Tensilica Fusion F1 DSP

The Fusion F1 DSP has the high configurability to scale the breadth of potential IoT and wearable applications, and its architecture provides very good DSP performance, both fixed and floating point, to process all of the sensor data available in these new devices. Developed on a highly configurable architecture, the Fusion F1 DSP is specifically designed to excel at “always-on” processing, including wake-on-voice and sensor Fusion F1 applications. Additionally, the Fusion F1 DSP is very efficient in running the narrowband wireless communications standards typically associated with IoT device communications.

Designed for low-energy, high-performance IoT/wearable applications, the Fusion F1 DSP is the industry leader for voice-trigger/ wake-on-voice applications. The Fusion F1 DSP also excels at sensor Fusion F1 processing. Its single-precision floating-point unit (FPU) efficiently runs the software suites of the leading sensor software providers.

No one processor will be able to serve the tremendous breadth of applications in the IoT market. That’s why it’s essential to give you the opportunity to choose a processor that only includes functionality that is needed for your particular application to keep the costs and energy consumption down. With the Fusion F1 DSP, you can quickly configure the processor with the options that are required. This approach allows performance, energy, and area optimization with very little additional engineering effort. In addition, this configurability allows an SoC provider to scale the same processor architecture, development tools, and hardware flow across the entire IoT/wearables portfolio.

OEMs are continuing to look to audio/voice/speech (AVS) as a means to differentiate their products and provide a compelling user experience. For instance, they’re adding features such as voice trigger and advanced voice recognition to their products. For the home audio market, this could mean supporting the latest object-based audio decoder or the latest 3-D audio postprocessing suite.

By enabling the AVS configuration option, you can gain software compatibility with the HiFi DSP family. The HiFi software ecosystem has the broadest set of codecs, pre and postprocessing and sensor Fusion F1, voice- and face-triggering software in the industry. Currently, there are more than 175 different software packages available in an ecosystem that includes more than 80 different companies that help OEMs create compelling products.

The Fusion F1 DSP provides a high level of flexibility to help you create the right processor for your end applications. Based upon the Tensilica Xtensa configurable processor, processors with the Fusion F1 DSP are fully configurable. Configuration options are pre-defined by Cadence and give you control over numerous features, including the memory subsystem, debug, and interrupts.

In addition to the pre-defined configuration options, you can also add custom instructions to further improve the performance and differentiation on any given application, with full development tools and flow support provided automatically. The Xtensa Processor Generator makes configuration and customization possible. All changes remain completely compatible with the existing software ecosystem.

A key advantage of the Fusion F1 DSP is its simple programming model. Because of the efficiency of the compiler, software developers can write their applications completely in C while maintaining or surpassing the performance of assembly on other DSP and CPU architectures.