2072/5000 Alif Semiconductor introduces advanced BLE and Matter wireless microcontrollers

Maggie April 24, 2024

Alif Semiconductor, founded in 2019, reportedly introduced advanced BLE and Matter wireless microcontrollers; The Balletto™ family of wireless MCUS is based on Alif Semiconductor's advanced MCU architecture with DSP acceleration and dedicated NPU for fast and low-power execution of AI/ML workloads.

Alif Semiconductor®, a supplier of advanced secure, connected, and energy-efficient artificial intelligence and machine learning (AI/ML) microcontrollers (MCUS) and converged processors, today announced the Balletto™ family. This family is an advanced Bluetooth ® Low Power (BLE) wireless microcontroller with hardware optimization for AI/ML workloads.

The Balletto MCU is packaged in an ultra-small WLCSP package, giving manufacturers the opportunity to incorporate sophisticated AI/machine learning capabilities such as speech recognition, adaptive noise cancellation, sound localization and beamforming in true wireless stereo (TWS) earbuds, as well as sensors incorporated into fashion wristbands and other types of space-constrained devices.

With outstanding DSP and AI/ML capabilities and large on-chip integrated memory, the Balletto series offers particularly strong performance in audio and sensor functions. The Arm® Cortex-M55 core achieved an EEMBC CoreMark® score of 704 in the 160 MHZ band, including the Arm Helium M-profile Vector Extension (MVE) that improves DSP performance by 500%.

In addition to the Cortex-M55 CPU, Balletto includes an Arm EthO ™-U55 neural processing unit (NPU) with execution speeds of up to 46 billion times per second and is supported by 2MB of high-capacity tightly coupled memory (TCM) to scale audio capabilities with hardware acceleration of AI/ML models. The processing performance of the neural network is 15 times that of the Cortex-M4 processor. This results in excellent audio encoding/decoding capabilities, including the high-bitrate, low-latency LC3 codec that powers BLE Auracast™ broadcast audio technology.

The chip has the characteristics of ultra-low power consumption, which greatly improves the overall performance

Because Balletto uses Alif Semiconductor's aiPM™ technology, the technology can dynamically power only the logical and associated memory that is being used at a given point in time, achieving the lowest overall system power consumption. The aiPM power management unit also has four system-level power modes, including a stop mode that consumes only 700 nanoamps.

Thanks to its own independent processor and memory, Balletto's radios can support concurrent operation of Bluetooth LE 5.3 and Thread via a single on-chip antenna, making them ideal for use in Mature-enabled home automation networks and systems. In addition, the excellent reception sensitivity of -101dBm and advanced algorithms help avoid interference from other devices, and can coexist with other protocols that share the 2.4 gigahertz band.

Balletto also features a dual PA architecture: The High Power amplifier (HPA) delivers +10dBm output for maximum coverage and signal strength; With a low power amplifier (LPA), the output is +4dBm, which optimizes power consumption.

Balletto includes the most advanced multi-layer Security Fabric security architecture, which was first introduced in Alif Semiconductor's Ensemble family of MCUS and fusion processors. It is managed by an advanced security enclave with its own dedicated processor and memory, providing a hardware root of trust and chain of custody.

Other key features of Balletto include:

Digital interfaces, including I3C, USB-HS (480Mbps) and dual CAN-FD channels

Analog front end with high precision ADC, including 24-bit Sigma-Delta, DAC

Up to 2MB of high-speed MRAM and 2MB of zero-wait-state SRAM

Eight-way SPI interface

Up to 77 GPIOs, including ports for audio input/output

Camera interface

MIPI Indicates the interface

Two-dimensional graphics processor

Sensor hub function

Paul Williamson, senior vice president and general manager of Arm's iot business, said: "Device manufacturers and developers will need more device-based intelligence if they are to deliver on the promise of edge AI across consumer, enterprise and industrial markets. Alif's new family of SoCs, powered by Arm, significantly improves the performance of machine learning applications, making small, cost-sensitive devices smarter and more powerful."

Reza Kazerounian, president and co-founder of Alif Semiconductor, said: With a platform adapted to the high bitrate audio and Matter ecosystem, Alif Semiconductor's Balletto series targets the BLE challenges facing today's wireless audio and smart home industries. With the introduction of multi-protocol support, coupled with our industry-leading AI/ ML-enabled MCUS, it will be easier than ever to build connected smart edge devices that benefit battery life."

Founded in 2019, Alif Semiconductor's vision is to meet the needs of a rapidly growing market by providing broad, scalable, and connected AI-powered computing solutions that are truly energy efficient. Driven by this need, Alif Semiconductor creates new embedded controllers or converged processors that unlock innovative, low-power technologies, unmatched functional integration, accelerated AI and machine learning edge processing, high security, ubiquitous wireless connectivity, and operating system diversity. Seamlessly integrate technology into everyday life.