TECHNOLOGY IN CONTEXT
Hybrid Architectures Tackel Multiple Challenges
Wireless Microcontrollers Combine Low-Power Processing and RF Connectivity in a Hybrid Architecture
The ability to configure a hybrid MCU makes possible a variety of strategies to optimize RF connectivity while saving power and extending battery life in mobile connected devices.
NISHA GANWANI, SILICON LABORATORIES
Page 1 of 1
Microcontrollers (MCUs) are indispensible in the world today. Their applications range from simple alarm clocks and coffee makers to complex electronics systems that control automobiles and airplanes. In the broad spectrum of devices driven by MCUs, one encounters a plethora of battery-operated products that pose the additional requirement of low power operation and in some cases wireless connectivity. Prime examples of low power embedded systems requiring an RF link include applications in home and building automation, wireless security, smart utility metering, asset management and portable medical systems.
To address the requirements of these applications, wireless MCUs have emerged in recent years as an embedded processing solution with a hybrid architecture that integrates an MCU core and wireless transceiver into a single device. This single-chip solution offers several advantages including reductions in BOM cost, component count, board space, power consumption and overall design complexity.
Reducing Power Consumption
One of the most important design considerations for battery-operated systems is reducing power consumption, which increases the battery discharge time and extends the operational life of a system. RF transceivers, however, require significant power to form data packets and communicate over long distances. A wireless MCU can implement features in a hybrid architecture to reduce overall power consumption without compromising RF performance.
Thanks to continuous improvements in process technology, embedded processors and RF transceivers are now readily available in device geometries scaling to 180nm and even smaller. The operating voltage of the internal circuits scales with the geometry, dropping to 1.8V and even lower. However, batteries continue to have a terminal voltage of 3.3V or even 3.6V. A common design methodology to support these terminal voltages is to integrate a low drop-out voltage regulator (LDO) on-chip. This structure takes the battery input and regulates the internal voltage of the chip to something lower, i.e. 1.8V or less. However, using a linear regulator to step a 3.6V input down to a 1.8V output has at best 50 percent conversion efficiency. This efficiency gets worse as the output voltage decreases.
More advanced embedded controllers have integrated switching regulators with much higher efficiency than their LDO counterparts. In many cases, these devices can have switching efficiencies exceeding 85 percent. This high efficiency has the effect of reducing the total current sourced from the battery and extending battery life.
Figure 1 illustrates the benefits of applying an advanced MCU with a DC/DC converter to a battery-powered wireless application. In this example, we greatly reduce the battery drain due to the transceiver: (Formula 1)
In other words, using a DC/DC converter reduces the current drain from the battery for use by the radio to approximately 62.5 percent compared to using a simple LDO. This approach has the net effect of reducing the radio power consumption by the same amount.
Another power saving strategy can come in the form of a data packet processing engine (DPPE). In traditional wireless systems, several CPU processing steps are required to prepare the suitable RF packets for communication. For example, a common requirement is to encrypt the data to secure transmission over-the-air. The Advanced Encryption Standard (AES) is a universal specification for encrypting electronic data. The strength of the AES algorithm depends on the number of bits used for encryption, i.e., 128, 192 or 256 bits. Traditionally, an increase in the number of bits translates to an increase in the power used by the CPU for AES processing.
Many RF protocols require a cyclic redundancy check (CRC) for detecting errors during transmission. Additionally, certain wireless protocols, such as the wireless M-Bus stack used for metering in Europe, require encoding the bit stream to eliminate the DC component (e.g., a string of 0s and 1s). This ensures frequent signal transitions directly proportional to the clock rate and helps in clock recovery.
An advanced MCU design may include hardware blocks for each of these functions, along with a Direct Memory Access (DMA) peripheral to perform a complete radio transaction without CPU intervention. A complete DPPE in hardware can accelerate packet processing by up to five times while consuming half as much current as operations performed in software. The result is power savings of up to 10x for radio packet generation (Figure 2).
The benefits of using a DPPE can result in significant power savings.
In certain wireless sensor applications, particularly utility meters, a device called a register encoder records the flow of natural gas or water. In a metering system, this flow can appear electrically as a series of switch closure events or pulses. In a traditional system, the CPU must wake up and sample the state of an I/O pin to determine if the switch is open or closed. If it is a physical reed switch, additional CPU bandwidth is needed to de-bounce the switch and manage pull-up resistors to guarantee it is a valid pulse as well as to minimize the current drain through the closed switch. Performing this function in software, even in the most optimized system, can consume well over 1 µA.
A better approach is to integrate a dedicated input capture timer on the MCU that can operate autonomously while the CPU is inactive. This technique has a number of advantages over a software-based approach. For example, the switch closures can be accumulated in a hardware register requiring little if any CPU intervention. Additionally, features such as switch de-bounce, pull-up resistor management and self-calibration can be integrated directly in the hardware. With two timer inputs, quadrature decode functionality can be supported to determine flow direction. This arrangement provides the capability of back-flow detection as well as an anti-tamper provision. A dedicated low-power input capture timer can consume less than 400 nA at 3.6V even with a sampling rate as high as 500 Hz. This is a significant improvement over performing this function in software and reduces the wireless meter’s overall power consumption.
An ultra-low-power MCU is designed to have extremely low standby currents with a real-time clock (RTC) running. Standby currents for a transceiver are quite excessive in comparison. When integrating the MCU and the transceiver in a hybrid architecture, the latter device can be configured to go into complete shut-down mode when not in use, thereby saving power. In this scenario, the MCU’s RTC block can be configured to wake up the radio as required for operation.
In a wireless MCU, a serial interface (such as SPI or 3-wire mode) can be internally connected between the MCU and a transceiver. The hard-wired interface provides access to the radio’s peripheral registers from software executing on the MCU core. This approach accelerates system design and firmware development. A highly integrated chip with fewer external pins reduces layout complexity as well as decreases package cost and BOM count, enabling a more cost-effective solution than is possible using discrete components.
Multichip Wireless MCU Module vs. Monolithic CMOS Chip
Building a monolithic CMOS chip that integrates the MCU and radio operations on a single die poses several challenges. A considerable design challenge is finding an optimum process that is suitable for both the flash memory and processing functions of an MCU as well as RF operation. A single-chip solution may drive an IC designer to pick a process optimized for one of the functions while taking a cost and/or performance hit on the other. Another significant challenge is the impact of the processor’s digital blocks on the transceiver’s RF performance.
In a multichip module (MCM) approach, the impact of noise from the MCU’s digital circuits on the RF frequency spectrum is minimized. Physical distances ensure that the MCU’s clock frequencies do not cause spurs and/or blocked channels on the radio. The impact on critical RF specifications such as sensitivity and range is also minimized, ensuring interoperability without compromising radio performance.
A MCM approach provides numerous benefits for a hybrid wireless MCU solution, including lower power consumption, reduced cost, complexity and BOM count—all with continued exceptional RF performance.
There are many critical design specifications for a battery-operated wireless embedded system including RF frequency, range of communication, type of modulation, software stack implementation and required operational life of the system between battery replacements.
Silicon Labs’ EZRadioPRO sub-GHz transceivers, for example, provide RF output power of +20 dBm, an RF sensitivity of -121 dBm and an RF link budget of >140 dBm. These transceivers support a range of modulations, e.g. On-Off Keying (OOK), Frequency Shift Keying (FSK) and Gaussian Frequency Shift Keying (GFSK).
Silicon Labs’ ultra-low-power C8051F96x MCU products have been designed from the ground up to seamlessly interoperate with EZRadioPRO sub-GHz transceivers. These MCU products have flash sizes up to 128K and RAM sizes up to 8K, and can implement standards-based software stacks such as wireless M-Bus and 6LoWPAN, as well as proprietary RF protocols for wireless communications. These low-power MCUs have been designed to provide ultra-low-power consumption in active, sleep and deep sleep modes.
The Si102x/Si103x wireless MCU family (Figure 3) pairs the ultra-low-power C8051F96x MCU core with the EZRadioPRO transceiver into a single-package solution that addresses the requirements of low-power embedded systems that use an RF link. By reducing overall system power, these hybrid devices enable developers to either reduce the size/cost of batteries or increase battery life while delivering end products with industry-leading RF performance.
Example of Wireless MCU Hybrid Architecture.