Ceva has introduced a multi-radio intellectual property package for IoT IC designers that combines Wi-Fi 6, Bluetooth 5.4 and 802.15.4 modems with a 2.4GHz transceiver aimed at TSMC’s 22nm process. Called Links100 (diagram right), it is the first of a series of IP products (diagram below) that will also include transceivers for Wi-Fi 7 with MLO (multi-link operation), and UWB ...
Tag Archives: CEVA
Ceva adds cores and generative AI to neural processing IP portfolio
Ceva has filled-out its NeuPro-M family of neural processing units, and boosted its architecture for generative AI algorithms. “NeuPro-M NPU architecture and tools have been redesigned to support transformer networks in addition to convolutional neural networks,” according to the company. “This enables applications leveraging the capabilities of generative and classic AI to be developed and run on the NeuPro-M NPU.” ...
Arm down in an up market
Arm’s revenues declined slightly in 2019 in a market that grew, says IPnest, reflecting similar situations in 2018 and 2017. Arm’s royalty revenue declined by 6-7%, says IPnest, but its licensing revenues – including physical IP – were up 13%. Companies ranked by semiconductor design IP revenue in 2019 (millions of dollars). Source: IPnest (March 2020). ARM once controlled 50% ...
Ceva DSP and voice AI support Tensorflow Lite integration
Ceva’s BX DSP cores and Whispro speech recognition software for conversational AI now also support Tensorflow Lite for microcontrollers, a framework for deploying tiny machine learning on processors in edge devices. Tiny machine learning allows AI to operate in low power, battery operated IoT devices for sensor data analytics of audio, voice, image and motion. Ceva says its approach to ...
Bragi and Ceva collaborate on software-customisable hearable devices
Ceva and Bragi have announced a collaboration to drive hearable devices where the user experience is customizable. Target devices include headsets, hearing aids, true wireless stereo earbuds and other smart audio devices. The partnership utilises Ceva’s BX DSP low-power architecture, with its LLVM compiler and Eclipse-based development tool chain, and Bragi 4, an OS specialized for hearables. In conjunction with ...
CEVA opens Bristol R&D centre; seeks engineers
CEVA has opened an R&D facility in Bristol to access the engineering talent the city has to offer, to strengthen its R&D capabilities and to expedite the development of new digital signal processing and AI products. ‘Bristol has one of the strongest talent bases in the U.K. for digital signal processing, with many highly-skilled and experienced engineers,” says CEVA CEO Gideonn Wertheizer, “this ...
CEVA improves cellular IoT core
CEVA announces its CEVA-Dragonfly NB2 core for the cellular IoT standardCat-NB2
Mobile World Congress: Ceva adds power DSP to Nokia 5G network
Ceva is working with Nokia in the development of baseband system-on-chips (SoCs) for 4.9G and 5G wireless infrastructure. Ceva has adapted its CEVA-XC architecture framework to address the massive increase in signal processing complexity in multi-RAT (Radio Access Technology) network architectures. Nokia’s ReefShark SoC is based on 3GPP 5G New Radio specifications, which help offset deployment costs and TCO, while ...
CEVA’s PentaG for 5G baseband processing
CEVA has introduced PentaG a 5G IP platform for enhanced mobile broadband (eMBB). Compliant to the 3GPP 5G New Radio (NR) Release-15 and software upgradable for Release-16, PentaG is targeted at smartphones, fixed wireless access and the many other embedded devices that can take advantage of multi-gigabit data rates to enable services and use cases such as 4K video streaming, ...
CEVA algorithms clarify voice input
CEVA has started licensing ClearVox – voice input processing algorithms aimed at enhancing speech intelligibility and voice clarity for the CEVA-TeakLite-4 and CEVA-X2 audio/voice DSPs. Voice-enabled speakers have to cope with multi-mic beamforming, multi-channel acoustic echo cancellation and noise suppression in their products. The ClearVox algorithms cope with different acoustic scenarios and microphone configurations, including optimized software for speaker direction of ...