Chirp Adds Sonar-on-Chip to ToF Battle

　　Whether they appear in drones, robotic vacuums, VR/AR headsets, smartphones or cars, sensors are proliferating in consumer and industrial systems, to capture “accurate range and position measurement” information.

　　Depth information is critical in an imaging world when things are fast transitioning from 2D to 3D sensing. Knowing accurate distances, for example, can augment position tracking in VR/AR headsets, and prevent robotic vacuum cleaners from running into walls or objects.

　　Although the market has already seen time-of-flight (ToF) sensors built on different technologies — such as IR and optical — Chirp Microsystems believes its new MEMS-based millimeter-sized ultrasound chip will become an effective alternative to the competition. Chirp boasts that it operates at ultra-low power with high precision range. It stands up to harsher environments while working both in total darkness and under the full sun, according to the company.

　　IHS Markit, currently preparing a report on ToF sensors, told EE Times that shipments “started to become relevant in 2016,” and then growing quickly. The market research firm predicts annual shipments to expand to 600 million units in 2019.

　　Michelle Kiang, Chirp's CEO, first talked to EE Times earlier this year about her Berkeley, Calif.-based startup and its ultrasound sensing technology. Since then, “We’ve been busy adding a new product — a longer range sensor device (whose operating range extends to 5 meters) and the development of sensing solutions including software and system integration,” Kiang said. Chirp’s ultrasonic ToF sensors, the CH-101 and CH-201, are both sampling today.

　　How does Chirp stack up?

　　Manuel Tagliavini, principal analyst for MEMS and sensors at IHS Markit, told us, “The Chirp solution has its working principle based on a reflected ultrasonic wave, while the standard ToF devices are based on the measurement of the time a laser pulse requires to be reflected.”

　　Depending on its final application, the ultrasonic solution has strengths and weaknesses, he noted. For example, the wider field of view enabled by the Chirp’s piezo technology is advantageous for IoT appliances like the Amazon Echo, in sensing the user’s presence in a room, he said. But an ultrasonic solution applied to the AF system for cameras, for example, Tagliavini said, provides a general scene distance without necessarily measuring “the single details and object’s distance.”

　　According to Chirp, its ultrasonic ToF sensors offer 180-degree of Field of View (FoV), compared to the 20-25 degrees offered in IR-based solutions. Kiang told us that while IR-based ToF sensors are mainly used for front-facing applications, ultrasonic sensors are more flexible when designed into a system. “You can install it in a non-front facing panel and you can still see a lot," Kiang said.

　　Comparing Chirp’s solution with optical ToF sensors, Jean-Christophe Eloy, President & CEO, at Yole Développement, agreed its main advantage is in the viewing angle. “The field of view of ultrasonic solutions is much larger than optical solution. The price is low, so it is easy to integrate multiple sensors.” He believes the ultrasonic solution, initially, “will be used for applications where optics is difficult.”

　　Two ultrasonic devices

　　Chirp is coming out with two ultrasonic ToF sensors. Kiang calls the CH-101, designed for a one-meter mid-range distance, “good for arm’s length applications” such as AR and VR.

　　The company’s CH-201, designed for a distance up to 5 meters, is for “room-scale applications,” she explained. It will apply best to home security applications such as Nest Cams or digital voice assistants as it can sense a human presence in a room.

　　Both the CH-101 and CH-201 come in a system-in-package that integrates a piezoelectric micromachined ultrasonic transducer (PMUT) and an ultra-low power SoC.

　　The SoC is there to run Chirp’s own ultrasonic DSP algorithms, such as rejecting noise and processing detected signals, explained Kiang. Asked whose microcontroller or DSP cores Chirp is using in its SoC, the CEO said that both are Chirp’s custom cores. “This SoC part is essentially an ASIC with a lot of hardware functions integrated.” The combined SiP — ASIC and MEMS (a transducer) part — receives and processes signals, and pipes that data to where it needs to be, she added.

　　Chirp can factory-program CH-101 and CH-201 to enable different functions, such as time-of-flight range-finding, proximity sensing, human presence sensing, the company said.

　　Chirp also stressed that the ultra-low power consumption of its ToF sensors quoted in the company’s data sheets includes “everything” — power needed for both MEMS and data processing. The company said that the CH-101, operating in a range from 1 cm to 1.2 m, consumes 15 ?W (10 cm max range) and 27 ?W (1.2 m max range) at 1 sample per second. Operated at 30 samples per second, it consumes 70 ?W (10 cm max range) and 340 ?W (1.2 m max range).

　　Both CH-101 and CH-201 use an ASIC with the same interface and pinouts, so that a drop-in replacement is possible, noted the Chirp CEO.

　　Why go after the smartphone market?

　　In announcing the new ultrasonic sensors, Chirp described smartphones as “another important target market.”

　　Acknowledging that Chirp’s ultrasonic sensors initially generated a lot of interest among smartphone vendors, Kiang said she was reluctant to test the market largely because smartphone manufacturers wanted Chirp’s solution at the same, already commoditized price as that for IR-based sensors, even though ultrasonic sensors function better.

　　Consider, for example, the “black hair issue,” she noted. When an IR-based proximity sensor gets closer to black hair or black material, it becomes unreliable, sometimes prompting the user’s cheek to start dialing a smartphone. Ultrasonic sensors have no issues with black hair.

　　Second, smartphone vendors’ interest in ultrasonic sensors has grown as handset designers seek to get rid of optical windows from a handset. Smartphone manufacturers, who now regard an optical opening as “an eyesore,” prefer ultrasound sensors to IR, Kiang said. “But they told us they wanted our solution at the same price as IR sensor. Naturally, we said, ‘No.’”

　　Today, the smartphone design trend has gone even further. The goal now is an all-glass, full-screen, no-bezel display. “The CH-101allows product designers to remove the optical proximity sensor from the front of the phone to create a bezel-less display,” Kiang explained. “That’s because the CH-101’s wide field-of-view allows it to measure range even when it is mounted on the top or bottom sides of the phone. Since this is the location of the microphone and speakers, it’s easy for designers to add another acoustic sensor at these locations.”

　　This time around, smartphone manufacturers appear willing to pay more to make the no-bezel screen possible.

　　While acknowledging a growing trend for bezel-free screens, Frédéric Breussin, MEMS and sensors business unit manager at Yole Développement, is skeptical. He observed that the focus in mobile phones, at the moment, is “really dedicated to optical functions that are easily integrated below the glass surface.” Breussin suspects, “Pushing the mobile phone players to look at ultrasonic sensors may be difficult.”

　　IHS Markit’s Tagliavini remains more optimistic. He expects Chirp’s ultrasonic sensors to get their shot at the smartphone market. Further, he noted, “In the long term the ultrasound solution could enable new functions, for example the 3D reconstruction and mapping of indoor environments with low power requirements.

　　Compete with ST and Ams?

　　As companies such as STMicroelectronics and Ams pour resources into leading smartphone vendors’ demand for 3D sensing cameras, how will Chirp compete?

　　Kiang said, “We get that question a lot.” But she stressed, “Look, we are not in a camera business. We aren’t competing with them directly on 3D camera. Rather, we’d like to think there is synergy between us.”

　　Today, Apple is using a “very expensive piece of technology” for its TrueDepth module in iPhone X, she explained. “However, we think there are other ways to add the third dimension to the 2D image sensors.”

　　Chirp is working with a CMOS image camera company to add depth-sensing, said Kiang. But she declined to name the partner or reveal its progress.

　　Market predictions

　　Yole offers ToF sensor market predictions on a granular level, dividing the market into several different segments, including simple ToF sensors, complex ToF sensors for movement detection, complex ToF sensor for 3D imaging for mobile phone applications and others. Here’s Yole’s forecasts:

• Simple ToF sensor for proximity sensing and laser ranger (this is the existing device from STM, Sony, AMS…) for mobile phone: market potential of $7000 million (see slide below) in 2022

• Complex ToF sensor for movement detection (and analysis of in cabin movement) in automotive applications: market potential could reach $200 million minimum in 2021

• Complex ToF sensor for 3D imaging for mobile phone applications: it could replace SL of Apple. It is a $6 billion market opportunity in 2022 at camera module level (with the optics…). But Yole highlights a very strong technology competition.

• Complex ToF sensor for 3D imaging for industrial, medical…: about $600 million in 2022, according to Yole

• So in total, the ToF market potential should reach around $7.5 billion, with a very strong competition with other 3D sensing technologies (like structured light) or simple proximity sensing/range finder