Internet of Things Connecting quickly and economically to the Internet of Things
The IoT allows devices to be connected wirelessly so we can interact with them anytime, anywhere. There are numerous use cases where, with just a press of a button or a voice command, devices are connected to the internet. IoT technology introduces new and innovative features to devices that have been around for decades while at the same time defining entirely new markets like smart water bottles that track your water drinking habits. Sachin Gupta, staff engineer product marketing and Richa Dham, senior marketing and applications manager, Cypress Semiconductor talk about some of the issues of wirelessly connecting devices
W
hile connecting all these devices makes our lives easier, adding connectivity brings a whole new set of challenges to system design. Most of the wireless technologies that are enabling IoT are in the crowded 2.4-GHz spectrum. This requires wireless design that is very specialised and non-trivial as it needs to operate reliably with high interference levels while still being low power. In addition, traditional appliances/system manufacturers are unlikely to have wireless system design expertise in-house. Because of the nature of these designs, volumes of IoT products tend to be very unpredictable. In such cases, it is important to make business decisions that ensure profitability. One such key decision is whether to invest heavily in in-house wireless-expertise or buy the technology as a standalone module. Both paths are viable, depending upon a manufacturer’s market goals and ability to invest. For many companies, the module approach is an attractive way to add connectivity to a system quickly without significant investment of time and cash.
The complexities of wireless connectivity In IoT, the most commonly used wireless technologies are Bluetooth Low Energy (BLE), Bluetooth classic, and Wi- Fi. However, BLE is quickly dominating the market for applications such as Human Interface Devices (HID), wearables, home automation, and asset tracking. There are various reasons for this, including BLE’s lower power consumption, lower protocol overhead,
and near-ubiquitous availability of Bluetooth on devices like mobile phones and personal computers. Wi-Fi, in contrast, is mainly used for connectivity to the internet and so serves as the main gateway technology to connect to the cloud. From a system design perspective, adding wireless connectivity (BLE or Wi-Fi) is not as straightforward as adding a wired communication interface like USB or UART. When a wireless interface is added to a system, there are several factors that drive end-system cost and complexity. Some of these factors are explained below. Board design complexity: Wireless systems require an antenna and matching network that add complexity to board design. Depending on the use case, an omnidirectional or more complex beam forming antenna might be required. The power supply is critical for stable system operation, and extra care is needed to ensure these circuits do not interfere with antenna performance. Another very sensitive part of the RF subsystem is the clock circuit. Most wireless protocols rely on accurate frequency matching, and any drift in the clock can mean that Rx and Tx are out of sync, resulting in data loss. It is commonly accepted that experienced RF layout engineers are needed to ensure a system will be able to achieve its expected performance. If board layout is not done properly, this may result in power supply noise, an unstable crystal oscillator, very short range, and a system that does not meet RF regulatory requirements. Thus, the product could fail commercially. However, to maintain an in-
Figure 2: High-level block diagram of a BLE module
house RF design expert can be costly as the skills required are very specialised. RF regulatory certifications: Every country has one or more RF regulatory certification requirements for wireless equipment. These certifications verify that devices maintain radiation levels within set limits and the operating frequency does not fall within any prohibited frequency band for a given application. These certifications require extensive testing. Generally, these tests are carried out at labs approved by the regulatory committee. Each time a device goes in for testing, a fee is levied. If the device does not pass the test, the fee will need to be paid again when the device is retested board reworking. Product qualification: To call a
product compliant to a given standard say Wi-Fi or Bluetooth, it needs to be qualified. That adds another cost. For example, a BLE product needs to undergo testing at a Bluetooth SIG qualified test facility to be able to use the Bluetooth trademark on its product (after the declaration and listing process is completed). There is an inherent advantage with product qualification including interoperability and a stamp that your product will be interoperable with that particular qualification.
Declaration and listing This is another cost that results from listing the product on the protocol consortium’s website and to legally use the brand name on your product. This cost is mainly required for using the branding advantage of the specific consortium. Figure 1 shows the product qualification, declaration, and listing flow for a Bluetooth product. Out of all these, the first three costs can be lowered by
using a module instead of silicon to add wireless connectivity. Figure 2 shows the high-level system block diagram of a BLE module. As shown in the diagram, the module includes crystals and an antenna that reduce board design complexities for system designers. A module is also the quickest and easiest way to add connectivity to a system. There are multiple choices available for any form of wireless connectivity; hence, selecting a module can be a big decision. Given the needs of varied applications – some are power sensitive and others need longer range – understanding the factors that impact module choice at an early stage is very important. From a system perspective, you can choose a module that meets system requirements. However, one must also look at the IDE that supports the module that you have picked and understand the migration cost from module to silicon when volumes go high and you want to move to silicon to produce a system that is more cost effective. The backbone of such a migration is the IDE that is used to develop firmware around module and silicon. For an easy transition, both must be supported in the same IDE without requiring substantial changes in firmware. It is also important to have support for vertical migration in case more features are to be added at a later stage of time. Though modules seem appealing for all applications, they may only be cost- effective up to volumes of around 250k units, depending upon your application. The reason is that at this point, the advantages of using silicon begin to outweigh the higher cost of modules. Thus, silicon becomes the preferred choice when volumes start going beyond 250k units unless there is not much price pressure and a company does not want to invest in additional head-count and certification process.
Figure 1: Product qualification, declaration and listing flow for a Bluetooth product 26 October 2018 Components in Electronics
www.cypress.com
www.cieonline.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68
