Spiking Neural Network Hardware Patent Landscape 2026
Spiking Neural Network Hardware Patent Landscape in 2026
Spiking neural network hardware IP is tightly concentrated: a single specialist, Innatera Nanosystems, holds the largest position among 69 patent families in scope, while the field is still expanding on a multi-year basis even as annual volume has eased from its 2022 peak. The United States leads as the primary filing jurisdiction, with AI-model computing as the overwhelmingly dominant technology branch.
Innatera leads a highly concentrated, specialist-dominated field
Innatera Nanosystems BV holds the top position in this landscape, followed by Intel Corporation at second rank and Fraunhofer Society at third. The top five filers account for 65% of the combined output of the hundred largest filers, signalling an unusually tight competitive structure for an emerging hardware category.
The tier gap between the leader and the rest is sharp: Innatera’s count is three times that of Intel and more than four times that of Fraunhofer. Below those three, most applicants hold only two or fewer patent families, confirming a long, thin tail of academic and research entrants rather than a broad commercial cohort.
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 1 | Innatera Nanosystems BV | 27 | |
| 2 | Intel Corporation | 9 | |
| 3 | Fraunhofer Society | 6 | |
| 4 | University of Dayton | 3 | |
| 5 | Institute of Semiconductors, Chinese Academy of Sciences | 2 | |
| 6 | The University of Aizu | 2 | |
| 7 | Korea Electronics Technology Institute | 2 | |
| 8 | UNIV OF ELECTRONICS SCI & TECH OF CHINA | 2 | |
| 9 | Electric Power Research Institute of State Grid Zhejiang Electric Power Co., Ltd. | 2 | |
| 10 | University of Windsor | 2 |
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 11 | Peking University | 2 | |
| 12 | Huawei Technologies Co., Ltd. | 2 | |
| 13 | Peng Cheng Laboratory | 1 | |
| 14 | Rama Krishna Pasupuleti | 1 | |
| 15 | Beijing Institute of Technology | 1 | |
| 16 | Liang Xiang | 1 | |
| 17 | UT-Battelle LLC | 1 | |
| 18 | Ohio State Innovation Foundation | 1 | |
| 19 | Hangzhou Dianzi University | 1 | |
| 20 | NEC Corporation | 1 |
Innatera’s position as a pure-play neuromorphic chip startup with a commanding lead over Intel — a large incumbent — implies that the specialist route to IP in this field has so far outpaced the big-platform approach. Challengers seeking differentiation will need to find angles not already claimed by the current top three.
The most recent 18–24 months of filings are likely under-counted due to standard patent publication lag; figures for 2024–2026 should be treated as provisional minimums. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.
Multi-year growth continues despite an eased peak; AI-model computing dominates the technology mix
The filing trend and technology composition charts together show a field that has grown substantially since 2017 but is concentrated in a narrow set of IPC classes, with several adjacent branches attracting minimal attention.
Annual filing trend
Filings grew from low single digits in 2017–2018 to a clear spike in 2019, subsided through 2020–2021, then surged again to a new peak in 2022. The three-year window ending in the most recent full year sits 57% above the prior three-year window, confirming net multi-year growth. Annual volume has eased from the 2022 peak; figures for 2024–2026 are further suppressed by publication lag and will revise upward.
↗ Hover for values · click a bar to ask EurekaTechnology composition
G06N (Computing based on AI models) accounts for the overwhelming majority of IPC records, with G06F (Electric digital data processing) a distant second and H04L (Digital information transmission) a small third. Branches covering image/video recognition, engine control, ignition systems, radiation measurement, and material analysis each appear only once or twice, pointing to a field that has not yet spread broadly across application domains.
↗ Hover for values · click a bar to ask EurekaHighly cited patent families surfaced by the query
Citation-heavy patent families returned by the query. Use this section as citation context, not as a curated list of the most topic-specific patents.
Hyper-parameter optimization method for spiking ne…
Provided are a hyperparameter optimizer and method for optimizing hyperparameters and a spiking neural network processing unit. The optimizer includes a statistical analyzer configured to receive training data and perform statistical analysis on the training data, an objective function generator configured to generate hyperparameter-specific objective… (excerpt from the patent abstract)
| # | Patent | Citations |
|---|---|---|
| 1 | Spiking neural network accelerator using external … | 74 |
| 2 | Procedural neural network synaptic connection modes | 33 |
| 3 | Spiking neural network simulator for image and vid… | 30 |
| 4 | Resilient Neural Network | 24 |
| 5 | Resilient neural network | 21 |
| 6 | Neuromorphic accelerator multitasking | 17 |
| 7 | 一种脉冲神经网络硬件电路 | 10 |
| 8 | Spiking neural network by 3D network on-chip | 9 |
Ranked by total forward citations. Citation counts favour older and broadly cited patent families, and broad or adjacent patents may appear when they match the search scope. Treat this section as citation context, not as a curated list of the most topic-specific patents. Some patent titles may be shown in their original, non-English language where an accurate translation could not be guaranteed.
What the structure means for R&D investment decisions
The combination of a growth-stage lifecycle, extreme concentration at the top, nascent academic collaboration, and US-led filing geography shapes where practical entry points exist and where risks are highest.
Growth stage, with annual volume easing from a 2022 peak
The lifecycle evidence places this field firmly in the Growth stage: the recent three-year filing window is 57% above the prior three-year window. Annual volume has eased from its 2022 high, but publication lag means 2024–2026 data are provisional. New entrants can still establish meaningful positions, but the window before consolidation is narrowing.
Growth stageTop five filers hold 65% of the hundred largest filers’ combined output
The top five applicants’ share of the hundred largest filers stands at 65%, with a single company — Innatera Nanosystems — accounting for the dominant slice. Intel Corporation and Fraunhofer Society form a secondary tier, but both sit well below the leader. The long tail is populated almost entirely by universities and research institutes with one or two patent families each, suggesting limited commercial depth outside the top three.
High concentrationOnly one active co-filing pair identified: Huawei and the Institute of Semiconductors (CAS)
The single documented collaboration pairs Huawei Technologies with the Institute of Semiconductors of the Chinese Academy of Sciences, with two co-filed patent families. No other co-applicant relationships appear in the evidence. The absence of broad cross-institutional collaboration is notable for a hardware field where chip design and algorithm co-optimization typically require joint expertise; it may reflect the early stage of the ecosystem or proprietary development preferences among the leading players.
Minimal co-filingUS leads filings; Europe and China present but secondary
The United States is the primary filing jurisdiction. Europe via the EPO and China represent secondary markets. WIPO PCT filings indicate some applicants are pursuing broad international coverage. India and Germany also appear as separate filing destinations. This distribution reflects both the location of leading assignees (Netherlands, US, Germany) and the commercial importance of the US and European markets for neuromorphic chip deployment.
US-led, global spreadGo beyond the landscape: Eureka’s TRIZ Solution agent breaks down an R&D problem and returns patented concept solutions, each with a technical approach and cited patent & literature evidence.
| Applicant | Collaborator | Co-filings |
|---|---|---|
| Huawei Technologies Co., Ltd. | Institute of Semiconductors, Chinese Academy of Sciences | 2 |
Co-filing pairs, ranked by the number of jointly-filed patent families.
Innatera dominates on volume; Intel and Fraunhofer anchor the second tier with distinct technology emphases
The top two commercial players and the leading research institute show meaningfully different technology strategies, and all three entered the most recent filing period as new or significantly accelerating participants.
Innatera Nanosystems BV
Innatera holds 27 patent families, nearly three times the count of the next-ranked applicant. Its technology focus is concentrated in G06N 3 (AI-model computing, 27 records), with secondary positions in G06F 15 (digital data processing) and H04L 49 (digital information transmission), indicating a vertically integrated hardware-plus-connectivity approach. Momentum is classified as a new entrant with 15 families filed in the recent period — the largest recent-window contribution in the entire landscape.
families: 27Intel Corporation
Intel holds 9 patent families, entirely within G06N 3 (AI-model computing, 9 records) with a small secondary position in G06F 12 (digital data processing). Unlike Innatera’s multi-branch approach, Intel’s portfolio is tightly scoped to the core neural-computing model class. Intel does not appear in the applicant momentum table for the recent window, suggesting its filing pace has not accelerated in parallel with Innatera’s recent surge.
families: 9| Applicant | Recent (3 yrs) | Trend |
|---|---|---|
| Innatera Nanosystems BV | 15 | ▲ new entrant |
| Fraunhofer Society | 6 | ▲ new entrant |
| University of Electronic Science and Technology of China | 2 | ▲ new entrant |
| Electric Power Research Institute of State Grid Zhejiang Electric Power Co., Ltd. | 2 | ▲ new entrant |
| Huawei Technologies Co., Ltd. | 2 | ▲ new entrant |
| Peking University | 2 | ▲ new entrant |
Under-served adjacent branches worth monitoring
Several IPC classes adjacent to the dominant G06N core show very low patent counts relative to the field’s overall activity, representing areas where technical coverage is sparse and where a targeted filing program could establish an early position.
H04L · Digital information transmission
With only 6 records and a 7% share of IPC classifications, digital information transmission — covering on-chip and chip-to-chip communication protocols essential for neuromorphic hardware — is the most populated of the sparse branches but still well below what its technical relevance would suggest. Innatera itself holds H04L 49 filings, confirming the branch’s relevance; however, no other leading applicant has staked a significant position here. An entrant with expertise in network-on-chip or spike-routing protocols could file into this branch without encountering dense prior art from competitors.
Search this in Eureka →G06V · Image/video recognition
Only 2 IPC records fall under G06V (image and video recognition), despite event-based vision sensors being one of the primary application targets for neuromorphic hardware. The cited patent landscape includes a spiking neural network simulator for image and video processing among the most-cited works, confirming technical relevance. The sparsity in this branch likely reflects that applicants are classifying vision-processing inventions primarily under G06N rather than G06V, but it also leaves a gap for application-layer hardware patents oriented toward event cameras and edge vision inference.
Search this in Eureka →How leading applicants differ across technology routes
Strength of each leader across the main technology routes.
| Player | G06N 3 · Computing based on AI models | G06F 15 · Electric digital data processing | H04L 45 · Digital information transmission | H04L 49 · Digital information transmission | G06F 12 · Electric digital data processing |
|---|---|---|---|---|---|
| Innatera Nanosystems BV | Strong · 27 | Emerging · 4 | Emerging · 3 | Emerging · 4 | Absent |
| Intel Corporation | Strong · 9 | Absent | Absent | Absent | Moderate · 2 |
| The University of Aizu | Strong · 2 | Absent | Strong · 2 | Moderate · 1 | Absent |
| Fraunhofer Society | Absent | Strong · 5 | Absent | Absent | Absent |
| University of Dayton | Strong · 3 | Absent | Absent | Absent | Absent |
| Institute of Semiconductors, Chinese Academy of Sciences | Strong · 2 | Absent | Absent | Absent | Absent |
| Peking University | Strong · 2 | Absent | Absent | Absent | Absent |
Frequently asked questions
The evidence covers 69 patent families in scope across the global landscape as tracked in this analysis.
Innatera Nanosystems BV leads with 27 patent families, followed by Intel Corporation with 9 and Fraunhofer Society with 6.
The field is in a Growth lifecycle stage. The recent three-year filing window is 57% above the prior three-year window on a multi-year basis, though annual volume has eased from its 2022 peak. The most recent years (2024–2026) are further under-counted due to standard patent publication lag.
The United States leads with 24 records, followed by Europe via the EPO with 15, China with 9, and WIPO PCT with 6. India and Germany also appear as filing destinations.
G06N (Computing based on AI models) dominates. H04L (Digital information transmission) and G06V (Image/video recognition) are the most notable underserved adjacent branches, with 6 and 2 records respectively, despite clear technical relevance to neuromorphic hardware deployments.
Only one co-filing collaboration is documented in the evidence: Huawei Technologies and the Institute of Semiconductors of the Chinese Academy of Sciences, with 2 jointly filed patent families. No other cross-institutional co-filing relationships appear in the data.
Map your own spiking neural network hardware patent landscape
Join 18,000+ innovators using PatSnap Eureka to map any technology landscape: search 2B+ patents and papers, surface key assignees, and generate a report like this in minutes.
Disclaimer. This page is generated from PatSnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. PatSnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.