Neuromorphic Computing, Memory and Sensing 2024

Scope of the report

Methodologies & definition

About the authors

What we got right, what we got wrong

3-Page summary

Executive summary

Context

• The pursuit of Moore’s law
• The brain as a model

Market forecasts

• Market Segmentation of neuromorphic technologies
• Forecast Methodology
• 2019-2034 Neuromorphic sensing and computing forecast

Market trends

• Mobile and Consumer
  • Smartphone
  • Entertainment
  • Consumer health
  • Home
  • Productivity
• Automotive and Mobility
  • ADAS
  • Infotainment
  • Chassis and Others
• Medical
  • Health Monitoring
• Industrial
  • Manufacturing
  • Logistics
  • Energy
  • Security
• Communication & Infrastructure
  Datacenter focus
• Defense & Aerospace

Ecosystem and supply chain

• Neuromorphic: noteworthy news of the Last Three years
• Neuromorphic companies market positioning
• Neuromorphic and in-memory computing
• Neuromorphic sensing
• Edge-AI Computing Players – In-memory computing and near-memory computing approaches

Technology trends

• Neuromorphic Sensing
• Neuromorphic Computing Processor
• An opportunity for FDSOI
• Memory technologies

Conclusion

Related Product

YG corporate presentation

Neuromorphic sensing and computing markets are expected to generate $8.4B by 2034

Neuromorphic sensing and computing are poised for significant growth. In 2029, the neuromorphic sensing market could reach $410M and $2.9B by 2034, while the neuromorphic computing market is expected to grow to $412M by 2029 and $5.4B by 2034. For neuromorphic sensing, mobile applications will lead in 2034, followed by entertainment, smart city, automotive, and home applications. For neuromorphic computing, however, data center applications are expected to be the biggest revenue generator in 2034, followed by automotive, entertainment, smart city, and mobile. The embedded emerging NVM market for near- and in-memory computing is expected to remain limited until at least 2029. Near-memory computing approaches – mainly implemented with MRAM – will lead the short-term growth, while analog IMC and other RRAM/PCM-based IMC approaches will ramp up in the following years (>2027).

Consolidating ecosystem with different strategies to harvest neuromorphic revenues

The neuromorphic ecosystem has seen increased activity, with notable investments and product demonstrations and announcements since last year (2023). Leading companies, such as IBM, Intel, Sony, and Samsung, are focusing on specific markets, while startups are pursuing more diversified approaches. Traditional processor companies are partnering with neuromorphic firms, enhancing technological differentiation. Some neuromorphic companies target extreme edge AI on battery-powered devices, while others focus on high-performance computing for data centers. In China, giant technology leaders such as Hikvision and Huawei are investing in neuromorphic technologies, mainly for security cameras. In memory technology, most companies are developing digital IC designs utilizing distributed SRAM, with a few adopting SRAM-based or embedded Flash memory cells. Emerging non-volatile memory technologies like PCM and OxRAM show promise but are still in the early stages of development.

Low latency, power efficiency, and online learning are the key neuromorphic assets

The economic feasibility of scaling semiconductor devices is declining despite Moore's Law holding true, as chip development costs rise sharply. Neuromorphic technologies, inspired by biological brains, offer power-efficient solutions for AI tasks, with benefits like low latency and high scalability. They enable real-time edge-AI applications, addressing privacy concerns while utilizing manufacturing nodes from 28nm to 12-7nm for the most advanced. Sensing options include standalone event-based sensors and mixed RGB with event-based pixel hybrids. Neuromorphic computing systems, featuring event-driven processing and spiking neural network algorithms, enable online learning and autonomous robotics. Digital IC designs based on SRAM distributed across logic cores or neurons dominate memory approaches, with analog in-memory computing emerging but facing technical challenges. Overall, neuromorphic technologies show promise for sustainable and efficient AI processing at the edge.

ABR, Accenture, Adesto Technologies, Aistartek, AI Storm, Alibaba, Alpsentek, Amazon, Ambarella, Ambient Scientific, AMD, AMT, Analog Inference, Anotherbrain, Antaïos, Apple, Applied Materials, ARM, Aryballe Technologies, Aspinity, Avalanche Technology, AWS, Axelera AI, Axis, Azure, BAE Systems, Baidu, Blumind, BMW, Bosch, BrainChip, Canon, CEA, Celepixel, Ceva, Cogito Instruments, Crossbar, d-Matrix, Dahua, Dialog, DTS, Empatica, ESA, Facebook, Fraunhofer, Fullhan, General Vision, GlobalFoundries, Google, Gorilla, GrAI Matter Lab, Groq, HPLabs, Gyrfalcon Technology, Hikvision, HLMC, Hprobe, Huawei, IBM, IMEC, Infineon, IniLabs, iniVation, Innatera, Insightness, Intel, Kalray, Knowm, Lenovo, Leonardo DRS, Lynxi, MediaTek,  Mentium Technologies, Meta, Microchip, Micron, Mythic, NASA, Natural Intelligence, Nepes, Neurxcore, Nimble AI, Northrop Grumman, Numem, Numenta, Nuvoton, Nvidia, NXP, Oculi, Omnivision, Oppo, Opteran, Polyn Technology, Prophesee, Qeexo, Qualcomm, Quantum Ventura, Rain AI, Raytheon, Renesas, Robosensing, Samsung, SCD, Schneider Electric, Sensigent, Sigmastar, Smartnvy, Smartsens, SMIC, Socionext, Soitec, Sony, Snap Inc, Space Machines Company, Spin Ion Technologies, SpinEdge, STMicroelectronics, Summer Robotics, SynSense, Synthara, Syntiant, TDK, TetraMem, Tiandy, TSMC, Ultraleap, UMC, Uniview, Untether AI, Visionchip, Vision Research, Voxel Sensors, Weebit, Witmem, Xiaomi, Xperi, and more…

Report's objectives

To provide market data on key neuromorphic adoption areas with details on:

• Semiconductor-level revenue forecasts, shipments, and ASPs, and penetration rate of neuromorphic sensing and computing for 6 markets, 13 applications, and 29 systems/end-systems.
• Market dynamics and segmentation breakdown by application, market, and end-system.
• Two axes of focus: sensing and computing.
• Memory architecture and related technology trends.

To deliver an in-depth understanding of the neuromorphic ecosystem:

• Who are the key CIS manufacturers, neuromorphic computing, near memory/in-memory AI accelerator manufacturers, and eNVM manufacturers, and how are they related?
• Who are the key suppliers to watch?
• How will the technology and associated market evolve?

To present key technical insights and analyses regarding future technology trends and challenges:

• Manufacturing technologies and structural design.
• Device technologies and applications across the different markets.
• Technology insights for game-changing applications like deep learning and neuromorphic approaches.

Key features

• 2019 - 2034 Neuromorphic sensing and computing market forecast in $. (by market, application, end-systems)
• 2019 - 2034 Neuromorphic technology diffusion in %.
• Consumer, automotive, medical, industrial, datacenter and defense/ aerospace market trends and neuromorphic opportunities
• Neuromorphic computing and in-memory computing competitors – mapping and technology segmentation and neuromorphic image sensor player mapping
• Neuromorphic sensing and computing technology trends
• Neuromorphic software description and readiness analysis
• Neuromorphic in-memory computing concept description
• Emerging Non-Volatile Memory, ecosystem and roadmap
• 2019-2029 emerging Non-Volatile Memory forecast

What's new?

New systems tracked. New modeling of automotive forecast. Separate scenario for sensing and computing technologies. Deeper sensing ecosystem and technologies analysis.