The point in time when the number of connected devices outnumbered the number of people on the planet corresponds to the birth of the Internet of Things (IoT).
The challenges of patenting IoT technology may be classified into four categories (for further details please see “Smart devices in IoT need a smarter patenting strategy”):
- Challenge one: scope of claim.
- Challenge two: joint and divided infringement.
- Challenge three: subject-matter eligibility post-Alice v CLS Bank.
- Challenge four: patent quality.
This report proposes solutions to overcome each of these challenges.
Solutions for claim scope-related trade-offs
Keeping in mind patentability requirements, the cost of obtaining the patent and the value proposition from the invention, the following approach should be taken to address these trade-offs. To work out the best types of claim for each particular innovation, consider the following factors:
- What is the business model of the company?
- For sellers of the actual IoT device, it is preferable to prioritise device claims and method claims that cover how the device operates.
- If the company only sells software, software-style claims can be prioritised.
- If the company is a backend service provider, the server side ‘in the cloud’ processes can be prioritised.
- Where does the point of novelty lie?
- If the invention deals with how the IoT device operates, it might not be possible to have a viable claim for server-side functionality.
- If the invention is software-implemented, consider software-styled claims.
- What is the relative licensing value for each type of claim?
- It might be possible to license a software claim across multiple different industries (eg, for operation on multiple different types of device).
- Who is the potential infringer?
- The general rule of thumb is to choose a claim that can be asserted against a competitor (as opposed to a customer or end user).
It is also important to account for unexpected prior art. As there is a huge amount of innovation in the IoT space right now, it is possible that during patent prosecution the invention might encounter prior art that is currently unknown (eg, unpublished patent applications). Therefore, the claim strategy should take this into account. For example, several back-up positions can be built, either as dependent claims or in the specification.
To begin with, a focused claim set can be relied on, and broader claims can then be pursued in a continuation application.
Overcoming the issues of divided infringement
Typically, the issues of divided infringement can be overcome by carefully drafting claims from the perspective of only one device in the IoT system.
But drafting that type of claim may not be possible, depending on the state of the art and whether the invention is even patent eligible. If this is the case, it is preferable to claim the components that would most likely be sold or operated by the competitors. An exemplary case of divided infringement in the IoT domain is discussed below.
Smart city: how joint infringement could occur – a real-life example
Smart city schemes are a real-life example that shows how the various claim strategies to overcome joint infringement might play out. Broadly, ‘sensing cities’ use smart sensors as municipal infrastructure. In patenting a sensing city, the following could be claimed:
- The smart sensor itself – however, that likely would not capture the big picture idea of how it is being used. Also, by itself the smart sensor might not be novel, and therefore not patentable.
- The cloud perspective (which describes how the data from the sensors is used) – however, it might be much more difficult to detect infringement because this is basically claiming a software process running on a private server.
- The whole system (which covers the sensors and the servers operating together) – however, this could lead to a divided infringement problem where one entity owns or operates the sensors, and another aggregates or processes the data.
Sometimes overcoming this problem may require compromise — it may not be possible to claim the invention from the perspective of a single device. However, in most cases, a smart drafter can find a way to craft independent claims without a divided infringement problem, which is usually a strong approach.
Overcoming subject matter eligibility issues
In wake of the recent decisions concerning patent eligibility, there is a need for strategic application drafting for IoT inventions. This need becomes obvious in view of the following case laws:
- The Alice two-part test:
- The laws of nature, natural phenomena and abstract ideas are not patent eligible.
- Is an abstract idea in the claim or is the claim directed to an abstract idea? If yes, determine whether the additional elements recited in the claim transform the nature of the claim into a patent eligible application.
- Diamond v Diehr:
- Although the claim may encompass a mathematical formula, if it “implements or applies that formula in a structure or process, which when considered as a whole, is performing a function which patent laws are designed to protect”, the claim is patent eligible.
- The claims in Diehr were directed towards an improvement in the existing technological process.
While Alice provided a glimpse of how to determine patent eligibility, an older decision in Diehr clearly indicates the underlying principle behind the determination of patent eligibility. For IoT inventions, the closer to Diehr that the patent application can recite the IoT invention and narrowly tailor the claim features that provide a practical application (preferably a real world application) for the invention, the higher the likelihood of obtaining the patent and withstanding the America Invents Act post-grant eligibility challenges.
An exemplary patent-eligible case that fits best for IoT ecosystems is Thales Visionix v United States. The invention was broadly directed towards a helmet-mounted display (HMD) used by F-35 fighter pilots. The system utilised a pair of identical sensors arranged as – one mounted on a helmet, the other mounted on an airplane – to determine the orientation of the helmet relative to the moving airplane. During flights, HMDs are subject to drift, in which even small measurement errors can accumulate into larger errors when estimating an object’s position. Visionix’s claimed solution uses a computer running a mathematical equation to periodically calculate the relative orientation of the helmet. The following observations regarding Visionix’s solution should be noted:
- There was no improvement to the claimed computer that executes the steps.
- The identical sensors were conventional or generic ones. However, the claimed elements (though generic in nature) operated together to achieve a particular goal.
- The claims “result[ed] in a system that reduces errors in an inertial system that tracks an objection on a moving platform”, “eliminate[d] many complications” of prior art solutions and were thus “unconventional”.
The key takeaway from this is that drafting an application for an invention directed towards a non-conventional, specific arrangement of sensors (even though the sensors themselves may be well known) may still provide grounds for patent eligibility.
Another relevant case, Fitbit v Jawbone, provides further insights in this regard:
- Improving the human interface for consumer IoT inventions (perhaps because of the apparent real-world application) seems to be persuasive in avoiding or overcoming Alice.
- The interaction of multiple nodes in an IoT ecosystem seems to be another common theme. Further, illustrating the underlying problem solution story may prove beneficial.
- Considering the small form factor of the claimed IoT devices (wherever applicable), it is imperative to discuss the inherent challenges and how the claimed solution addresses those challenges arising due to small form factor.
The following points give good insights regarding patent eligibility of IoT technology landscape:
- Industrial IoT inventions, if properly claimed, will continue to be patent eligible.
- A new arrangement or combination of generic IoT elements directed towards solving an old standing problem will remain patent eligible.
- Drafting the claims in a manner that addresses a technology-centric problem can be found persuasive while ascertaining patent eligibility of the claimed subject matter.
A list of software-styled patents which were found to be eligible by the US courts and which are relevant to the IoT ecosystem can be found here.
Overcoming patent quality issues
The sole purpose of obtaining a patent on an IoT invention may be to protect the interests of the inventor(s) pertaining to what the inventor(s) may regard as the invention. In this regard, it may be important to focus on quality patents that are enforceable. Drafting patent applications that are far too narrow may lead to patents that may not be enforceable and should be avoided at all cost. Specifically, ‘narrow applications’ mean the undue limiting of the claims and embodiments beyond the point of novelty. Features that do not contribute to the absolute novelty should be recited at a broader level. Similarly, features other than those which bring out the ‘significantly more’ aspect of the invention should be accorded their broadest possible scope. Of course, they can be identified with particular examples in the specification to provide enablement.
Another important aspect that may improve the quality of the patent would be drafting focused dependent claims. The prime importance of dependent claims emerges from the fact that the dependent claims act as back-up claims for the independent claims. Accordingly, this objective should be prioritised over all other priorities of dependent claims. In as much as possible, features that form part of prior arts, particularly obvious to combine features, should not be preferred over important aspects of the invention that may narrow down one or more features of the independent claim.
Therefore, the best approach is one centred on drafting realistically broad claims (with high infringement potential) and one which addresses these challenges.