The Intersection of Semiconductor Design & IoT: Opportunities and Challenges

The Intersection of Semiconductor Design & IoT: Opportunities and Challenges

The Intersection of Semiconductor Design & IoT: Opportunities and Challenges

Nilesh Ranpura, Director, eInfochips, 0

An IIM Ahmedabad alumnus, Nilesh has been associated with eInfochips for over 25 years now, prior to which he held the role of R&D Engineer, VLSI engineer and Director Engineering [ASIC] at eInfochips. In a recent interaction with Bimlesh Prasad (Correspondent, CEO Insights), Nilesh shared his valuable insights on the impact of semiconductor design on IoT and many such interesting aspects. Below are the key excerpts from the exclusive interview –

How is designing semiconductors impacting IoT in recent times?
Semiconductor design has undergone a serious metamorphosis over the last two decades. Today, it has options to add more features onto the chip, increased its speed and cut-down on its power consumption. IoT has the requirement of multiple wireless protocols, interface processors and connectivity to the cloud. Semiconductor design offers IoT the ability to consume less power and the flexibility to design complex and high purpose solutions. As a result, it is safe to say that IoT applications would not have been widely successful and accepted if the semiconductor industry had not evolved over the last 20 years. Additionally, every wireless protocol and process performance that is changing in the IoT domain will require constant improvement every year over speed, functions and many other aspects. Semiconductor design offers these variations much smoothly and at a faster rate.

Most of the businesses today grapple with the critical challenge of Time-to-Market. Due to the highly competitive nature of the current market across almost all verticals, businesses are constantly looking for ways to keep their TTM as less as possible to stay ahead in the market. Semiconductors ensure that IoT products are equipped with the most advanced technologies and consume less power. Also, recent tech advancements have also enabled semiconductor chips to become smaller and more compact, which will act as a great advantage in the final product production.

Could you tell us how semiconductor designs can address security and performance related challenges in IoT.
IoT implements multiple wireless protocols and has a lot of connectivity options with other elements such as the cloud. Once the communication comes into the outside world, a key aspect that needs to be taken care of is the security of data. There are open security standards and standardized security algorithms which are mandatory for secure communication across all interconnected devices. IoT being a non-computer device, the security measures are implemented within both software and hardware layers. For instance, security algorithms such as AES can be directly implemented on the semiconductor, making them more secure and faster. On the other hand, today, semiconductors are designed to ensure that their speed/performance can be doubled every 18 months. The perfect example for this is the
evolution of 2G connectivity to 3G, 4G and now 5G, which showcases enhanced speed as well as performance.

Shed light on the role of semiconductor design in ensuring the integrity and confidentiality of IoT data.
IoT security is a prime concern which can be taken care of by implementing tools such YubiKey, which is a hardware based encryption tool regarded as being highly secure in the IT industry. Any IT industry device goes through multi-factor authentication, which is possible only with the presence of hardware either in the form of random number generation on the key, biometric fingerprint or any other secure mechanism. The core of making all this happen is the semiconductor. Also, data and privacy across devices like the computers and laptops are protected through firewall algorithms and architecture, which are again implemented on hardware, i.e. the semiconductor chip. To conclude, semiconductors offer the most secure hardware implementation configured through its memory comprising configurable software as a UI to change the configuration. While semiconductor design gives complete framework for security and privacy, the percentage of security and privacy, that has to be given, is decided by the software.

Organizations must be prepared for leveraging these emerging developments to evolve to the next level by applying IoT-ready frameworks.

Tell us about a few emerging trends resulting from semiconductor design-IoT intersection.
IoT design started with a small embedded hardware coupled with wireless protocols and connectivity to cloud as a level-1 device. This has worked very well in the case of applications such as washing machines, set-top boxes or any agriculture devices. Now, we see that the intersection is happening in such a way that everything is evolving to the next level of IoT implementation. For instance, there are many POS devices in banks, malls, hospitals and many other setups. These are few of the key areas where we can see the intersection of semiconductor and IoT. Another major area is robotics in industry 4.0, which has many appliances and machines which will have interaction of IoT and semiconductors. Organizations must be prepared for leveraging these emerging developments to evolve to the next level by applying IoT-ready frameworks to the devices, making them more digitally controlled.

How should IT bridge the gap between semiconductor design and IoT integration for seamless collaboration?
Be it semiconductor design or IoT design, IT acts as the spinal cord of any digital implementation across any organization. The first thing IT must address in this regard is the large server farm implementation to process and store data, which will be crucial in deriving meaningful insights for IoT applications. Secondly, IT works very well pertaining to issues concerning supply chain. While semiconductor design is largely supply chain driven, IoT is majorly implementation-driven. Thus, the supply chain must be carefully tracked and managed, which is possible only through IT. Furthermore, IT has to handle a lot of memory issues. For instance, since semiconductor and IoT design generate massive amounts of data, IT has to bridge the storage issues by doing centralized storage or distributed storage; sync it and creating backups for utilizing the data for AI models or normal usage.