Dialog and Arm collaborate to deliver lowest power Wi-Fi chip in the industry – Internet of Things (IoT) blog – Arm Community blogs

Dialog and Arm collaborate to deliver lowest power Wi-Fi chip in the industry - Internet of Things (IoT) blog - Arm Community blogs

Short battery life of IoT devices is a key bottleneck to adoption. According to an IDTechEx report, at least 80% of the potential for IoT will be denied because of the battery replacement requirement of IoT devices. The goal of IoT is to connect devices, for greater convenience, insight and control, with more efficient operation. However, these benefits are compromised if these IoT devices require regular battery maintenance or a permanent connection to a mains power supply.

IoT device makers understand this problem quite well. If you are a device maker, the short battery life of an IoT device can hurt your brand, annoy your customers, and make your product look unfit for purpose. So, optimizing battery life without sacrificing the core features of an IoT product is a key challenge for product makers.

To address the problem of frequent battery replacement in IoT devices, Dialog has developed the DA16200, the world’s lowest power Wi-Fi SoC leveraging Dialog’s VirtualZero technology. The DA16200 specifically pushes the technology limits to reduce the power consumption of Wi-Fi devices more than ever before. As a result, Wi-Fi-based IoT products can now offer multi-year battery life like Zigbee and Z-wave based sensors. Given the ubiquity of the Wi-Fi standard and the fact that Wi-Fi devices do not need to connect with a home hub, this low-power Wi-Fi technology can simplify smart home architecture and increase the adoption of wireless products in IoT (See Fig.1).

Low power Wi-Fi simplies the smart home

Fig.1: How Low power Wi-Fi simplifies the smart home

DA16200 IoTMark-Wi-Fi Results

Low-Power Wi-Fi is an emerging IoT segment. Although each chip vendor provides power consumption numbers in the datasheet, so far, there was no benchmark that would provide battery life insights to a potential customer in real-world scenarios. To provide a comprehensive assessment of the energy efficiency of embedded platforms, EEMBC’s IoTMark -Wi-Fi has recently developed the first industry benchmark that emulates the real-world behavior seen by battery-operated wireless devices. The benchmark for IoT connectivity is targeted at always-connected devices that are expected to have long battery life and reasonably low Wi-Fi throughput demands.

The DA16200 platform has achieved the highest score on the IoTMark -Wi-Fi benchmark which is at least 50% better than the closest competitor in the market. DA16200 has achieved a score of 815 which roughly translates to 815 days (2+ years) of battery life for a product running on 2 AA alkaline batteries. This is the first time in industry that a Wi-Fi chip has enabled a multi-year battery life for IoT devices. We are proud to have achieved these results and we believe that it results in an accelerated adoption of IoT as we add ultra low-power features on top of simplicity and ubiquity of Wi-Fi protocol.

DA16200  Core Technology:

DA16200 optimizes sleep and wake-up times using its proprietary dynamic power management (DPM) algorithms built upon standard Wi-Fi features. The technology offers three sleep modes depending on the use case requirements of the IoT device makers:

  • Sleep mode 1 (Unconnected Sleep) is the lowest power operation mode with only 0.2 microamperes (μA) current. In this mode, most of the chip blocks are powered off and the chip doesn’t keep the networking time. The SoC can be woken by an external trigger delivered to the chip’s wake-up pins or its digital and analog IOs.
  • In sleep mode 2, Connected Sleep, the RTC functionality is retained while consuming the only 1.8uA. The SoC wakeup time is less than 100ms in response to an external event or completion of a programmed internal timer.
  • In sleep mode 3, always-connected Wi-Fi mode ensures wake up of less than 2ms upon detection of an incoming Wi-Fi data packet while consuming less than 50 μA average current. The chip regularly checks for standard Traffic Indication Map (TIM) or Delivery Traffic Indication Map (DTIM) information elements embedded in Wi-Fi management frames, it then wakes up to begin processing normal 802.11 traffic, like any network station when required. More than 200 commonly used Wi-Fi access points are tested for DPM algorithms to ensure the low power performance of DA16200 on any Wi-Fi network.

DA16200 DPM Technology

Fig.2: Introduction to DA16200 DPM Technology

Another aspect of the DA16200’s low-power architecture is the use of an embedded processor that is low power, while providing the processing capabilities needed in Wi-Fi applications. The Arm Cortex-M4 processor provides architectural sleep features to support the sleep modes mentioned. The efficient instruction set provides excellent program execution performance, high code density, as well as wide range of computation support feature such as a floating-point unit and bit field manipulation instructions. Our VirtualZero Technology combined with the low-power features of Arm Cortex-M4 processor has allowed us to build the lowest power Wi-Fi platform in the industry.

As short time to market is a key concern for developers, we ensure that our unique power-saving features do not increase the development effort for our customers. DPM operation and the details of power management remain abstracted from the developers which simplify the process of product development. To enable developers to quickly prototype IoT designs using DA16200, widely used FreeRToS and gcc based Wi-Fi development kits are available. These kits combine a Wi-Fi module with a USB interface, keys, power profilers, debuggers, and connectors speeding up the development and debugging of DA16200-based designs.

For more information on how to take advantage of the lowest power SoC to simplify the smart home architecture, visit

Source link

What do you think?

Written by Ola


Leave a Reply

Your email address will not be published. Required fields are marked *

ten + 18 =


Big Data Drives Office Productivity into Overdrive - SmartData Collective

Big Data Drives Office Productivity into Overdrive – SmartData Collective

Computer Vision to Detect License Number Plate - Analytics Vidhya

Computer Vision to Detect License Number Plate – Analytics Vidhya