle with its android smartphones. I believe rather than embedded intelligence in vehicles, it is more beneficial for both the automotive OEMs and end users to extend the capabilities of the smartphones by using automotive apps. This strategy decouples the mismatch in the product life cycle of automobiles and consumer electronics technology. However, many applications in ADAS will need embedded intelligence.

    It is obvious that tech firms will keep on innovating and create new trends to penetrate in each of these environments. It is also infallible that we consumers will become prey of these tech trends and depend more on these machines rather than our brains. Then I really question whether these smart things are really making us smarter or just offer us an illusion that we are becoming smart.

    What do you think?

    Posted on

    Artificial Intelligence is Everything!

    Artificial Intelligence is Everything!
    by Daniel Nenni on 07-08-2016 at 7:00 am
    Categories: AI, Ceva, Events, IP

    My first brush with AI was a LISP class for my undergraduate degree. LISP, originated from MIT in 1958, was the language of choice for AI research and spawned a new class of computer hardware called LISP Machines in the 1980s. My first personal experience with AI was the HAL 9000 system from the 2001 Stanley Kubrik movie Space Odyssey. Today I have my own personal AI systems (Amazon echo and Apple Siri) that I rely on every day.

    Most people don’t realize this but AI is already an active part of our daily lives: in our cars, in our phones, and in our homes. In fact, in regards to our cars, our lives will literally depend on AI, absolutely. I also believe the collective intelligence of the human race is on a downward trend so we will need all of the help we can get!

    The challenge of AI of course is compute power which is good news for the semiconductor industry because that “need for speed” will consume leading edge silicon like there is no tomorrow. The fabless semiconductor ecosystem is already gearing up for this deep learning experience on embedded systems and this webinar is a quick example:


    Summary
    As Artificial Intelligence (AI) marches into almost every aspects of our lives, one of the major challenges is bringing this intelligence to small, low-power devices. This requires embedded platforms that can deliver extremely-high Neural Network performance with very low power consumption. However, that’s still not enough.

    Machine Learning developers need a quick and automated way to convert and execute their pre-trained networks on such embedded platforms. In this session, we will discuss and demonstrate tools that complete this task within few minutes, instead of spending months on hand porting and optimizations.

    REGISTER HERE

    Join CEVA experts to hear about:

    • Overview of the leading deep learning frameworks, including Caffe and TensorFlow
    • Various topologies of neural networks, including MIMO, FCN, MLPL
    • Overview of most common neural networks such as Alexnet, VGG, GoogLeNet, ResNet, SegNet
    • Challenges in porting neural networks to embedded platforms
    • CEVA “Push button” conversion approach from pre-trained networks to real-time optimized
    • Programmer Flow for CNN Acceleration

    Target Audience:
    Computer vision engineers, Deep learning researchers, Project managers, marketing experts and others interested in embedded vision and machine learning.

    Speakers:
    Liran Bar, Director of Product Marketing, Imaging & Vision, CEVA
    Erez Natan, Neural Network Team Leader, Imaging & Vision, CEVA



    About CEVA, Inc.
    CEVA is the leading licensor of signal processing IP for a smarter, connected world. We partner with semiconductor companies and OEMs worldwide to create power-efficient, intelligent and connected devices for a range of end markets, including mobile, consumer, automotive, industrial and IoT. Our ultra-low-power IPs for vision, audio, communications and connectivity include comprehensive DSP-based platforms for LTE/LTE-A/5G baseband processing in handsets, infrastructure and machine-to-machine devices, computer vision and computational photography for any camera-enabled device, audio/voice/speech and ultra-low power always-on/sensing applications for multiple IoT markets. For connectivity, we offer the industry’s most widely adopted IPs for Bluetooth (Smart and Smart Ready), Wi-Fi (802.11 a/b/g/n/ac up to 4×4) and serial storage (SATA and SAS). Visit us at www.ceva-dsp.com and follow us on Twitter, YouTube and LinkedIn.

    Posted on

    IBM z13 Helps Avoid Costly Data Breaches

    IBM z13 Helps Avoid Costly Data Breaches
    by Alan Radding on 07-07-2016 at 12:00 pm
    Categories: Security

    A global study sponsored by IBM and conducted by the Ponemon Institute found that the average cost of a data breach for companies surveyed has grown to $4 million, representing a 29 percent increase since 2013. With cybersecurity incidents continuing to increase with 64% more security incidents in 2015 than in 2014 the costs are poised to grow.


    z13–world’s most secure system


    The z13, at least, is one way to keep security costs down. It comes with a cryptographic processor unit available on every core, enabled as a no-charge feature. It also provides EAL5+ support, a regulatory certification for LPARS, which verifies the separation of partitions to further improve security along with a dozen or so other built-in security features for the z13. For a full list of z13 security features click here. There also is a Redbook, Ultimate Security with the IBM z13 here. A midsize z, the z13s brings the benefits of mainframe security and mainframe computing to smaller organizations. You read about the z13s here on DancingDinosaur this past February.


    As security threats become more complex, the researchers noted, the cost to companies continues to rise. For example, the study found that companies lose $158 per compromised record. Breaches in highly regulated industries were even more costly, with healthcare reaching $355 per record – a full $100 more than in 2013. And the number of records involved can run from the thousands to the millions.


    Wow, why so costly? The researchers try to answer that too: leveraging an incident response team was the single biggest factor associated with reducing the cost of a data breach – saving companies nearly $400,000 on average (or $16 per record). In fact, response activities like incident forensics, communications, legal expenditures and regulatory mandates account for 59 percent of the cost of a data breach. Part of these high costs may be linked to the fact that 70 percent of U.S. security executives report they don’t even have incident response plans in place.


    The process of responding to a breach is extremely complex and time consuming if not properly planned for. As described by the researchers, the process of responding to a breach consists of a minimum of four steps. Among the specified steps, a company must:

    • Work with IT or outside security experts to quickly identify the source of the breach and stop any more data leakage
    • Disclose the breach to the appropriate government/regulatory officials, meeting specific deadlines to avoid potential fines
    • Communicate the breach with customers, partners, and stakeholders
    • Set up any necessary hotline support and credit monitoring services for affected customers

    And not even included in the researchers’ list are tasks like inventorying and identifying the data records that have been corrupted or destroyed, remediating the damaged data, and validating it against the last known clean backup copy. Am surprised the costs aren’t even higher. Let’s not even talk about the PR damage or loss of customer goodwill. Now, aren’t you glad you have a z13?


    That’s not even the worst of it. The study also found the longer it takes to detect and contain a data breach, the more costly it becomes to resolve. While breaches that were identified in less than 100 days cost companies an average of $3.23 million, breaches that were found after the 100-day mark cost over $1 million more on average ($4.38 million). The average time to identify a breach in the study was estimated at 201 days and the average time to contain a breach was estimated at 70 days. The cost of a z13 or even the lower cost z13s could justify itself by averting just one data breach.


    The researchers also found that companies with predefined Business Continuity Management (BCM) processes in place found and contained breaches more quickly, discovering breaches 52 days earlier and containing them 36 days faster than companies without BCM. Still, the cheapest solution is to avert breaches in the first place.


    Not surprisingly, IBM is targeting the incident response business as an up and coming profit center. The company increased its investment in the Incident response market with the recent acquisition of Resilient Systems, which just came out with an updated version that graphically displays the relationships between Indicators of Compromise (IOCs) and incidents in an organization’s environment. But the z13 is probably a better investment if you want to avoid data breaches in the first place.


    Surprisingly, sometimes your blogger is presented as a mainframe guru. Find the latest here.


    DancingDinosaur is Alan Radding, a veteran information technology analyst writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

    Posted on

    Latest Pinpoint Release Tackles DRC and Trend Lines

    Latest Pinpoint Release Tackles DRC and Trend Lines
    by Don Dingee on 07-06-2016 at 4:00 pm
    Categories: Dassault Systemes, EDA

    After reading previous SemiWiki coverage on Dassault Systèmes and their ENOVIA Pinpoint solution, one big item seemed missing: how does this thing actually work? With all due respect to our other bloggers who covered when Dassault Systèmes acquired Pinpoint from Tuscany Design Automation, why Qualcomm is using Pinpoint, and what it does for Continue reading “Latest Pinpoint Release Tackles DRC and Trend Lines”

    Posted on

    IoT Tutorial: Chapter7 – IoT data and IoT-BigData Convergence

    IoT Tutorial: Chapter7 – IoT data and IoT-BigData Convergence
    by John Soldatos on 07-06-2016 at 12:00 pm
    Categories: IoT

    Introduction to IoT Data and their Characteristics
    Most IoT applications up to date involve the collection and processing of IoT data i.e. data stemming from IoT sources such as sensors, wearables and other internet connected devices. In the majority of cases the business benefits of an IoT application stem from the processing of IoT data. Typical examples include:

    • Security applications involving processing of information from multiple cameras deployed in urban areas in order to timely identify security events.
    • Urban mobility applications relying on the processing of data from traffic sensors in order to identify and alleviate traffic congestion.
    • Healthcare applications involving the collection and processing of behavioral information of a subject (based on streams from cameras, accelerometers and wearables) towards identifying lifestyle patterns.
    • Sports and fitness applications processing information from wearables in order to track statistics and training parameters for athletes.
    • Smart city applications entailing collection and processing of information from smart meters towards energy management at various timescales.

    As evident from the above examples, data-intensive IoT applications involved processing of data from various sensors and devices. In several cases these applications can also combine data from other sources such as open data sources and social media. Furthermore, these IoT applications process IoT data in various timescales ranging from real-time processing for operational applications (e.g., traffic rerouting in case of congestion) to data processing at a weekly, monthly or yearly basis as part of strategic level applications (e.g., transport planning).

    Apart from applications (such as the above-listed ones), whose business logic is the IoT data processing itself, there are also other IoT applications which focus on actuation and real-time control rather than on providing data to their end-users. Typical examples of such applications including for example CPS systems controlling robots in manufacturing plants or actuators in connected cars applications. Despite their emphasis on control (rather than data provision) these applications are in most cases also driven by IoT data processing, since decisions are usually based on the collection and analysis of IoT data from different data sources.

    IoT data feature certain characteristics, which distinguish them radically from other types of data sources and respective applications (e.g., classical transaction applications). These characteristics include their streaming and real-time nature, their spatial and temporal characteristics, as well as their special security and privacy requirements (e.g., in cases where collection and processing of personal data are involved). The special characteristics and related challenges for IoT data processing applications can be listed as follows:

    • Heterogeneity of IoT data streams: IoT data streams tend to be multi-modal and heterogeneous in terms of their formats, semantics and velocities. Hence, IoT analytics applications expose typically variety and veracity. BigData technologies provide the means for dealing with this heterogeneity in the scope of operationalized applications.
    • Varying data quality: Several IoT streams are noisy and incomplete, which creates uncertainty in the scope of IoT analytics applications. Statistical and probabilistic approaches must be therefore employed in order to take into account the noisy nature of IoT data streams, especially in cases where they stem from unreliable sensors.
    • Real-time nature of IoT datasets: IoT streams feature high velocities and for several application must be processes nearly in real-time. Hence, IoT analytics can greatly benefit from data streaming platforms, which are part of the BigData ecosystem.
    • Time and location dependencies of IoT streams: IoT data come with temporal and spatial information, which is directly associated with their business value in a given application context. Hence, IoT analytics applications must in several cases process data in a timely fashion and from proper location. Cloud computing techniques (including edge computing architectures) can greatly facilitate timely processing of information from given locations in the scope of large scale deployments.
    • Privacy and security sensitivity: IoT data are typically associated with stringent security requirements and privacy sensitivities, especially in the case of IoT applications that involve the collection and processing of personal data.
    • Data bias: As in the majority of data mining problems, IoT datasets can lead to biased processing and hence a thorough understanding and scrutiny of both training and test datasets is required prior to their operationalized deployment. To this end, classical data mining techniques can also be applied in the case of IoT applications.

    IoT Data-Intensive Applications Lifecycle
    The development of IoT applications entails the following activities, which are usually combined towards developing and deploying non-trivial IoT data applications:

    • IoT Data Collection, including interfacing to IoT sources (i.e. internet connected devices) and enrichment of these data with appropriate contextual metadata, such as location information and timestamps. As already outlined, the collection process needs typically to deal with the heterogeneity of the IoT data sources and their data streams, including heterogeneity of interfaces to data sources and of data formats.
    • IoT Data Validation, including validation of the format and source of origin of the data. The process includes also the validation of their integrity, accuracy and consistency.
    • IoT Data Semantic Unification and Interoperability, which deal with the unification/homogenization of the semantics of IoT streams stemming from different sources, as a prerequisite for their unified processing.
    • IoT Data Structuring and Storage, which involves the persistence of validated and interoperable data in an appropriate database such streaming database, object database or even graph database.
    • IoT Data Analysis,which deals with the application of data mining and machine learning techniques (e.g., regression, neural networks, decision tree, clustering) towards transforming IoT data streams to actionable knowledge.
    • Deployment of IoT analytics algorithms,which involves the actual deployment and operationalization of machine learning and data mining schemes for data analytics.
    • IoT data visualization,which emphasizes the presentation of IoT data in a graphical format, including their browsing across the temporal and spatial dimensions of the IoT datasets.
    • IoT data repurposing and reuse,which entails access to IoT datasets towards reusing them across different applications.

    IoT and BigData Convergence
    The above-listed IoT data processing challenges and activities are very closely related to the wave of BigData technologies. Indeed, IoT data are characterized by the Vs that are commonly associated with BigData technologies. In particular, BigData systems refer to data processing and management systems, which feature one or more of the following characteristics (Vs):

    • Volume: Very high data volumes, beyond those that can be handled by state-of-the-art data management systems.
    • Velocity: Data streams with very high ingestion rates, which cannot be handled by state-of-the-art systems and databases.
    • Variety: Data featuring extreme heterogeneity in terms of velocities, formats and semantics.
    • Veracity: Data that are characterized by uncertainty and unreliability.

    IoT analytics applications are typically characterized by:

    • High-data volumes, since in several cases they have to collect and process streaming information from thousands of sensors.
    • High-velocity streams, since they usually involve streaming data that are collected and in several cases processed in real-time.
    • High-Variety, since it is usual to interface and leverage data from heterogeneous sensors and internet-connected devices.
    • High-Veracity, as sensor data are typically noisy and prone to errors and the unreliability of the devices.

    Nevertheless, IoT data have also several differences from conventional BigData analytics, in particular:

    • IoT data collection consumes bandwidth, network, energy and other resources. Furthermore, data collection depends on multiple layers of the network.
    • IoT data analytics should consider optimized data analytics considering the available resource and cross-layer optimisations (i.e. the so called deep IoT analytics).
    • Contrary to conventional BigData systems, IoT analytics solutions should work across multiple systems and platforms.
    • IoT analytics applications integrate in several case physical, cyber and social dataset.
    • IoT analytics and IoT processing are in several cases part of real-time control systems, through providing actionable information.

    Note that IoT analytics systems are commonly deep IoT analytics involving multiple platforms (e.g.. IoT/cloud platforms, publish/subscribed platfoms), networks, IoT data sources etc. i.e. the whole ecosystem of IoT platforms and technologies. Such systems combine data from multiple sources, (near-) real time analytics, visualisation and semantic representations towards transforming raw IoT data to insights and actionable knowledge. The development and deployment of deep IoT analytics systems is challenging, given that they integrate and/or transcend multiple networks, clouds, IoT platforms and more, thus requiring optimization across multiple levels.

    Beyond the systemic aspects of IoT-based data-intensive applications, the development of IoT analytics applications requires the blending and integration of machine learning schemes and data science with IoT platforms. This is discussed in one of the next chapters of the tutorial.

    Resources for Further Reading

    View all IoT Tutorial Chapters