Silicon Labs SiWx917 Wi-Fi 6 SoC Tested to Deliver Robust Connectivity and Multi-year IoT Battery Life

Battery life is one of the first characteristics buyers compare on wireless IoT devices. The longer it is, the better. According to the interoperability and power consumption test done by Novus Labs, the SiWx917 Wi-Fi 6 SoC proved to deliver robust, energy-efficient connectivity even in extreme congestion scenarios. As a result, SiWx917 was estimated to provide smart locks with up to 5 years of battery life, giving device makers an invaluable competitive advantage. Read this blog to learn more about SiWx917 power consumption, throughput, and battery life in different network congestion scenarios.

Wi-Fi IoT Devices and Connected Sleep Mode

Wi-Fi IoT devices such as smart locks, security cameras, and sensors spend most of their time sleeping while being also associated to the Wi-Fi access point (AP) and connected to the cloud. This so-called connected sleep mode (a.k.a. WLAN associated mode) reduces IoT device’s power consumption and extends battery life while keeping the device reachable from the Wi-Fi network, i.e., the cloud application and user can start data communication with the device whenever needed. Typically, the active communication period lasts just seconds or minutes after which the device resorts back to the connected sleep mode. Because of the very “sleepy” IoT duty-cycle, the device’s ability to stay connected to the Wi-Fi AP at a low power consumption in the connected sleep mode is the most critical factor affecting how long the device’s battery lasts.

SiWx917 Wi-Fi 6 SoC

Silicon Labs SiWx917 Wi-Fi 6 SoC is renowned for its exceptional energy efficiency. Its 22µA power consumption for Wi-Fi 6 connected sleep mode (TWT 60s) is among the lowest in the industry. Its combined system power for Wi-Fi 6 connected sleep and MCU is just 37µA. That’s less than any other Wi-Fi product in this market segment at the time of writing.

However, measurements in ideal conditions with a single Wi-Fi AP do not give the full picture of energy efficiency. Wi-Fi channel congestion and AP models can affect interoperability and power consumption.

To get a more holistic view on how SiWx917 performs in different network congestion scenarios, how it interoperates with different AP models, and how these aspects translate into battery-life, we hired Novus Labs, the global independent wireless engineering specialist house, to put SiWx917, our flagship Wi-Fi 6 SoC through their most rigorous wireless test program.

SiWx917 Wi-Fi Interoperability and Power Consumption Test

The purpose of the test was to find out how well SiWx917 performs in terms of connection robustness, power consumption, and throughput across 100 popular, globally used Wi-Fi AP models.

For each AP a 25-minute test was conducted to measure robustness and power consumption under different network congestion levels: 0%, 20%, 40%, 70%, 90% - 5 min each. The end-results have been tabulated by Novus Labs and can be used by IoT device makers evaluating Silicon Labs SiWx917 as a measure of robustness to compute battery life under various real-life conditions and predict variations across APs in their use case.

The following sections cover the three tests conducted by Novus Labs:

Test 1: SiWx917 Power Consumption in Standby Associated Mode

Novus Labs’ test 1 measured the power consumption of SiWx917 in the standby associated mode (connected sleep). It was conducted over multiple weeks against 100 APs with a 1-sec listen interval (DTIM10) and 55-sec MTTQ keep alive (no TWT). The results clearly show that SiWx917 sustains its superb energy efficiency across the tested APs, even under the harshest channel congestion scenarios.

The graph below shows the average current consumption for each congestion scenario (0%...90%) tested across all the Wi-Fi 4 (802.11n) and Wi-Fi 6 (802.11ax) APs with a 1-sec listen interval (DTIM10) and 55-sec MTTQ keep alive.

Results: The average power consumption in the clean channel scenario was just ~90µA while the ‘close to saturation’ traffic scenario power consumption resulted in ~217µA, which is still highly efficient considering extreme channel congestion. All tests were passed with 0 disconnects.