Enclustra FPGA Solutions | Digital Signal Processing | Digital Signal Processing | Design Services

Digital Signal Processing

Enclustra offers a unique combination of signal processing and FPGA/SoC technology expertise. As a result, algorithms are implemented optimally by design which reduces resource and power consumption and hence finally overall cost. Where required, Enclustra has the know-how and experience to hand-code your custom algorithms in RTL or HLS. On the other hand, proprietary and open-source libraries are used for common operations to achieve low development cost.

We offer turn-key projects from concept to implementation but are also open to working hand-in-hand with customers throughout the project to achieve optimal flexibility. Enclustra can work on concepts, implementation of existing algorithms or both.

Our goal is your success. Nobody knows your application better than you do and our mission is to support you in getting this know-how mapped to lines of code, LUTs and DSP slices in the optimal way.

Technology Expertise

Enclustra is focused on FPGA/SoC technology for over 15 years now. We know the available devices and their features by heart and keep technology constraints in mind when designing algorithms. As a result, any algorithm design done by Enclustra is guaranteed to nicely fit into real-world FPGA/SoC devices.

Achieving leading edge performance is Enclustra’s daily work and Enclustra’s enginerring team has Hundreds of person-years of cumulative experience in timing-closure, floorplanning and low-level-optimization, including newest-generation devices. For projects with lower performance requirements, the same optimization know-how is used to optimize for different key requirements, such as inexpensive target devices or low-power operation.Quality is key to get projects done in time. Therefore, Enclustra’s fine-tuned development processes include regular peer-reviews and fully automated regression tests, which are developed and delivered along with the production code.

DSP Expertise

Digital signal processing (DSP) is a common application area for FPGAs and SoCs. To optimally serve customers in this field, Enclustra built up extensive DSP know-how and thus cannot only support you with pure implementation services, but also with architecture and algorithm design from scratch and even with requirements engineering in the starting phase and consulting throughout a project. You have an idea of what you want to achieve but no architecture or exact specification yet? No worries. Enclustra can define the most optimal architecture in close collaboration with you. Different implementation options (filter architectures, number-formats, etc.) and trade-offs are discussed with you, so all decisions are well-founded and not to be regretted later in the project. All commonly used development tools and Programming languages are well-known at Enclustra. Whether your preferred language is Python, MATLAB, Simulink, SciLab, C/C++ or .NET/C#: We are fluent.

Applications

  • Wireless and wired communications
  • Test and measurement equipment
  • Audio processing
  • Control systems

Why choose Enclustra?

  • 100% project success record
  • Low project risk due to bit-true modelling flow
  • Deep DSP know-how from theory to implementation
  • We develop – you own!
    • Delivery of all sources and IP rights
    • Know-how transfer at the end of the project
  • Flexibility and close collaboration with customers

Development Flow

The standard Enclustra DSP development flow includes creation of Python or MATLAB bit-true models before starting FPGA/SoC implementation. Self-checking unit-tests are used to verify models and implementation are matching. Bit-true models enable you to evaluate the exact processing behavior and performance early in the project, even before FPGA/SoC implementation has started. As a result, costly iterations on the actual FPGA/SoC implementation are avoided and project risk is significantly reduced. Alternatively, Enclustra can also follow any customer-defined development flow, if required.