Version classiqueVersion mobile
OpenEdition Books

L’Internet des objets

 | 
Pierre-Jean Benghozi
, 
Sylvain Bureau
, 
Françoise Massit-Folléa

The Internet of Things

Chapter 3

Technical evolution of the Internet of Things

Texte intégral

1Although much progress has been made, a certain number of issues must be resolved before the IoT can be used in the same way as the Internet. In this chapter, we will first list the main technological obstacles that hinder the development of the IoT, and then we will address the matter of competition between technologies (second part) and the current principal standardization processes (third part).

The main technological requirements

2Currently, three major points must be addressed : improvement of the performance of solutions, better interoperability between systems while guaranteeing optimal security, and the design of an efficient data management system.

Guaranteeing the performance of solutions used

  • 1 van Liehout, M.,L. Grossi, G. Spinelli, L. Kool, L. Pennings, R. Stap, T. Veugen, B. van der Waaij, (...)

3The definition of three types of performance helps in understanding of the current issues : first, sheer technical performance within laboratories ; second, technical performance in a real environment ; and third, the technical and economic equation contained in the solution. Over the past few years progress under laboratory conditions has been impressive. It is now possible to find RFID solutions that can function under water (thanks in particular to NFC – Near Field Communication – technology and ultra-high frequency) and in highly metallic environments1. Similarly, reading ranges and rates have increased considerably over the past decade. Lastly, the consumption of energy by RFID solutions has dropped markedly ; although much progress is still to be made on this point in order for systems to function long-term while retaining a low ecological impact (this issue is particularly sensitive for active and semi-passive chips).

  • 2 van der Togt, R., Jan van Lieshout, E., Hensbroek, R., and others, e. 2008. Electromagnetic Interfe (...)

4Conversely, the reliability of solutions in certain usage contexts still needs to be improved. For instance, researchers from a University of Amsterdam laboratory have recently shown that the radio frequencies emitted by RFID solutions could cause interference with some vital medical equipment such as pacemakers, ventilators and the devices used for dialyses2.

5Finally, the most reliable solutions remain too expensive for large-scale dissemination. It is of course possible to find passive RFID chips on the market for less than 10 euro cents, but they will neither be equipped with a very secure system, nor have an effective reading rate. To promote rapid dissemination, it is necessary to develop solutions that better respect the technical and economic equations of companies, for the current cost of change is often too high. Realistically, less costly solutions must be produced, not only regarding chips but also extending to all the system components (readers, antennae, middleware, etc.) which enable the storage of a moderate volume of data on an ever-shrinking surface, while at the same time maintaining an efficient reading rate. Where active chips are concerned, energy consumption will have to be cut in order to increase battery life, ensuring that batteries last as long as the object identified by the chip. As a final point, the issues of chip recycling and the ecological impact of RFID solutions will have to be resolved.

Ensuring the durability of solutions

6When looking beyond the mere question of system performance, three types of problem should be examined : flexibility (the selected solution can evolve with the context), interoperability (the selected solution can communicate with other solutions), and security (the solution is secure for both data and users).

Guaranteeing flexibility

7There is no point in purchasing a costly solution that must be changed after a few months. Where supply chains are concerned, it is essential that the entire system evolves and adapts to modifications. For example, a solution which functions when the weather is fair but not when it is wet is inconceivable (indeed humidity has an impact on the performance of RFID solutions). Similarly, if delivery procedures change, the system must be flexible enough to adapt to new rules without disrupting the whole system, whether at hardware or software level. If the company decides to change or alter its ERP, would they also need to change the middleware or all or part of the RFID solution ? Although this case is somewhat theoretical, it is a good illustration of issues that could arise as RFID solutions evolve.

Guaranteeing interoperability

8The success of the IoT will also depend on the interoperability of systems. As a first stage, it is essential that all the components of a solution be interoperable between each other. For instance, the RFID reader must be compatible with the chips used. To satisfy this requirement it is necessary to create an open system in which the different stakeholders may operate. Here, solutions that can adapt to heterogeneous contexts must be developed in order to guarantee reliable functioning throughout the period of use. Interoperability is also relevant in the convergence between RFID and mobile solutions. Up to this point no definitive solution to this problem has been designed. Decisions made by telecommunications companies will naturally be far-reaching, but for the moment their strategy in this field appears undecided.

9As we will see later, the question of interoperability is not limited to the merely technical. Indeed it affects several domains : the legal (in particular the proliferation of patents), the economic (the complex question of royalties) and also governance (this will be examined, with a focus on the role of the European Commission and of the major standardization institutions).

Guaranteeing security

10Any system must also ensure the security of data. If an RFID solution is deployed between a client and a supplier, the two parties must share a certain quantity of information, but at the same time the data that is useful for the supplier does not necessarily have to be disclosed to the client. This point is clearly very sensitive in the healthcare sector, for the information held on each patient must not be accessible to all involved (insurance organisations, pharmaceutical industry, employers, etc.). Of course, this issue is not new, but the technical set-up of the IoT could have a strong impact on storage options (quantity, and location of storage, local or dispersed) and on data access (it is possible to read data contained in a chip remotely, perhaps at the expense of the user). Technical systems that ensure data security must be put into place ; and beyond this, there is the need for a framework that gives guidance and also prevents prejudicial and illicit practices. Lastly, and specifically, the impact of radio frequencies on the human body must be examined (this question will be addressed in the final two chapters of the report).

11The technical levers and constraints are numerous, and we have provided only a brief description of the critical factors. Other points, such as the relation of the IoT to nanotechnologies and robotics, might have also been mentioned.

Designing an efficient data management system

An unprecedented mass of data

  • 3 Gonzalez, H. and J. Han, 2008. RFID Data Warehousing and Analysis, in L. Yan, Y. Zhang, L. T. Yang (...)
  • 4 ibid.

12According to Venture Development Corporation, Wal-Mart would produce almost 7 terabytes of data each day if there were a chip on each item sold3. What is to be done with this “flood of data” ? Do current solutions used in particular in Business Intelligence have the means to satisfactorily manage the quantity and complexity of this data ? The data emanating from RFID chips is highly specific, as it must give information on space (location) and time (including the movement of objects in time). For this data to make sense, it has first to be cleaned in order to avoid repetition (sometimes, however, under the pretext that there is a large mass of data, the problem is rather the loss of critical data !). At times it must also be aggregated and, depending on the required granularity, it is not always necessary to have access to all the data. In other cases it may be preferable to note information on a pallet and not on specific objects. Lastly, it is necessary to be able to structure this data. It will be necessary to invent new datawarehouses in order to provide managers with dashboards to assist in decision-making. At present, although some research projects and a few applications4 offer some solutions, there is still a long way to go before there is real-time integration of data originating from different geographical locations.

A need for a new referencing framework

13Going beyond these issues that pertain to companies, one may conjecture what type of research engine might be adapted to the IoT. This question has not yet been discussed to any great extent in existing reports, although research tools will be as important in the IoT as they are in the current Internet. The increased use of Enterprise Search is already a sign of the potential importance of these new markets. Their development will probably require the resolution of new technical challenges, linked specifically to 2D and 3D engines.

14These questions cannot be answered from a purely technical standpoint. The history of technologies shows that the most effective solutions are not always those that are adopted. To fully grasp the reasons for this, it is essential to first examine the competition that exists between diverse technological solutions.

Competing technological solutions

Different types of RFID

15Above and beyond the wide choice of chip types and radio frequencies, there are several, competing RFID solutions available on the market. The major players – be they chip or sensor manufacturers, software editors, integrators, or consulting firms – all attempt to further their own technologies, standards and services. Besides, leading users who influence the market through the extent of their orders – for example, Wal-Mart, Metro or the Department of Defence – select first and foremost RFID solutions for the efficient reconfiguration of their own value chain and not for the development of a more general standard. In the case of supermarket distribution, these solutions must in particular improve the vertical integration of the supply chain, facilitate online payment, and improve customer profiling in order to implement a customized marketing policy. In addition to their impact on markets, these major clients have also played a significant role in the past years as a result of the financing of research in the field (the case of Auto-ID labs is a striking example) or by participating directly in consortia which work on the definition of common tools and standards (the agriculture and health sectors are particularly representative). In fact, pure RFID players are rare, even though some are beginning to develop considerably, such as TagSys, Verichip and Alien Technology. Indeed, the market remains widely dominated by the traditional participants, and in consequence the RFID solutions developed appear to some extent to target segmented markets. Thus participants in the high-frequency chip market sometimes differ from those present on the low-frequency chip sector. However, the durability of this division is questionable and it is believed that in future market segmentation may be based on use rather than on technical parameters.

Alternative solutions to RFID

16We have hypothesized that RFID chips will be essential to reliable functioning of the IoT. Yet other solutions may be envisaged. Intel, a company that has invested large sums in R&D linked to RFID, has recently sold this branch of its business to a Californian producer of RFID readers (Impinj). This does not mean that the future of RFID is in jeopardy : the transfer does not totally cast doubt on Intel’s involvement in the RFID market, but rather depicts a change in the type of investment (the agreement with Impinj included an exchange of shares). However, the future of the IoT will not necessarily be bound to that of RFID technology. Many laboratories and firms are working on the development of alternative solutions that could, in the future, replace RFID solutions, such as, for instance, acoustic identification systems, the use of microwaves, of optical systems, or the detection of DNA. Some are also considering integrating chips into the design of objects themselves, which would strongly impact on the issues surrounding RFID as we perceive them today.

17At the heart of this competition between technologies the major industrial and public players make strategic choices via standardization organizations.

The progress of IoT standardization

18The question of standards involves technical, economic, legal and political issues. Rather than give a comprehensive account of these questions or list important standards for the IoT, we will centre this analysis on three crucial questions : the dependence of current standards on former norms, the difficult conciliation between the pertinence and equity of standards selected, and the challenge of effective interconnection between standards.

Dependence on existing standards

  • 5 As already mentioned, EPCglobal brought together EAN international and Uniform Code Council, two in (...)

19IoT standards are based on the main principles that are at the core of the Internet (DNS, TCP/IP, etc.) and the barcode system. The key players involved in this standardization are one and the same5. Thus the TCP/IP protocol and XML language remain a reference for the IoT. Similarly, the guiding principles of the Electronic Product Code are very close to those used for the barcode system.

20This legacy of the Internet and the barcode system to the IoT is likely, for technical reasons, to limit certain developments and applications. Moreover, if billions of units of extra data have to travel through the network, it is doubtful whether the Internet will be able to adapt : not only for physical reasons regarding the infrastructure (namely due to router capacity), but also for issues of logic (in the long term, the principles designed for the Internet would not be viable for the IoT).

21Some believe that incremental measures could suffice in facing up to these new challenges as well as to those of IoT security and reliability. The shift from IPv4 to IPv6 (Internet Protocol version 6) is an example of this approach within the framework of the historical Internet paradigm. This new protocol will allow for a significant increase in the number of addresses available (from 4.29*1010 addresses to 3.4*1038) and for an improvement of router performance and security.

  • 6 A wide-scale research project was launched at Stanford : http://cleanslate.stanford.edu/

22However, other research teams6 advocate discarding former standards and seizing the opportunity to develop the IoT from a clean slate, thus permitting a complete redesign of the main functioning principles of this new Internet. In terms of evolution, or revolution, the technical, economic, and, more widely, political impacts will not be the same.

A standard of standards

23Today, the standardization of networks is carried out either by private industrial groups, or by international standardization organizations. Where industrial groups are concerned, EPCglobal is one of the major players involved in regulation, whether public or private ; standard conformity certification is out-sourced to a private company (MET). Then, AIM Global and Global Data Synchronisation are private cooperative organizations that are central to the process of connecting applications, objects and firms. ISO, ITU and ETSI are at the core of the network of public representation in telecommunications standardization. W3C, IETF and IEEE are the founders of Internet standards. Lastly, ICANN is the organization that currently carries out domain name management on the Internet.

24However, the impact of IoT on society will be such that it is essential to gain consensus between private and public players with regard to the choice of standards, and to ensure that their choices are not guided by solely economic considerations, but rather that they take into account questions of privacy, sovereignty, etc.

25Moreover, it is important that IoT standards not be limited to those defined by a few major players, and that they do not become geographically fragmented.

26Standardization represents a key technical challenge not only because of the nature of the technical choices (frequency, power, encoding, etc.) but also as a result of the diversity of levels and types of standardization. Those involved in these issues are highly diverse and the selected models alternate between proprietary and private, voluntary and shared solutions (industrial model and standards), de jure public (telecommunications model) and self-regulation. But the aim must remain to succeed in developing open standards by promoting a transparent dialogue between the different participants. This is a necessary condition for the development of competition and for the harmonization of solutions available on the market.

“Granularity” and interoperability of standards

27The question of the granularity of standards is crucial. It consists of ensuring that each standard is sustainable on the international, national and regional levels, as well as at the level of each specific industry. Granularity is just as important as interoperability for avoiding fragmentation where there has been a lack of anticipation in the choice of interfaces associated with technical systems, such as nanotechnologies, sensors or ONS. It is essential to examine these issues upstream, otherwise there is a risk that the development of innovative applications, themselves a source of value creation, will be constrained or slowed.

28Certain standards are now beginning to impose themselves as a result of actions by organizations such as EPCglobal. Nonetheless one must be careful that other stakeholders are not excluded from decision-making. It is essential to reflect on the equity of the standardization process : for instance what role Europe should play in the choice of these standards should be considered.

Interoperability between standards

29Beyond simply the improvement of standards, it is fundamental to plan for their interoperability, particularly since the IoT is a system of systems which necessarily requires a high level of interoperability. In this context, some significant breakthroughs have been achieved in the case of NFC (Near Field Communication) technologies. NFC solutions enable the connection of two electronic devices a short distance from each other, via a wireless system, in order for them to interact (this type of system is used in Japan as a means of payment). At the beginning of 2008, the members of the ETSI (European Telecommunications Standards Institute) selected a software protocol which makes it possible to control communication between the NFC system and a SIM card. This type of standardization is essential where large-scale production is envisaged. Similarly, on the mobile phone market, many positive developments have recently taken place with the adoption of the LTE (Long Term Evolution) standard by many key players in this sector (AT&T, China Mobile, Vodafone, Verizon, Motorola, Nortel, etc.). This trend will have a significant impact on achieving a better definition of 4G (fourth generation) : until now the definition of this standard has been relatively ambiguous, which could deter some investors.

  • 7 A report by the IDABC (Interoperable Delivery of European eGovernment Services to Public Administra (...)
  • 8 http://www.bridge-project.eu/

30Over and above these breakthroughs on the mobile phone market, the role of public institutions remains essential in promoting standards that guarantee interoperability. The European Commission thus defends the principle of open standard via the IDABC7. The aim is to ensure governance by a non-profit organization, which enables all stakeholders to participate democratically in the creation and the evolution of the standard. Besides, the dissemination of this standard is widespread and free (or at nominal cost). Also, copying and distribution may be executed freely (for instance, there are no licences). As an example, the European Commission supports the BRIDGE project (Building Radio Frequency IDentification solutions for the Global Environment)8 launched in 2006. This aims to implement an open standard based on the application developed by EPCglobal for the use of RFID solutions that manage the traceability of products throughout the supply chain.

Notes

1 van Liehout, M.,L. Grossi, G. Spinelli, L. Kool, L. Pennings, R. Stap, T. Veugen, B. van der Waaij, and C. Borean, 2007. RFID Technologies: Emerging Issues, Challenges and Policy Options, in I. Maghiros, P. Rotter and M. van Liehout, eds: 278. Seville: Institute for Prospective Technological Studies, JRC European Commission: 44.

2 van der Togt, R., Jan van Lieshout, E., Hensbroek, R., and others, e. 2008. Electromagnetic Interference From Radio Frequency Identification Inducing Potentially Hazardous Incidents in Critical Care Medical Equipment. JAMA., 24(299): 2884-2890.

3 Gonzalez, H. and J. Han, 2008. RFID Data Warehousing and Analysis, in L. Yan, Y. Zhang, L. T. Yang and H. Ning eds, The Internet of Things. From RFID to the Next-Generation Pervasive Networked Systems: 53-80. New York: Auerbach Publications: 54.

4 ibid.

5 As already mentioned, EPCglobal brought together EAN international and Uniform Code Council, two institutions at the heart of the bar code standardization process.

6 A wide-scale research project was launched at Stanford : http://cleanslate.stanford.edu/

7 A report by the IDABC (Interoperable Delivery of European eGovernment Services to Public Administrations, Business and Citizens) published in 2004 defines the open standard in the following way: “the standard is adopted and will be maintained by a not-for-profit organization, and its ongoing development occurs on the basis of an open decision-making procedure available to all interested parties (consensus or majority decision etc.). The standard has been published and the standard specification document is available either freely or at a nominal charge. It must be permissible to all to copy, distribute and use it for no fee or at a nominal fee. The intellectual property - i.e. patents possibly present - of (parts of) the standard is made irrevocably available on a royalty-free basis. There are no constraints on the re-use of the standard.” (IDABC (2004): “European Interoperability Framework for Pan-European eGovernment Services”: 9).

8 http://www.bridge-project.eu/

© Éditions de la Maison des sciences de l’homme, 2009

Conditions d’utilisation : http://www.openedition.org/6540