Image compression and navigation on the agenda
LONDON — Imagination Technologies used the Linley Fall Conference in Silicon Valley to announce two new IP cores, one for low power consumption location information and another for visually lossless image compression.
Imagination’s new global navigation satellite system (GNSS) IP offering, the Ensigma Location GNSS IP core, supports GPS, GLONASS, Galileo, and BeiDou as well as several satellite-based augmentation systems (SBAS) including WAAS and EGNOS. The IP is optimized for battery-powered remote IoT sensors and edge devices, wearables, health monitors, consumer mobile products, automotive after-sales products such as insurance boxes and road tolling equipment, and asset tracking devices.
For multimedia applications, the company’s PowerVR PVRIC4 is a new generation of its image compression technology for SoC customers targeting devices such as DTVs, smartphones, and tablets to reduce costs without a discernible loss of image quality. It claims random-access visually lossless image compression, ensuring bandwidth and memory footprint savings of at least 50% and enabling systems to overcome performance bandwidth constraints.
Low power GNSS IP
A growing number of battery-operated products incorporate positioning, navigation, and timing (PNT) services. Consumers want to track their devices but don’t want to recharge batteries frequently. In industrial and agricultural environments, users need to track mobile assets to improve efficiency and reduce operational costs, but it isn’t possible to frequently change batteries across numerous devices in disparate locations. In addition, new regulatory requirements mandate the use of location services in some products, for example, to address spectrum sharing requirements in IoT devices and base stations.
“Our customers are increasingly combining connectivity such as Wi-Fi, Bluetooth, and 802.15.4 with GNSS technology in one chip,” said Martin Woodhead, executive vice president for Ensigma Communications at Imagination.
For this, Woodhead said that the Ensigma GNSS IP provides a power-efficient solution to integrate into customers’ own silicon to save on total solution cost.
Malik Saadi, managing director and vice president for strategic technologies at ABI Research, said that silicon and system vendors designing their own silicon need low-power, high-quality GNSS IP. Until now, they’ve had to either settle for a restricted set of features or use a companion chip to provide comprehensive functionality. He said that ABI Research has consistently maintained that the integration of fully featured GNSS with other wireless technologies has the ability to enhance the overall performance of the chip and lower the overall power consumption without compromising cost.
The Ensigma GNSS IP can share location information with external wireless technologies such as LTE or Wi-Fi and has options to share peripherals such as memory and system clock to enable reduced total system cost. It also features high sensitivity for indoor location, support for assistance information using external sources such as LTE and Wi-Fi to improve time to first fix, and radio frequency interference (RFI) detection and mitigation.
The GNSS IP enables developers to choose a configuration that works best for their specific system implementation, from standalone to highly integrated. The IP can be configured for lowest power or highest integration and is designed to fit into any existing solution.
The IP includes dedicated hardware blocks that enable lower power consumption compared to a software-only solution. In addition, it supports not only continuous fix techniques but power-efficient “capture and process” for devices that require only periodic location updates. This feature further conserves battery life by capturing data, such as fitness information from a wearable device, and storing it for later processing.
Ensigma GNSS IP is designed to be used with a wide range of GNSS receivers; the initial solution is optimized for use with SaberTek’s ultra-low-power GNSS receiver. The broad portfolio of Ensigma wireless communications IP comprises solutions for 30 wireless standards in Wi-Fi, Bluetooth, 802.15.4, audio, and TV broadcast.
The PowerVR PVRIC4 is a standalone IP block for SoC manufacturers to integrate into their multimedia pipelines. The block is already used by Imagination partners, including Chips&Media, who has access to PVRIC4 as part of a recent collaboration to deliver an optimum solution for a system leveraging a PowerVR GPU and WAVE5 video codec.
Linley Gwennap, principal analyst at the Linley Group, called it a significant advancement for GPU compression technology. “The increasing demands on memory bandwidth and footprint are a real concern for SoC manufacturers, so the ability to combine lossless and visually lossless compression delivers both cost and bandwidth savings.”
PVRIC4 features a dual-pipeline framebuffer compression engine. A new lossy pipeline, used only if the lossless pipeline does not achieve 50% compression, ensures that even difficult-to-compress, “noisy” images are compressed with the highest fidelity. A decision logic block determines which output should be used to guarantee the compression ratio, and due to highly tuned algorithms, the image quality change is imperceptible. This hybrid solution means that PVRIC4 offers SoC manufacturers the best of both approaches, with high fidelity ensuring bandwidth and frame buffer allocation savings on graphics and video content. This is all performed in hardware and achieved without performance overhead.
Images rendered with the PowerVR PVRIC4 IP, one compressed and the other not, illustrating that image quality change is imperceptible. (Source: Imagination Technologies)
The PVRIC4’s bandwidth savings translate into better battery life for consumers and cost savings for system manufacturers, enabling more RAM and bandwidth to be freed for other uses, such as enabling simultaneous fast 5G downloads while the GPU is in use or a reduction in the number of DRAM devices used in the system.
Nigel Leeder, executive vice president of PowerVR at Imagination, said that increasing demands introduced by higher-resolution 4K and 8K displays, combined with the desire to reduce system costs, results in a need to minimize memory bandwidth for its customers. By introducing visually lossless compression into its GPUs, a reduction in memory bandwidth and usage is possible. PVRIC4 will be available as a feature in next-generation PowerVR GPUs and is available for licensing now as a standalone IP block.