aPm EDGE | Ambient IoT

The Ambient IoT aPm Edge device is a state-of-the-art, sleek, and disposable hardware solution designed for modern IoT frameworks.

Approximately the size of a credit card, this multifunctional device includes multi-sensing and multi-connectivity capabilities for continuous environmental monitoring across diverse environments.

It features an integrated neural decision processor, enabling advanced machine learning models to operate efficiently at the edge..

Key Features

Credit card size (86mm x 54mm x 5mm)

Energy harvesting ( Light, RF, Movment)

On-Board AI

Multi-Sensing (Temperature, Humidity , Pressure, Movement , Air Quality , Light )

Multi-Connectivity ( Terrestial & Satellite)

External sensor Integration ( Wiliot, inPlay, Ambient Iot)

NFC

External Display

Compact Size

IP65 Rated

Firmware OTA Supported

Highights

Compact Design: With a size equivalent to a credit card, the APM Edge device is easily deployable in various settings, ensuring minimal footprint and unobtrusiveness.

Continuous Connectivity: The Altair 1250 NB-IoT/LTE-modem ensures that the APM Edge device is always connected to the APM Edge Cloud, providing real-time data transmission and processing capabilities.

Syntiant EDGE ML The Syntiant® NDP120 Neural Decision Processor™ is a special purpose chip that runs machine learning models on edge devices, enabling real-time sensor data processing for applications like pharmaceutical cold chain and fresh produce supply chains, thereby enhancing their efficiency and reliability.

Atmosic MCU: Serving as the central processing unit, the Atmosic MCU seamlessly controls all the chipsets within the device. It ensures efficient functioning, power management, and effective coordination between different hardware components.

Interactivity with Ambient IoT Sensors: Beyond its primary sensing capabilities, the APM Edge device is also equipped to read environmental data transmitted by nearby Ambient IoT sensors. This feature allows it to serve as a central hub or aggregator for localized sensor networks, enriching its data collection and analysis capabilities.

Chipsets

ATMOSIC ATM33e MCU

The ATM33/ATM33e Series SoC has designed-in features that make it the best option for a low-power Bluetooth Low-Energy product.

ATM33e Series SoC has an on-chip RF Energy Harvester with a dedicated antenna port as well as a separate input for energy from photovoltaic, mechanical, or TEG harvesting devices.

The independent RF Wakeup Receiver is designed to look for an incoming paging or wakeup signal while the rest of the SoC remains in a very low power state.

The separate receiver supports short-range reception of a configurable signal from a Bluetooth device, mobile phone, or a dedicated transmitter.

The Power Management Unit is very efficient at providing the core and I/O power for the SoC but can also be bypassed if a power source is available elsewhere in the system.

The ATM33/ATM33e Series SoC also provides an integrated Sensor Hub which is a configurable hardware element that can read data from external sensors and write to RAM or an external flash device on the quad SPI interface, while all other power domains are powered down.

The sensor hub can also trigger a wakeup of the MCU if the data read falls outside programmed thresholds.

Bluetooth LE

■ Bluetooth Low-Energy 5.3 compliant ■ 2 Mbps, 1 Mbps 500 kbps, and 125 kbps PHY rates ■ Supports Bluetooth Angle-of-Arrival (AoA) and Angle-of-Departure (AoD) direction finding

MCU

■ 64 MHz ARM® Cortex® M33F MCU

■ 64 KB ROM, 128 KB RAM, 512 KB NVM, 1 MB extended flash storage (selected package)

■ Retention RAM configuration: 16 KB to 128 KB in 16 KB step sizes

■ 16 MHz / Optional 32.768 kHz Crystal Oscillator

Security

■ ARM® TrustZone®, HW Root of Trust, Secure Boot, Secure Execution & Debug

■ AES-128/256, SHA-2/HMAC 256 Encryption/Cryptographic Hardware Accelerators

■ True random number generator (TRNG)

Energy Harvesting

■ On-chip RF Energy Harvesting

■ Supports photovoltaic, TEG, motion, and other energy-harvesting technologies

■ External Harvesting/Storage Interface

RF and Power Management

■ Fully integrated RF front-end

■ Sensor Hub

■ RF Wakeup Receiver

■ 1.1 V to 4.2 V battery input voltage with integrated Power Management Unit (PMU)

■ Radio power consumption with 3 V battery - Rx @ -95 dBm: 0.85 mA - Tx @ 0 dBm: 2.5 mA

■ SoC typical power consumption with 3 V battery including PMU

- Retention @ 32 KB RAM: 1.8 µA

- Hibernate: 1.3 µA

- SoC Off: 400 nA

- SoC Off with Harvesting Enabled: 700 nA

Interfaces

2C (2), I2S, SPI (2), UART (2), PWM (8), GPIOs (15, 18, 21 or 31 depending on the package option)

■ Quad SPI

■ 11-bit application ADC, 4 external, 5 internal channels, up to 2 Msps

■ Two mono or one stereo digital microphone input (PDM)

■ 8 x 20 Keyboard Matrix Controller (KSM)

■ Quadrature Decoder (QDEC)

■ SWD for interactive debugging

ALTAIR ALT1250 Modem

■ 3GPP Release 14 (3GPP releases 15-17 to be supported by SW in the future)

■ CAT-M1: R14 Up to 588 Kbps in DL, and 1119 Kbps in uplink

■ CAT-NB2: R14 Up to 127 Kbps in DL, and 158 Kbps in uplink

■ Ultra-low-power with 1uA sleep current (PSM mode for LTE-M, NB-IoT)

■ HFDD (Half Duplex FDD) and TDD

■ Frequency Band Support • OneSKU™ frequency range: - Low Band: 617–960 MHz - Middle Band: 1700–2200 MHz

■ Supported Band List: - LTE-M: 1, 2, 3, 4, 5, 8, 12, 13, 14, 18, 19, 20, 25, 26, 27, 28, 66, 71 - NB-IoT: 1, 2, 3, 4, 5, 8, 12, 13, 14, 17, 18,19, 20, 25, 26, 28, 65, 66, 70, 71, 85 • 410-466MHz (bands 31, 72, 73, 87, 88) can be supported using additional RF FE components.

■ Application network protocols: supporting IPv4/IPv6 with TCP/UDP, PPP, FTP, HTTTP, TLS, HTTPS, SSL, DTSL, MQTT, CoAP, LWM2M

■ Operating temperature: • Maximum temperature: 85c • Minimum temperature: -4

■ Integrated PMU with Voltage Regulators and Real-Time clock (RTC)

■ ARM cortex-M4 integrated MCU

■ Low power always-on sensing hub-based ARM Cortex-M0+ integrated MCU

■ GNSS, Wi-Fi and cellular-based location engine for location services

■ Sub 1GHz and 2.4GHz for short range communication and GW

■ Carrier grade iSIM-based iSE

■ Enhanced application layer security-based iSE

■ AI low power acceleration for edge processing

■ Interfaces: UART, SPI master, and slave, I2C master and slave, PCM/I2S audio, PWM, LED, GPIOs, USIM or eIUCC, auxiliary ADC, capture/compare timer, anti-tampering, QUAD SPI flash, and PSRAM extension ports

Epishine Photovoltaic Energy Harvesting

Light Energy Harvesting: Epishine photovoltaic cells are designed to efficiently capture and convert light energy from both indoor and outdoor sources into electrical power.

Flexible and Thin: These cells are highly flexible and ultra-thin, making them suitable for integration into a wide range of applications and surfaces, including curved and irregular shapes.

High Efficiency: They offer high conversion efficiency, maximizing the amount of electrical power generated from available light.

Low Light Performance: Epishine cells are optimized for low-light conditions, ensuring reliable energy harvesting even in environments with limited natural light.

Sustainable and Eco-Friendly: Manufactured using environmentally friendly processes and materials, these cells contribute to sustainable energy solutions.

Wide Application Range: Ideal for use in IoT devices, sensors, wearables, and other low-power electronic applications where continuous energy supply is critical.

Ease of Integration: Designed for seamless integration into various products, reducing the need for external power sources and maintenance.

Long Lifespan: Durable and long-lasting, providing consistent performance over extended periods, reducing the need for frequent replacements.

Syntiant NDP120 Neural Decision Co-Processor

The Syntiant® NDP120 Neural Decision Processor™ is a special purpose chip for sensor processing in always-on applications in battery-powered edge devices.

It is capable of running machine learning models on these edge devices, allowing for efficient local data processing

In the context of the pharmaceutical cold chain, it can monitor and analyze temporal data from various sensors, ensuring the safe and efficient transportation of temperature-sensitive pharmaceutical products.

For fresh produce supply chains, it can provide real-time monitoring and analysis of data from various sensors across different locations.

This ability to process temporal-spatial data can help maintain the integrity and safety of both pharmaceutical products and fresh produce during transportation.

Overall, the NDP120 can be instrumental in enhancing the efficiency and reliability of both pharmaceutical and fresh produce supply chains by leveraging machine learning at the edge.