Internet of Things

Hello everyone, On behalf of HEAnet I would like to hear your ideas on how Internet of Things/IOT could benefit your educational environment and what HEAnet could do to help.

Let me start by setting the scene.

What is Internet of Things/IOT

Internet of Things (IOT) is a concept which pops up everywhere nowadays. In a recent IOT newspaper article some examples were listed, including, rubbish bins which signal their content to the bin collector vans so they can be directed using optimum collection routes. Another example is alarm clocks which are linked to your e-diary and will adjust your wake-up time accordingly, governed by user preferences of course. Remote central heating and coffee machines were also mentioned. All of the above are based on the assumption that they operate on the Internet and in the Cloud.

IDA Ireland describes IOT as starting with sensors to measure the physical world. Devices collect and transmit data to a data host (possibly in the Cloud), where it (big) data can be analysed for the benefit of users and others.

I hope these examples don’t recall George Orwell’s 1984?.. All technology can be misused of course, but it can also be used to improve all our lives.

Can IOT be of value to the education sector?

Having asked around, I identified a couple of few areas where the use of IOT could be worthwhile:

  • educating people about what IOT means – the addition of IOT related modules or courses in the curriculum, as per WITs “world-first” BSc: http://tinyurl.com/ot5rknu
  • IOT can also be used in pedagogy. A student provided a scenario on researching an essay. When reading the literature, an eye-tracking device monitors the reader. A heart beat monitor checks for interesting text.. While a highlighting e-pen can add additional bookmarks and context. Together with the personal bookshelf, analysis and advice can be performed. And all of this can help the student with further cognitive support for his/her research.
  • Another area where IOT can be used is supporting a smart campus/institute. I have heard from campus facilities’ departments that understanding how students move around on campus, have library access, use the laboratories and interact with learning management systems could provide worthwhile information about student needs and wishes. This (big data) has to be analysed to identify its real value. Running a pilot can reveal a lot more than simply discussing it. An example of a smart campus can be seen in this video of UCD.
  • Besides the educational aspects, there is the research aspect. Projects on health, industry, public, etc. which would look at how IOT can optimise products, services and life experiences. For instance, researchers at DCUs Insight Centre are working on life-logging which could possibly help people with dementia. Other SFI supported projects on IOT are: CONNECT and SmartBay.

Can IOT be of value to you (and your work)?

The above IOT examples are perhaps too generic. So let’s look at it from your perspective. What I am really looking for is your examples/ideas/experiences:

  • What does IOT mean for you?
  • Is IOT important/not important and why?
  • Will IOT provide challenges/opportunities to society, personal, organisational and/or technology levels? If yes, which challenges/opportunities?
  • What IOT inhibitors do you see; are they political, policy, security, data ownership, privacy, a vague definition of IOT etc? How can they be overcome?
  • Are there any IOT forums already in existence in Ireland or elsewhere?
  • Can IOT devices be manageable; looking at software upgrades, (mis)use aka security, privacy, etc?
  • How can HEAnet, as a service provider, support the IOT activities in the educational environment?: connectivity, protocols, storage, compute power, data models, (cognitive) analytics, providing forums, providing leadership, etc?
  • Do you see a role change for the ICT sector from data provide to phone provider to energy provider (Power over Ethernet: PoE)?
  • Any other IOT related thoughts would be very welcome.

Your feedback will help test the waters on IOT in the educational environment and also help in determining what a service provider, such as HEAnet, could/should contribute. Thanks in advance for your comments.

All the best, Victor

 

Disclaimer

All content provided on this HEAnet blog is for informational purposes only. The owner of this blog makes no representations as to the accuracy or completeness of any information on this site or found by following any link on this site. HEAnet’s blog owners will not be liable for any errors or omissions in this information nor for the availability of this information and will not be liable for any losses, injuries, or damages arising from the display or use of this information.

36 Comments

  1. This might be an interesting webinar on the education side of Internet of Things. It looks to talents needed around Internet of Things and that is thus important for the educational system:
    https://vts.inxpo.com/scripts/Server.nxp?LASCmd=AI:4;F:QS!10100&ShowKey=32522&AffiliateData=email1&mkt_tok=eyJpIjoiTTJGaFl6VmtNelEzTnpWbCIsInQiOiJIWUFOOXhoTklSWVBcL3BEeTRHRWUyd3I5RDcydE1VVFpabGtNQ085WjJWK0d2Q0VEMnJUMjQrYkFTYWJhQWxhc25VSFZWNzE5NEliMmcxcEVvTlhyNjVYREl5b2hKRGlMejM4b1hrTWtYQWc9In0%3D&Referrer=http%3A%2F%2Fmkto-trk.cisco.com%2Fp0z00W0cUV0z1fNHmZ0ULX2

    I myself am not able to join up. So if someone watches, I am interested in your findings. Thanks.

    All the best,

    Victor

  2. Here is an article around IOT and educational institutes: http://er.educause.edu/articles/2016/7/raising-expectations-for-iot-systems-vendors
    It also provides a checklist when purchasing/tendering for IOT related systems. Very worthwhile to read, IMHO.
    Furthermore it gives some four factors that contribute to growing system complexity:
    . The complexity of the technology itself
    . The growing number of systems involved
    . The growing number of interdependencies between those systems
    . The energy and resources required to manage multiple interdependent client-to-vendor and vendor-to-vendor relationships

    Let me know your ideas and how they effect your organisation?

    Thanks.

    All the best,

    VIctor

  3. Hello all of you,

    There is the yearly Internet Of Things World Forum (now in Berlin): http://www.iotwf.com/iotwf2016/agenda
    I went last year (see my blog above). This 2016 forum (again) does not look to have many talks around the educational sector. Security though seems to be discussed at several sessions. So might be of interest, but if you go there for education sector impacts, I am not sure it is worthwhile.
    Do people know if there are conferences around IOT and the education sector? Let us know. Thanks.

    All the best,

    Victor

  4. Another link around security and IOT (seems to be one of the key points around IOT): https://www.researchgate.net/publication/301200106_Security_Vulnerabilities_and_Challenges_in_IoT_End_to_End_Systems_and_Current_Security_Implementations

    Interestingly they have a slightly different layering of IOT systems than IDA Ireland.
    IDA Ireland has: Collect (Physical world & Data harvesting); Connect (Data transmission & Data hosting) ; and Transform (Data analysis & Value discovery).

    Proposed IOT systems layers: Perception (sensing infra & gateways); Communication (Networks & Messaging); and Application (Data processing and management units & Applications).
    So a slight difference between the two models. I like the IDA Ireland one as it looks to go a little higher in its remit: Value discovery (although one could think the Application does that).

    For each of these layers the security aspects are being discussed (chapter 3 and 4). IMHO it does not really provide solutions, it produces more an overview of threads. The perception layer might pose the newest threads: minimal maintenance; power/compute restrictions; remote locations; the sheer amount; etc.
    The other layers seem to be better understood.

  5. There were a few presentations at TNC16 relating to IOT:
    Winners and Losers in the Internet of Things, and What We Can Do About It
    by Maria Farrel:
    https://tnc16.geant.org/web/media/archive/9A (change timeslider at 40:46)

    A good presentation about the cultural aspects around IOT.

    and
    A Nerd and Internet of Everything
    by Victor Reijs
    https://tnc16.geant.org/web/media/archive/2A (change timeslider to 1:09:20)
    This lighting talk (5min) stresses the importance of human conversation and how ‘easy’ it is to rely on algorithms:
    not everything that can be counted counts, and
    not everything that counts can be counted. (by William Bruce Cameron)

    Enjoy

    All the best,

    Victor

  6. Things are progressing in Internet2 and IOT. There is a working group now on ‘IoT Systems Risk Management Task Force’ (chaired by Chuck Benson, who contributed earlier to this blog). That group will thus look at the risk that a campus could experience when IOT takes off, such as: security, privacy, software updates, proliferation of devices, Building Management Systems and IOT, procurement guidelines, etc. If people are interested in communicating about this, please let me know.

    EDUCAUSE Review just released their summer edition which has an IoT Systems theme:
    http://er.educause.edu/articles/2016/6/the-internet-of-things-iot-systems-and-higher-education
    and
    http://er.educause.edu/articles/2016/6/the-internet-of-things-riding-the-wave-in-higher-education

    All the best,

    Victor

  7. Internet2 (the USA NREN) make the following questionnaire around Internet of things:
    https://www.surveymonkey.com/r/NQ6CMGP
    It’s not too late to provide your input about the state of Internet of Things (IoT), smart campus, and related security issues on your campus.

    Thanks for helping Internet2 with gathering this info.

  8. So we had the last day of the IOTWF 2015. The main message of this wrap up day were that change is inevitable in IT (due to the ideas behind IOT). They predict that by 2020 the connectivity revenues of an ISP will be halved (to 40%). This is a major change, mainly due to value added services. The technology behind IOT is not seen as the most important aspect, it are the business opportunities due to enhancing the outcomes for the customers. So the traditional triangle where the network is the base and the outcomes are at the top is totally reversed. The outcomes are the platform at the top and the network is the tip at the bottom.
    I was part of a panel around IOT Collaborative Innovation Community. The panelists came from industry (IBM Watson IOT), standards (IEEE Internet Initiative), research (University of Wisconsin-Madison) and providers (Internet2 and HEAnet). The importance of collaboration in the ecosystems of IOT is essential and this is reflexed by the session. Research is needed in this fast evolving environment, standards have to emerge to provide stability to the industry which needs to implement technologies and services so the providers can serve the educational environment. Looking at the in depth questions received from the audience, there is an eagerness and potential.

    Looking at the four areas I mentioned at the start of this blog, I see the following opportunities:
    . Disseminating courses/degrees around IOT
    There will be a significant demand on: IT education, OT education and ethics needed across disciplines. IOT Talent Consortium can help in this environment.
    . Determining how IOT can be used in the pedagogical processes.
    A person read/looks/browses/underlines text in physical or virtual sources. Sensors pick this up and provide cognitive support for new links/ideas/etc. Or a lecturer can evaluate sources used by students and adjust curriculum. It is important to incorporate the idea of Sense -> Provide, instead of former idea of Build -> Sell.
    . Initiating projects that will support a smart campus/institute.
    Learn from building and smart city projects, and include that in new/old buildings of campuses. Other ideas for educational enterprise whihc could map initiatives like Booking.com and Uber on educational sector, so centralised courses booking (without owning any educational resources).
    . Supporting researchers who work in the IOT environment
    Determine what research areas are of interest to support above: Managing complex multi-components/domain ecosystems; integrating humans and machines; reduce security attack surface; and melding human understanding and interaction into machine/algorithmic systems.

    The IOTWF 2015 was quite oriented towards Industrial IOT. Perhaps there is an opportunity to diversify in the future and get the other sectors more explicit, like health and education. But nether the less: this IOTWF helped to see similarities with industry, such as: see IOT as a disruptive technology; deal with legacy equipment and/or slow refresh rates; expect the need for inter department organisational changes, take security seriously in all parts, work on cyberhuman interrelations, allow risk taking in this emerging environment, and prioritise the outcome for the customers above all (more important then technology).
    But there are differences between industry and education, for instances: the sheer amount of BYOD and BYOX, education is more a consumer/public environment, and privacy is very important.
    All of this will need to be covered by the ecosystems that IOT will generate in the coming future. Lets work on that and make IOT the new IT. Ideas are welcome how to progress this in the Irish educational and research sector.

    All the best,

    Victor

    P.S. There were a few session that were inspiring, when they become online I will provide the links.
    P.S.S. I plan to talk with several Irish initiatives on IOT soon (some of them already reacted on this blog or through personal e-mail).

  9. Ok, a blog on the second day of the Internet of Things World Forum (IOTWF). It looks this forum has a lot of attention towards Industrial Internet of Things (IIoT); companies traditionally in the engineering an devices are well presented.
    They see the convergence of IT (Information technology) and OT (Operations Technology) as IOT, thus decomposing the abbreviation in a different way. IOT is seen to be important for the ecosystem (and not for its individual silos), so this maps with my own views. I am also glad to see that IOT does not have to be only around physical devices, but can also be around virtual devices/sources, like e-mails, files, etc.

    There is an area where industry and education might have a similarity: educational institutes and industry both have to deal with slow refresh times and legacy systems. Perhaps the reasons are different, but this similarity might help finding analogies how to introduce IOT.
    An example is, that industry should move from ‘build -> sell’ to ‘sense the customer -> respond’. This principle can also be used in making an IOT curriculum.
    This large industrial sector will need education in network, cloud services and analytics. So there is an opportunity for extending ones curriculum. The educational system is not really covered at IOTWF; accept perhaps the yesterday mentioned IoT Talent Consortium. Luckily education will be handled tomorrow in a panel discussion in which I take part.

    Interesting is that beside privacy the word ’empathy’ around IOT has been mentioned (see earlier comment in this blog). Also the smart city project of Dubai wants to ‘create a happy environment’ for its citizens. If any analytics outcome is in some way communicated to humans, such considerations are important to cater for.
    Possible research subjects mentioned during the forum are around: complex multi-component systems; methodologies how to integrate humans, machine learning and data; multi-domain environments; smaller security attach surfaces; and machine-human systems that meld human understanding and interaction.

    Tomorrow is the last day of the IOTWF 2015 and you will get another blog comment around that. Feedback is welcome.

    All the best,

    Victor

  10. So I have been at the first day of the Internet of Tings World Forum (http://www.iotwf.com/iotwf2015/week-at-a-glance ), at which are some 2300 participants. Interesting concepts pops up.

    Internet of Things (IOT) is related to many layers, departments and budgets of an organisation. So setting up an organisational strategy might be difficult, as no relevant real use cases exist or proven. So choosing low hanging fruits is wise to pursue. During the presentations it is clear that IOT is seen as an innovation, failure is thus possible and small proof of concept trials (PoC) should be pursued. Someone else stated “If one does not innovates, death is nearby.” By using the golden circle (https://gumroad.com/l/GoldenCircle# ) a good ‘why’ analysis will be part of any PoC.

    It might be the case that an organisation still does not want to take the risk of doing such a PoC. It also seems that finding an integrator for the IOT layers will not yet be easy to find. In that case another organisation needs to evaluate if aggregating at an overarching level can reduce/manage the risk. If feasible HEAnet could investigate: if IOT is deemed important for the sector and if IOT is in HEAnet’s remit.

    So any feedback on the above would be very welcome. Please mention your use cases and/or your willingness to start in our sector a PoC (or extending existing work).

    About IOT as a technology. It is seen as a disruptive technology, as it can (or will) change the way organisations and societies will function in the future. Aspects of security, privacy and standardisation are all essential. IOT is of course not all new: sensors were made in the past, but were not provide ‘open’ data. Analytics existed also, but doing it on a large number of (public) devices is something new (but also not that new in some applications). So IOT is really looking at these different layers: from devices up to making sense/knowledge out the data.
    The network is seen as underpinning and an essential enabler of IOT, but the cloud as data host/compute is of course another enabler. Interesting to hear a new term: ‘fog’, which means that not all data needs to go to the cloud, but it can be processed locally (in the fog). In that case the equipment at the edge (for instance CPE) needs to have enough processing capacity, which can also be used for network Function Virtualisation. Results from the fog can then be send to the cloud. IOT needs also be able to include legacy equipment/services, to allow for a smooth transition.

    Beside the technology, IOT can be seen as a subject to be educate and to learn. See for instance the IOT Talent Consortium (http:///www.iottalent.org/ ). This might be a possible forum for educators to understand what education or courses are needed around IOT by the industry and leaners/students in the coming years.

    More was said, but I hope this will stimulate some feedback from you.

    All the best,

    Victor

  11. On my way to the Internet of Things World Forum, where I will be part of a panel discussion (coming Tuesday) on ‘Collaborative Innovation Communities of Universities, Research & Education Networks, and Industry Leaders Enable IoT’.
    See for the program: http://www.iotwf.com/

    If you have points I need to watch for or points that I should discuss or bring to the attention of the panel/audience, please don’t hesitate to tell me. I will provide from Sunday a daily overview of my findings.

    All the best,

    Victor

  12. Thanks Barbara, Chuck and others,

    As you say: the (device/bandwidth/data/compute) growth, security and organisational aspects are prominent around IOT.

    Around the growth aspects. In the last few years we see a YoY (Year On Year) growth of around 1.35 for the IP traffic. In the past (90s) we were able to cater for a factor of 2 YoY in IP traffic. Analysing the growth of data (in Bytes) one sees similar YoY numbers. And this is in line with Moore’s law, which also shows such behavior in the YoY (https://en.wikipedia.org/wiki/Moore's_law ).
    To be able to cater for this known growth, we need to be able to invest so we can renew equipment and provide more capacity and newer services/protocols. And investments might sometimes be a problem, certainly in times of recession! So it is of strategic importance to make sure that new investments can be done.
    This needs planning and continuous education of ourselves and the funding providers. Working together on this planning and awareness aspect might help our message. Any further thoughts?

    Another aspect I read in the provided links is that IOT devices will need proper access control: some institutes assign at present ‘user like credentials’ to devices, but more sophisticated certification methods seem to be needed. Investigating this could be worthwhile for the education and research community (such as how can eduGAIN cater for this?). What do you think?

    As Chuck brought forward; how do we manage the 100s and 1000s of new devices (incl. BYOD: Bring Your Own Device)? And managing means here in a broad sense: mange feature set of IOT devices, support contract including risk assignments, connection plans for IOT devices, etc.
    I think working on a common approach on this aspect can be worthwhile indeed. By sharing policies, ideas, contracts, etc?

    I liked the statement of ‘The best way to [support IOT] is to plan ahead rather than try to handle it after the fact.” So it is important to work together on the aspects which are important for our community. It is just a matter of finding the common aspects;-)

    Thanks again for your continued comments/feedback.

    All the best,

    Victor

    • Identity is certainly something that should be factored into the IoT space, the research focus right now is on ‘things’ like smart devices and sensors. However, we will need to be able to express the ownership of ‘things’, and as you say authorisation and access control to the ‘things’. HTTP based Identity ecosystems like academic federations (Edugate, eduGAIN) and open equivalents like OpenID Connect do not currently cater for ‘things’ yet the HTTP layer will be the most common destination for the data produced by these things. Much of the IoT literature I’ve read so far emphasises the constraints these devices can face where even an IP connection is difficult, never mind putting data on the connection. I was tempted to say we cam expect some form of HTTP gateway where IoT data is collected in one or more formats that can tolerate lossy data and is sent onto HTTP endpoints, but I am reminded of the GSM days when WAP gateways were the next big thing because of device and wireless network constraints, perhaps we will have greater capability in low-power sensors in the not so distant future than we envisage today that would cause a new wave of IoT (IoT 2.0) devices that can carry HTTP, and thus identity and authz data.

      Another area where I see IoT and security/identity converge is the potential for sensors to serve as an inexpensive second factor in two factor logon. While smartphones can serve this purpose today, we are overly reliant on the false sense of security they offer, I expect to see growth of malware like Zeus that intercepts 2F auth requests on a device, whether sent via Voice, SMS or via 2F apps. IoT devices could offer a secure alternative where the device serves a single purpose; authorizing a transaction or login and does not allow in the installation of other software. I thought falling cost of 2F had started when YubiKey arrived but the growth of 2F apps on Smartphones seems to have put a stop to the downward trend for dedicated 2F devices, but I hope this is a temporary pause.

  13. Last week I encountered an aspect of mobile/BYOD/IOT devices.
    A friend of mine was traveling for a surprise party to the family in Australia. I found it very interesting to hear that before traveling, the geo-location function was disabled of the device. So the surprise stayed a surprise until the end!

    This type of awareness will be essential when being immersed in more and more IOT/BYOD devices. This is of course also related to privacy awareness. So do we have support/awareness programs, so we can educate/help all the people who will get in touch with IOT/BYOD?

    • Awareness is not really something related to technology, but this is very important as IOT will be part of human environment. So we need to think about these awareness aspects to make sure we gain from it all. Some people don’t like the word ‘challenges’, but lets see what ‘opportunities’ IOT gives;-)

      Social media and multi player on-line gaming are discussed in several books, for instance by Sherry Turkle and Susan Greenfield. IOT is not mentioned much (too new?). There are though a few IOT related sections in Turkle’s ‘Reclaiming conversation’ for instance on page 52 and 53. She says “because technology can help us solve a problem, doesn’t mean it was a problem in the first place”. She sees some aspects of IOT as trying to make life friction-free; as machines know what we want and can sometimes predict what we want. For instance a machine (and its involved analytics) could/will reroute our travel to avoid meeting unwanted/unknown people or let us meet only known designated friends. According to her, this will stimulate a life without conflict (the Goldilocks effect). So she asks the question, if we want a life without conflict managed by machines? Do we want such a boring life?
      Having no conflict will be found by many as nice and that can become a kind of addiction, which might reduce ourselves and reduce our empathy.
      Being aware of this process is important/essential, so we can handle IOT in a human manageable way.

  14. I believe that the largest impact on education can be the physical world equivalent of the transformation that search engines have brought about. When I was starting out in education and research, books and journals on the library shelf were the only immediate source of information – anything not there had to be obtained through inter-library loan and could take weeks. Now educators and learners at all levels have all of the world’s information instantly at their fingertips. This has transformed how learning is prepared, delivered and acquired.

    If the IoT takes off as anticipated, we may eventually have the physical world equivalently at our fingertips – a wealth of data, with much in close to real time, about what is happening in the physical world. Further, a growing wealth of accumulated data that, appropriately analysed, will deepen our understanding of what is happening in the physical world and therefore deepen what we can impart to learners. You are teaching veterinary – you could show a live stream of the health of a dairy herd and historical data on national dairy herd health trends. You are teaching process engineering – you could show real-time data from a manufacturing process and how the control system is keeping the process stable, historical data on process variations. You are teaching medicine – you can show data on the historic cardiac health of a population of at-risk patients and the impact of cardiac monitoring on health management and mortality reduction, show a (volunteer) patient’s pulmonary artery pressure live in your lecture regardless of where the patient is. Here in Nimbus at CIT, we have an extensive building energy and occupancy IoT testbed monitoring a number of buildings on campus. The data from this is available in real time through a web interface and can be used by teachers and learners in architecture, civil engineering, energy engineering, data analytics, ICT, process and control engineering… We are developing similar testbeds for other infrastructure, with the objective of maximising both the research and educational value.

    Pick your own discipline area and use your imagination…

    I realise that this is an idealistic vision. There are many difficulties, including the market value of IoT data leading to it being locked away in pay-for-access silos and the conflict between security/privacy and the value of making IoT data publicly available for education and research. There is also a major hurdle in simply analysing and meaningfully visualising the data. However, if we grasp this vision, we can begin to plan for educational use of the insights and knowledge generated from IoT data and for the data analytics required to do this. We can begin to plan for national regulatory regimes that protect privacy while allowing data to be used for educational and societal benefit. We can plan how we allow national educational access to data from the growing number of IoT testbeds in Ireland.

    From an IT infrastructure viewpoint, the problem remains broadly the same as with current search engines: the infrastructure that can deliver us millions of search results in a fraction of a second can, if the trajectory of increasing bandwidth and access speeds continues, probably deliver us more than enough IoT data to be going on with. It’s being able to extract meaning and value from the search results/IoT data that’s the real challenge. It’s hard enough to find the value in web searches that drown you in information – if the IoT is going to generate 10 or a 100 times more data than the current internet, then we should be encouraging the nation’s children to pursue careers in data analytics.

    • Hello John and others

      Interesting this move from books to raw data. Book are based on raw data, but they use of course condensation and analysis of this raw data. By using the raw data (together with some existing knowledge from any source), one might/will indeed reveal new knowledge, I am sure.

      The raw data can come from sensors, but other sources can also act as ‘sensors’. For instance e-mails are now sensors to Google’s analytics for determining our advertisement threads. Another more ICT oriented example: (SNMP) monitoring probes in equipment act as sensors and here it shows that it is not always easy to find the right anlytics to provide high level advice to network/service managers.
      I like Mary Keeling’s presentation at the HEAnet conference (https://www.heanet.ie/conferences/2015/talks/id/166 ), who uses raw data in the form of facebook or Twitter messages, to inform organisations.

      By not only looking at physical sensors, we might be able to reuse some analytics ideas from the past. And for the reverse: IOT devices can stimulate us to generate better analytics for non physical raw data.
      Of course extracting real value of this data is the pivotal aim. And as you say John, we should work on the privacy aspects around all this data (not only from physical devices).

      I like your statement: “If we grasp this vision, we can begin to plan for educational use of the insights and knowledge generated from IoT data and for the data analytics required to do this.” Starting to work on IOT is a jump into the dark, but lets grasp that excitement.

      I hope that HEAnet can be part of this access to the growing number of IOT testbeds in Ireland! Let us know how we can help here; be it with connectivity and other services, stimulating standardisation and/or providing fora to discuss.

      All the best,

      Victor

  15. Great to see the new blog. You might like this … I have been working, after hours, on an software architecture for a marine internet of things or as I have coined a Sea of Things! The architecture is designed around the IoT centric mBed hardware ‘device’ architecture e.g. a NXP LPC1768 or a u-blox C027 – low cost, low power, connected devices and reasonably capable. The marine environment brings restrictions in terms of power and bandwidth – which are scarce resources. Current design supports standalone operation or connectivity back over 2G, 3G, VHF, ethernet or WiFi. Sensors integrated so far include accelerometer, gyro, A/D, magnetometer, tilt compensating compass (pitch, roll, yaw), humidity, temperature, real time clock, FET and relay switches, current and power consumption, bump / impact detection, walkabout polygons and circles, a solenoid, altitude/pressure sensor and USB storage. Lots of data available ! And not forgetting a PP9 and a complimentary iPhone app visualising the data in real time. All in a package the size of a deck of cards !

    Next phase is building a low cost football sized buoy, loading with the mbed and sensors, and deploying from a fixed or rotary wing aircraft in Galway Bay, and yes we will use the altitude sensor to detect when to release the deployment descent parachute and have a nice ‘apollo’ style sea landing. Why ? Well it is a low cost marine proof of concept device deployable to sense the good / the bad and report when needed all in real time and deployable at a moments notice. Device provides any amount of sensor data e.g. gps tracking so we can track for example tidal influences, thermo couples for water temperatures. In terms of the education and research sector – (coastal) schools could easily add more low cost sensors packages, perform experiments and analyse data gathered in a fun or academic way.

  16. IoT systems offer many potential benefits to Higher Ed. The nature of IoT systems also introduces a new flavor of risk that we’ll need to figure out how to manage.

    One of the unique aspects of IoT systems that I’m concerned about is the ‘seam’ that exists between our organizations and IoT systems vendors. For example, the fact that a single IoT systems vendor transaction/contract can bring on 100’s or 1000’s of new networked/computing/sensing devices on our networks & the risk that that introduces is often not well-accounted for. Also, because the devices can often be somewhat ‘invisible’ in practice (ie energy meters in utility closets, environmental sensing devices hidden in the infrastructure of buildings, video cameras mounted high/out of direct sight, etc), it’s easy to not be aware of the magnitude of the install in our physical spaces & networks.

    I think we need to be more formal & detailed in setting expectations & measures of performance for our IoT systems vendors. I think we need to redefine what constitutes a delivered product or service. Examples of requirements include:

    — were unneeded services turned off at all of the IoT endpoints/devices? How does the IoT systems vendor demonstrate & document this?
    — were default logins/passwords changed for the endpoints? what was the new login/password schema used?
    — does contract language include risk assignments? (who’s responsible for what when things go awry)?
    — do we plan for who does repair/preventative maintenance/troubleshooting for the high number of endpoint devices?
    — is there a commissioning/decommissioning plan for endpoint devices (testing, documentation, etc)?
    — has an endpoint/device patch plan been established? has the labor associated with the patching effort been identified?
    — etc

    So, with all of the benefits that IoT systems bring, I think the risks introduced by IoT systems are a different flavor from what we’ve worked with before & as Higher Ed institutions, we’ll need to get specific about how we manage that seam between the organization and the IoT system vendor.

    (I dive a little deeper into some of these ideas in the posts below if you’re interested.)

    Best
    Chuck

    http://longtailrisk.com/2015/04/28/systems-in-the-seam-shortcomings-in-iot-system-implementation/
    http://longtailrisk.com/2015/05/15/developing-an-iot-vendor-strategy/
    http://longtailrisk.com/2015/07/14/whos-building-your-iot-data-pipeline/

    Chuck Benson
    Chair UW IT Service Management Board &
    Assistant Director for IT, Facilities Services
    University of Washington
    cabenson@uw.edu
    206.221.4001

  17. NRENS contributed massively with operational experiences in the first steps of the IP era. It is essential to take over the responsibility to gain and invest knowledge in this new field of IOT. The IoT field leverages IPv6 technology but with a new stateless addressing adaptation (6lowpan). How this scheme affects security ? The IoT encompasses a new routing scheme. I expect that NRENs will contribute and report back in operations practices.

  18. Great to see a new blog on the IoT topic, something which will be increasingly relevant in the marine domain. As more connected sensor data feeds become available there is great scope to develop marine monitoring and operational decision-making.
    Some IoT-related issues in the marine space include sensor maintenance, the cost of sensors and connectiveness. More research is needed in these areas, while a wide area comms network for IoT devices would help with connectivity.
    On a related note the Marine institute is currently developing a framework for integrating and making available marine data called Ireland’s Digital Ocean (www.digitalocean.ie) and we are very keen to add new sensor feeds as they become available. We will shortly be including data from the new underwater observatory at the marine renewable energy test site in Galway Bay (see http://www.oceanenergyireland.com for more info) which is connected via a HEAnet circuit and will provide streaming underwater sensor data from Galway Bay online in seconds.

Leave a Reply

Your email address will not be published. Required fields are marked *