Your IP: easy to protect, easy to reuse
By Nathan John, Director of Strategic Marketing, Cypress MicroSystems, Bothell, Wash., EE Times
December 19, 2002 (10:32 a.m. EST)
Design engineers are constantly faced with the problem of creating and protecting intellectual property (IP) in new products. IP is the knowledge that enables companies to gain competitive advantages. It can add new capabilities to a company's products and get them to market faster. To maintain this advantage, many companies are waking up to the value they have accumulated in their IP and are seeking out ways to protect it.
In the dog-eat-dog world of electronics, competitors are eager to uncover the "secret sauce" of a new product as soon as it is shipped. If your board contains discrete components, a simple reverse-engineering process will quickly reveal its circuitry advantages. However, chipmakers now offer IC technologies that make it difficult or impossible for your competition to access your IP.
There are several types of ICs that electronic designers can use to protect their intellectual property, and they have pros and cons de pending upon time, cost and ability to re-use the IP.
Patents are one traditional method of protecting your valuable IP. Patents can provide a powerful market advantage to their holder. In essence, a patent grants monopoly rights to make and sell a product that incorporates this IP. In electronics, the length of time for the patent usually offers this monopoly for longer than the useful life of the technology itself. Many companies reward their employees for patents because of the strong benefits they can bring in protecting IP.
However, the patent process is both time-consuming and expensive. An average patent development and filing process can take weeks, or even months. During this time, engineers spend a significant amount of time focused on documentation diverting energies away from their core design responsibilities.
In addition to lost productivity, patents are an expensive project. When you consider all of the associated fees including those for application preparation, legal guidance and Patent Office procedures filing a patent can cost your company between $10-15K. Costs continue to escalate if you initiate legal proceedings to enforce your patent. Even though patents are a valuable form of protection, they are never an absolute guarantee that someone else will use your ideas.
A more effective approach to IP protection is using a design methodology that keeps it hidden from the outside world. To address the issue proactively, you can either use a custom IC or a programmable device, both of which can lock your IP inside. The level of protection provided by these different devices varies with the individual technology that you chose.
Creating a custom device not only provides the exact functionality that your application requires, but ensures that your IP cannot be copied by others. You can build your entire circuit inside, creating a System-on-Chip (SoC). And, the SoC can include both digital and analog capabilities, with no way for a nyone to view the architecture from outside the chip. Many IC manufacturers provide libraries of IP blocks that allow you to incorporate many standard functions. These same vendors now offer CPU cores that give you the capability to further enhance your design with software.
The custom-IC design approach provides you with the ultimate level of IP protection. However, creating a SoC is not a project to take lightly. For one thing, SoCs require large non-recurring engineering (NRE) charges and significant lead-time. You will also spend considerable time developing a specification from which the silicon design will commence.
Another drawback to custom SoC creation: a limited upgrade path. As soon as you complete your SoC design or in some cases even before you complete it you will inevitably think of new features that should be added to the chip. At the very minimum, this will delay the delivery of your device. Last-minute changes can also increase NRE charges, which impact t he final cost of the part.
There is also some level of risk that comes from transferring your IP to another company. Ultimately, even though you own the IP that is inside the chip, you need to pay the supplier every time you want to modify this IP, or use it for another purpose. The capability that you gain in protecting your IP costs you dearly in the lack of easy re-use.
MCU lock up
Locking your IP into a microcontroller (MCU) is another way to keep snooping eyes at bay. A microcontroller is a single chip computer with a CPU, program storage and data storage all in one device. Many microcontroller products offer a locking mechanism that prevents your software IP from being read outside of the device. Your competitors can see the inputs and the outputs, but do not know what steps your software is making internally.
When using an MCU, the level of IP protection is strong, but not to the extent of using a custom IC. Due to a fixed architecture of internal resources, someon e could look at the signals going into the MCU and determine the resources on the other end of the connection thereby deriving some aspects of your solution.
Various types of programmable logic devices, such as CPLDs and FPGAs, enable you to develop a custom digital logic. Some of these devices have internal non-volatile memory for storing the programming information, while others require a separate boot memory. The devices equipped with internal memory are designed to lock information inside. Because their internal architecture is defined by each designer, it is virtually impossible for a competitor to derive your IP from inside this black box.
The newest generation of devices, which require external memory, allow you to specify an encryption code to offer the same high level of protection as a custom IC. Overall, these programmable logic devices score very well on their level of protection they provide, as well as the level of flexibility they offer allowing you to re-use your IP at any time. However, these devices are unable to provide an equivalent level of protection and flexibility for analog signals, because they are "digital only" devices.
A new type of programmable technology, called Programmable System-on-Chip (PSoC), gives you the best combination of all these benefits. PSoC devices allow you to make your own unique architecture by combining analog and digital circuit elements - just like in a custom SoC. The technology not only allows you to define analog and digital building blocks, but also interconnect them in unique ways to best fit your application.
This gives you all of the flexibility and inherent protection of the custom solution, without the drawback of long lead-times and high NRE. At the same time, you do not have to transfer your IP to a supplier and incur the penalties of mid-cycle design changes.
Designers have many different legal and technological methods at their disposal to protect their intellectual property. Each has i nherent advantages and disadvantages. Two attributes to consider if you have this type of decision to make are the level of protection for your IP that is provided, and the level of flexibility you have to migrate this IP into future products. T
The PSoC method provides a level of protection that is better than any other programmable device and only slightly less than a full custom. At the same time, you retain complete control over your IP, and can utilize it as you see fit, a level of flexibility that is equal to any programmable technology, and far better than the flexibility found in a custom device. If intellectual property protection is a priority at your company, you may want to investigate this technology for future designs.