i.MX applications processor has NPU for advanced ML at the edge

The latest member of NXP’s EdgeVerse portfolio has been launched at CES in Las Vegas (7 to 10 January). The i.MX 8M Plus heterogeneous application processor is the first i.MX family to integrate a dedicated neural processing unit (NPU) for advanced machine learning (ML) inference at the industrial and IoT (Internet-of-Things) edge.

The NPU delivers 2.3Toperations per second and is combined with a quad-core Arm Cortex-A53 sub-system running at up to 2GHz. There is also an independent real-time sub-system with an 800MHz Cortex-M7, an 800MHz audio DSP for voice and natural language processing, a dual camera image signal processors (ISP) and a 3D GPU for rich graphics rendering. The i.MX 8M will enable edge devices to make intelligent decisions locally by learning and inferring inputs with little or no human intervention, says NXP. Target applications are people and object recognition for public safety, industrial machine vision, robotics, hand gesture, and emotion detection with natural language processing for seamless human-to-device interaction with fast response time and high accuracy.
The applications processor is based on 14nm LPC FinFET process technology. The i.MX 8M Plus can execute multiple, highly-complex neural networks simultaneously, these include multi-object identification, speech recognition of 40,000+ English words and medical imaging. The NPU is capable of processing Mobilenet, a popular image classification network at over 500 images per second, adds NXP.
Developers can off-load machine learning inference functions to the NPU, allowing the Cortex-A and Cortex-M cores, DSP and GPUs to execute other system-level or user applications tasks. The vision pipeline is anchored by dual integrated ISPs that support two high-definition cameras for real-time stereo vision or a single 12Mpixel resolution camera and includes high dynamic range (HDR) and fisheye lens correction for real-time image processing in surveillance, smart retail applications, robot vision and home health monitors.
For voice applications, the i.MX 8M Plus integrates a HiFi 4 DSP that enhances natural language processing with pre- and post-processing of voice streams. The Cortex-M7 domain can be used to run real-time response systems while the applications processor domain executes complex non-real-time applications. It reduces overall system-level power consumption by turning off the application processor domain while keeping only the Cortex-M domain alive for wake word detection. For advanced multimedia, and video processing, the processor can compress multiple video feeds using the H.265 or H.264 HD video encoder and decoder for cloud streaming or local storage, 3D/2D graphics, and Immersiv3D audio with Dolby Atmos and DTS:X.

In industrial scenarios, it can be used in machines that inspect, measure, precisely identify objects and enable predictive maintenance by detecting anomalies in machine operation. It can also support making the factory human machine interfaces (HMIs) more intuitive and secure by combining accurate face recognition with voice/command recognition and gesture recognition. The i.MX 8M Plus integrates Gigabit Ethernet with time sensitive networking (TSN), which combined with Arm Cortex M7 real-time processing provides deterministic wired network connectivity and processing, NXP explains.
Other features for industrial use are error correction code (ECC) for internal memories and the DDR interface.
The family is expected to be qualified to meet the stringent industrial temperature range (-40 to +105 degrees C ambient).

NXP at CES 2020: booth, CP-18
http://www.nxp.com  

> Read More

Jacinto 7 processors have functional safety features to advance ADAS 

Deep learning capabilities and advanced networking characterise the Jacinto 7 processor platform that Texas Instruments launched at CES. The first two automotive devices in the platform, TDA4VM processors for advanced driver assistance systems (ADAS) and DRA829V processors for gateway systems, include specialised on-chip accelerators to segment and expedite data-intensive tasks, like computer vision and deep learning.

The TDA4VM and DRA829V processors also incorporate a functional safety microcontroller, making it possible for original equipment manufacturers (OEMs) and Tier-1 suppliers to support both ASIL-D safety-critical tasks and convenience features with one chip, points out Texas Instruments. They share a single software platform, to reduce system complexity and cost as developers can reuse their software investment across multiple vehicle domains.

The TDA4VM processor offers on-chip analytics combined with sensor pre-processing, enabling OEMs and Tier-1 suppliers to support front camera applications using high-resolution 8Mpixel cameras to see farther and add enhanced features, such as drive assist. They are also capable of simultaneously operating four to six 3Mpixel cameras while also fusing other sensing modalities such as radar, lidar and ultrasonic on one chip, adds Texas Instruments. This multi-level capability enables TDA4VM to act as the centralised processor for ADAS. It also enables the critical features for automated parking, like surround view and image processing for displays and enhanced vehicle perception.

The DRA829V processor integrates the computing functions required for modern vehicles. It is believed to be the first in the industry to incorporate a PCIe switch on-chip in addition to an eight-port gigabit time sensitive network (TSN) -enabled Ethernet switch for faster computing functions and communications throughout the vehicle.

It also supports ASIL-D safety-critical and non-safety-critical operations and enables  OEMs and Tier-1 suppliers to support mixed-criticality applications on one device. Higher bandwidth on-chip also allows developers to manage software development and validation in vehicles, for more dynamic updates and upgrades.

Pre-production TDA4VM and DRA8329V processors are available now, together with Jacinto 7 processors development kits and TDA4VMXEVM and DRA829VXEVM evaluation modules.

Volume production is expected to be available in the second half of 2020.

http://www.TI.com

> Read More

Nordic Semiconductor prepares for Bluetooth LE Audio

This year, the Bluetooth Special Interest Group (SIG) will release the Bluetooth LE (Low Energy) Audio specification. To support this forthcoming specification, Nordic Semiconductor has partnered with Bluetooth LE stack developer, Packetcraft, to develop an LE Audio evaluation platform.

It demonstrates the benefits of LE Audio and is designed to support the new LE Audio specifications which promise lower power consumption than classic Bluetooth audio, significantly extending battery life, improved audio quality and enables the development of devices capable of both wireless data transfer and audio streaming. The technology also supports broadcast for audio sharing.

Nordic’s LE Audio comprises a hardware reference design based on its nRF52832 Bluetooth LE system on chip (SoC), Cirrus Logic’s CS47L35 smart codec with an integrated low power audio DSP, Packetcraft’s Bluetooth LE host stack and link layer supporting LE Audio and an LE Audio software development kit (SDK). The platform allows developers to start evaluating the technology for Bluetooth LE Audio wireless speakers, over-the-ear headphones and true wireless ear buds.

It is designed to operate in either source (for example, audio source, voice call headset source, and peer-to-peer call) or sink (audio playback, headset playback, and peer-to-peer call). A pair of devices (source and sink) is required to complete an LE Audio link. The LE Audio solution also features Cirrus Logic’s SoundClear for uplink noise reduction and echo cancellation, playback enhancement, voice control, and hearing augmentation.

An acoustic connector is incorporate that can accommodate up to six microphones or two speakers, a 3.5mm headset jack, a 3.5mm source jack, and a USB connector for charging, debug (using Nordic development tools), and acoustic tuning (using the Cirrus Logic WISCE platform). When wirelessly streaming audio from a source device to wireless (sink) earbuds, the LE Audio evaluation platform extends battery life by around 40 per cent compared to contemporary off-the-shelf classic Bluetooth solutions, claims Nordic Semiconductor.

http://www.nordicsemi.com

> Read More

“World’s smallest” 3D image sensor authenticates faces

At CES this week, Infineon will present what it claims is the world’s smallest 3D image sensor for face authentication and photo effects.

Infineon Technologies has collaborated with software and 3D time of flight (ToF) system specialist, pmdtechnologies, to develop what it claims is the world’s smallest and most powerful 3D image sensor. The Real3 chip measures 4.4 x 5.1mm and is the fifth generation of ToF deep sensors from Infineon.

Andreas Urschitz, president of the power management and multi-market division at Infineon, said: “We see great growth potential for 3D sensors, since the range of applications in the areas of security, image use and context-based interaction with the devices will steadily increase.” The 3D sensor also allows the device to be controlled via gestures, so that human-machine interaction is context-based and touch-free.

The depth sensor ToF technology enables an accurate 3D image of faces, hand details or objects, required to ensure that an image matches the original to verify payment transactions using mobile phones via facial recognition. This function requires an extremely reliable and secure image and return transmission of the high-resolution 3D image data. The same applies to securely unlocking devices with a 3D image. The Infineon 3D image sensor also implements this in extreme lighting conditions such as strong sunlight or in the dark.

The IRS2887C 3D image sensor also has additional options for photo taking, such as enhanced autofocus, bokeh effect for photo and video and improved resolution in poor lighting conditions. Real-time full-3D mapping also allows authentic augmented reality experiences.

Production will begin in the middle of 2020.

In addition, Infineon Technologies offers an optimised illumination driver (IRS9100C).

Infineon Technologies provides semiconductors to “make life easier, safer and greener”. It has approximately 41,400 employees worldwide.

http://www.infineon.com/real3

> Read More

About Smart Cities

This news story is brought to you by smartcitieselectronics.com, the specialist site dedicated to delivering information about what’s new in the Smart City Electronics industry, with daily news updates, new products and industry news. To stay up-to-date, register to receive our weekly newsletters and keep yourself informed on the latest technology news and new products from around the globe. Simply click this link to register here: Smart Cities Registration