![]() |
市場調查報告書
商品編碼
2100016
6G:市場佔有率分析、產業趨勢與統計、成長預測(2025-2030)6G - Market Share Analysis, Industry Trends & Statistics, Growth Forecasts (2025 - 2030) |
||||||
※ 本網頁內容可能與最新版本有所差異。詳細情況請與我們聯繫。
據 Mordor Intelligence 稱,2025 年 6G 市值為 2.6 億美元,預計到 2030 年將達到 38.8 億美元。
這相當於年複合成長率為72.6%。

[1] IEEE Staff,“面向 6G 的兆赫通訊”,IEEE Xplore,ieeexplore.ieee.org。這種快速成長是由兆赫頻段、人工智慧 (AI) 在整個網路協定堆疊中的整合以及將覆蓋範圍擴展到傳統基地台範圍之外的非地面電波網路的早期商業化所驅動的。本報告按設備(行動設備、物聯網和邊緣設備等)、組件(硬體、軟體和服務)、最終用戶產業(汽車和交通運輸、製造和工業等)、頻段(亞太兆赫和兆赫)以及地區進行細分。
Terabit特級擴增實境(XR)工作負載如今已超過100 Gbps,迫使通訊業者重新設計回程傳輸與去程傳輸拓樸結構。 NTT Docomo在2024年進行的一項實驗在兆赫頻段實現了280 Gbps的吞吐量,證明了頻寬在工業數位孿生和全像會議方面的成熟度。在製造業領域,約翰迪爾的低延遲專用網路改進了預測性維護週期和機器人協作,為工業4.0環境下的6G樹立了標竿。醫療領域的需求同樣嚴苛。一項使用5G實現的99毫秒往返遠距手術演示已經為6G醫療通訊鏈路樹立了性能標準。邊緣AI模型訓練進一步增加了吞吐量的負載,因為它需要在數千個節點上即時同步大規模參數更新,而確定性延遲對於分散式學習至關重要。
NVIDIA 的 AI Aerial 平台將即時機器學習推理引入無線層,推動整個網路轉型為分散式智慧。目前,輕量級變壓器模型在 Raspberry Pi叢集上每秒可產生 5-12 個令牌,證明了獨立於雲端的本地推理的可行性。到 2030 年,全球物聯網終端數量將接近 750 億,利用環境射頻波的能源採集無線電對於實現無電池運作和支援持續連接模型至關重要(ResearchGate)。邊緣運算可以減少長距離流量,但在實際應用中,數百萬個自主設備必須在微秒內達成共識才能安全協作,這提高了可靠性要求。
資本密集度是最大的障礙。在完成大規模5G 部署後,通訊業者正在削減開支,優先考慮軟體增強,而不是批准部署高密度兆赫(THz)小型基地台網路。兆赫的傳播特性導致每平方公里需要比 5G 更多的基地台,從而推高了土地購買和電力成本。開放式無線接取網路 (Open RAN) 可減少廠商鎖定,但會增加整合複雜度並減緩投資回報週期 (MDPI)。雲端原生核心網路透過將成本從資本投資轉向計量收費模式,將成本分攤到長期,但其經濟效益只有在部署規模擴大後才會顯現。
2024年,行動領域佔據了6G市場佔有率的46%,這主要得益於智慧型手機的主導地位。然而,物聯網和邊緣設備的複合年成長率高達73.22%,是業界最高的,凸顯了流量模式向以機器為中心的轉變。固定無線接取設備的出貨量增加,使得即使在農村家庭也能享受到類似光纖的速度,這在行動裝置普及之前,為通訊業者提供了暫時的收入來源。
邊緣設備整合了本地人工智慧晶片,可在數據源端最佳化數據,從而在不影響分析精度的前提下提高頻譜效率。三星在7GHz頻段開發的X-MIMO原型展示了行動電話天線技術的創新如何抵消都市區「峽谷效應」造成的高頻傳輸損耗。同時,耐環境的工業感測器和汽車通訊模組正透過內建能源採集等容錯功能得到增強,從而拓展了其應用場景。
自主無人機、農業機器人和擴增實境(AR)頭顯的普及顯著提高了節點密度,進而推動了基礎設施需求的進一步成長。企業對專用6G網路(通常以物聯網閘道為核心)的需求,使得設備多樣性既成為服務供應商的技術挑戰,也成為其創造收入的驅動力。隨著人工智慧工作負載向邊緣遷移,韌體更新和安全修補程式將日益成為網路流量的重要組成部分,這進一步凸顯了零接觸設備管理平台的重要性。
即使到了2024年,硬體仍將佔6G市場規模的54%,其中包括無線前端、相位陣列天線和化合物半導體晶片組。然而,隨著通訊業者採用雲端原生核心和AI驅動的編配,無需更換硬體即可實現無線功能,軟體堆疊正以77.10%的複合年成長率快速成長。
開放式無線接取網路(Open RAN)將基頻和無線功能分離,使軟體供應商能夠迭代改進排程、波束成形和網路切片等技術。 NVIDIA 的「AI Aerial」套件展示了即時強化學習代理程式如何透過動態調整調變方案,在毫秒時間內最佳化小區吞吐量。隨著通訊業者向基於訂閱的網路功能收費模式轉型,涵蓋規劃、整合和生命週期管理的服務部分也受益於持續的收入成長。
儘管複合半導體技術的進步對於提高兆赫功率放大器的功率效率仍然至關重要,但其價值正逐漸轉向能夠從每赫茲頻率中提取更多容量的演算法控制。因此,供應商之間的差異化如今不再取決於晶片的前置作業時間,而是取決於更新速度,這正在重塑由設備製造商、超大規模資料中心業者和軟體新創公司組成的傳統生態系統。
預計到2024年,亞太地區將引領6G市場,佔36%的市場佔有率,並有望實現最快成長,到2030年複合年成長率將達到74.08%。中國將於2024年發射首顆6G測試衛星,而韓國的公私合作藍圖旨在2028年實現6G商業化。日本的「超越5G促進聯盟」已成功展示了在兆赫頻段超過100Gbps的傳輸,凸顯了該地區設備製造和現場測試一體化生態系統的完善。人口稠密的城市地區透過降低人均基礎設施成本和從早期用戶獲得收益,進一步加速了投資週期。
北美受益於聯邦政府的大量津貼以及在雲端原生領域的領先地位。一項4.2億美元的聯邦創新基金正在加速開放原始碼入網(RAN)的普及,並賦予國內供應商在標準化方面的話語權。矽谷在人工智慧領域的領先地位使美國公司能夠定義網路智慧框架,儘管對進口鎵和磷化銦晶片的依賴仍然是供應鏈的薄弱環節。與加拿大和墨西哥的跨境頻段協調正在縮小漫遊差異並擴大用戶群體。
在歐洲,政策重點在於協調永續性與監管。智慧網路與服務聯合營業單位)正投資5億歐元,以推動環保型無線和能源採集設備的研發。德國和英國在衛星和地面電波網路測試方面處於主導,而法國和義大利則利用其航太領域的專業知識來改進星間光鏈路。各國監理政策的差異正在減緩泛歐一體化的發展,但通用的研究主題和漫遊指令正在縮小這些差距。中東和非洲雖然目前規模較小,但預計將透過依靠低地球軌道(LEO)衛星回程傳輸跨越光纖鴻溝,一旦行動電話價格降至大眾市場閾值以下,就能迅速趕上。
According to Mordor Intelligence, the 6G market is valued at USD 0.26 billion in 2025 and is projected to reach USD 3.88 billion by 2030, reflecting a compound annual growth rate (CAGR) of 72.6%.

[1] IEEE Staff, "Terahertz Communications for 6G," IEEE Xplore, ieeexplore.ieee.org The surge is propelled by terahertz-band spectrum, artificial intelligence (AI) integration across the network stack, and the early commercialisation of non-terrestrial networks that extend coverage beyond traditional cell sites. This report is Segmented by Devices (Mobile Devices, Iot and Edge Devices, and More), Component (Hardware, Software, and Services), End-User Vertical (Automotive and Transportation, Manufacturing and Industrial, and More), Frequency Band (Sub-Terahertz (sub-THz) and Terahertz), and Geography.
Terabit-class extended-reality workloads now exceed 100 Gbps, forcing operators to redesign back-haul and fronthaul topologies. NTT DOCOMO's 2024 lab trials delivered 280 Gbps in the terahertz band, validating spectral maturity for industrial digital twins and holographic conferencing. In manufacturing, John Deere's low-latency private network improved predictive maintenance cycles and robotic coordination, setting a benchmark for 6G in Industry 4.0 settings. Healthcare demands are equally stringent; telesurgery demos over 5G at a 99-millisecond round-trip have already shaped the performance baseline for 6G medical links. AI model training at the edge compounds throughput pressure because large parameter updates must synchronise across thousands of nodes in real time, making deterministic latency a pre-requisite for distributed learning.
NVIDIA's AI Aerial platform brings real-time machine-learning inference into the radio layer, driving a network-wide shift toward distributed intelligence. Lightweight transformer models now generate 5-12 tokens per second on Raspberry Pi clusters, underscoring the feasibility of local inference without cloud dependency. As global IoT endpoints approach 75 billion by 2030, energy-harvesting radios using ambient RF are essential for battery-free operation, underpinning continuous connectivity models ResearchGate. Edge computing cuts long-haul traffic but heightens reliability needs, since millions of autonomous devices must reach consensus in microseconds to collaborate safely in real-world environments.
Capital intensity is the primary headwind. Operators are trimming spend after heavy 5G roll-outs, prioritising software overlays before green-lighting dense THz small-cell grids. Terahertz propagation limits demand more base stations per square kilometre than 5G, pushing site-acquisition and power costs upward. Open RAN reduces vendor lock-in yet raises integration complexity, delaying return on investment MDPI. Cloud-native cores spread cost over time by shifting from capital expenditure to consumption-based models, but the financial benefit only materialises once adoption scales.
Other drivers and restraints analyzed in the detailed report include:
For complete list of drivers and restraints, kindly check the Table Of Contents.
The mobile segment commanded 46% of 6G market share in 2024 on the back of smartphone dominance, but IoT and edge devices are expanding at a sector-leading 73.22% CAGR, underscoring a shift toward machine-centric traffic patterns. Rising shipments of fixed wireless access units enable rural households to achieve fibre-like speeds, creating an interim revenue stream for operators before handheld mass adoption.
Edge devices embed local AI chips that prune data at the source, improving spectral efficiency without sacrificing insight accuracy. X-MIMO prototypes in the 7 GHz band from Samsung demonstrate how handset antenna innovations will offset high-frequency path loss in urban canyons. Meanwhile, ruggedised industrial sensors and vehicle communication modules add resilience features such as built-in energy harvesters, widening use-case diversity.
The proliferation of autonomous drones, agricultural robots, and AR headsets increases node density by orders of magnitude, driving incremental infrastructure requirements. Enterprise demand for private 6G networks, often anchored by IoT gateways, positions device diversity as both a technical challenge and a revenue multiplier for service providers. As AI workloads migrate to the edge, firmware updates and security patches will account for a growing share of traffic, reinforcing the primacy of zero-touch device-management platforms.
Hardware still represented 54% of the 6G market size in 2024, covering radio front-ends, phased-array antennas, and compound-semiconductor chipsets. However, the software stack is scaling at 77.10% CAGR as operators deploy cloud-native cores and AI-driven orchestration that permit over-the-air feature activation without hardware swaps.
Open RAN splits baseband and radio functions, allowing software vendors to iterate on scheduling, beamforming, and network slicing. NVIDIA's AI Aerial suite illustrates how real-time reinforcement-learning agents optimise cell throughput in milliseconds by tuning modulation schemes on the fly. Service segments-covering planning, integration, and lifecycle management-benefit from recurring revenue as carriers move to subscription pricing for network functions.
Compound-semiconductor advances remain critical for power efficiency in terahertz power amplifiers. Yet value is tilting toward algorithmic control that extracts more capacity from each hertz. Consequently, vendor differentiation hinges on update velocity, not silicon lead-time, reconfiguring ecosystem alliances among traditional equipment makers, hyperscalers and software start-ups.
Asia-Pacific led with 36% of the 6G market in 2024 and is scaling fastest at a 74.08% CAGR through 2030. China launched the first 6G test satellite in 2024, while South Korea's public-private roadmap aims for commercial 6G in 2028. Japan's Beyond 5G Promotion Consortium has already demonstrated terahertz transmissions above 100 Gbps, underscoring the region's integrated device-fabrication and field-trial ecosystem. Dense urban clusters reduce per-capita infrastructure cost and create early-adopter revenue pools that further accelerate investment cycles.
North America benefits from substantial federal grants and cloud-native leadership. The USD 420 million federal innovation fund catalyses open-source RAN deployments, giving domestic vendors a seat at the standards table. Silicon Valley's AI leadership positions U.S. firms to define network-intelligence frameworks, though reliance on imported gallium and indium phosphide chips remains a supply-chain weak spot. Cross-border spectrum alignment with Canada and Mexico narrows roaming gaps and enlarges the addressable subscriber base.
Europe channels policy energy into sustainability and harmonised regulation. The Smart Networks and Services Joint Undertaking deploys EUR 500 million to nurture green radios and energy-harvesting devices. Germany and the United Kingdom spearhead testbeds for satellite-terrestrial networks, while France and Italy exploit aerospace expertise to refine optical inter-satellite links. Fragmented national rules slow pan-European roll-outs, yet common research agendas and roaming directives mitigate divergence. The Middle East and Africa, though smaller today, rely on LEO satellite back-haul to leapfrog fibre gaps, positioning those regions for rapid catch-up once handset prices fall below mass-market thresholds.