Ceva-XC21 Crowned “Best IP/Processor of the Year”

In a resounding affirmation of innovation in semiconductor intellectual property (IP), Ceva, Inc. (NASDAQ: CEVA) has been honored with the prestigious “Best IP/Processor of the Year” award at the 2025 EE Awards Asia, held in Taipei on December 4. The accolade went to the Ceva-XC21, a groundbreaking vector digital signal processor (DSP) core that redefines efficiency in 5G and 5G-Advanced communications. This victory underscores Ceva’s unwavering commitment to delivering high-performance, low-power solutions that propel the next era of connected devices, from cellular IoT modems to non-terrestrial network VSAT terminals.

The EE Awards Asia, organized by EE Times Asia and now in its 12th year, stands as Asia’s premier recognition for excellence in electronics engineering. Attracting nominations from global industry leaders, the awards celebrate breakthroughs in categories spanning IP cores, power management, AI accelerators, and more. This year’s event, coinciding with the EE Tech Summit, drew over 500 engineers, executives, and innovators, highlighting Asia’s pivotal role in shaping global semiconductor trends. Ceva’s win in the IP/Processor category—amid stiff competition from giants like Arm and Synopsys, signals the XC21’s transformative potential in an era where power efficiency and scalability are non-negotiable for 5G deployment.

At the heart of the Ceva-XC21 is its advanced architecture, building on the proven Ceva-XC20 foundation while introducing true dual-threaded hardware for contention-free multithreading. This design features dual processing elements, instruction, and data memory subsystems, enabling seamless parallel execution of complex workloads. The processor supports a versatile 9-issue Very Long Instruction Word set, accommodating integer formats (INT8/16/32) alongside half-precision, single-precision, and double-precision floating-point operations. A dedicated instruction set architecture (ISA) accelerates 5G New Radio (NR) functions, making it ideal for enhanced Mobile Broadband, ultra-Reliable Low-Latency Communications, and massive Machine-Type Communications.

What sets the XC21 apart is its scalability and efficiency. Available in three variants (Ceva-XC210, XC211, and XC212) each offers configurable single- or dual-thread options with 32 or 64 16-bit x 16-bit multiply-accumulate (MAC) units. This modularity allows designers to tailor the core to specific needs, from compact RedCap devices to high-throughput industrial terminals. Compared to its predecessor, the widely adopted Ceva-XC4500, the XC21 delivers up to 1.8x performance uplift in the XC212 variant while slashing core area by 48% in the XC210 model. The XC211 maintains equivalent performance at just 63% of the previous die size, achieving a CoreMark/MHz score of 5.14 for superior control code execution.

These metrics translate to tangible benefits: unprecedented power savings for battery-constrained IoT endpoints, reduced bill-of-materials costs for consumer gadgets, and enhanced AI/ML integration for smarter edge processing. Interconnectivity is equally robust, with up to six AXI4 bus interfaces via an AMBA matrix for high-bandwidth data flows, ensuring effortless SoC integration. Software support further eases adoption, including a unified programming model compatible with the Ceva-XC4500 ecosystem, an optimizing LLVM C compiler, and comprehensive debug tools like JTAG and real-time trace.

“We are thrilled and deeply honored by this recognition from the EE Awards Asia jury,” said Amir Panush, CEO of Ceva. “The Ceva-XC21 embodies our vision of democratizing 5G Advanced connectivity making it accessible, efficient, and future-proof. In a market projected to see 5G connections surpass 2 billion by 2026, this DSP empowers our licensees to innovate without compromise, from smart wearables to satellite backhaul systems.”

This isn’t Ceva’s first triumph at EE Awards Asia; the company previously clinched the same category in 2023 for the Ceva-XC22 and in 2024 for the NeuPro-Nano NPU, cementing its legacy as a trailblazer in edge IP. The XC21’s success reflects broader industry shifts: as 6G horizons emerge, demand for versatile, energy-efficient processors intensifies. Analysts at Gartner forecast that by 2028, 75% of enterprise-generated data will be processed at the edge, necessitating IP like the XC21 to handle multi-protocol stacks (LTE/5G/NTN) with minimal overhead.

Looking ahead, Ceva’s roadmap hints at even bolder integrations, blending XC21’s vector prowess with AI accelerators for hybrid edge-cloud paradigms. For developers, the implications are profound: shorter time-to-market, lower power envelopes, and scalable designs that future-proof against evolving standards. As Gideon Wertheizer, Executive VP of Research and Development, noted, “Winning ‘Best IP/Processor’ validates our relentless focus on architecting for tomorrow’s challenges today.”

In an industry often dominated by raw compute power, the Ceva-XC21 reminds us that true excellence lies in balance, performance without excess, innovation without waste. This award not only elevates Ceva’s profile but also accelerates the proliferation of intelligent, connected ecosystems across Asia and beyond. As 5G matures into a ubiquitous fabric, the XC21 stands as a cornerstone, weaving efficiency into the wireless future.

Contact Ceva Here.

Also Read:

United Micro Technology and Ceva Collaborate for 5G RedCap SoC and Why it Matters

Ceva Unleashes Wi-Fi 7 Pulse: Awakening Instant AI Brains in IoT and Physical Robots

A Remote Touchscreen-like Control Experience for TVs and More