The wave of intelligence is revolutionizing industries globally, with 5G, AI, and IoT playing key roles in this trend. During Computex 2024, Fibocom CEO - Tiger Ying shared his insights during the trade show.
Fibocom, as China's first stock-listed wireless communication modules and solutions provider (stock code: 300638), has continuously employed innovative thinking and strategic market positioning in recent years. CEO Tiger Ying pointed out that, regardless of the technological or application aspects, in facing the transformative effects brought by these three major technologies, the company is well-prepared and ready to assist clients in seizing vast smart business opportunities.
Remain Industry-Focused and is Confident to Accelerate 5G Deployment
In the realm of 5G, this communication standard garnered significant market attention upon its release. The industry believes that 5G'sfeatures such as high speed, low latency, and large-scale connectivity not only enhance user experiences in the consumer sector but also serve as a crucial driver for deepening communication technology applications across industries.
While the development of 5G has not been as fast as initially expected, communication giant Ericsson predicted in its 2023 report that the golden crossover period between 4G and 5G would be delayed until 2028.However, Tiger Ying pointed out that from the perspective of technical standards' development speed, this delay is still within a normal cycle.
He mentioned that 5G's current development is focused on Enhanced Mobile Broadband (eMBB), Massive Machine Type Communications (mMTC), and Ultra-Reliable Low Latency Communications (URLLC). However, large-scale commercialization has not been achieved. This trend also highlights the versatility of 5G IoT architecture and applications.
With the acceleration of 5G Advanced, it will truly replace4G as the mainstream technology for IoT applications and successfully drive market development within five years.
Fibocom has responded to the trend of 5G development by formulating product strategies and solutions, actively expanding into different market sectors. Tiger Ying stated that the company's primary task is to promote the large-scale commercialization process of eMBB.
They have already launched a diverse range of product portfolios. This includes the RedCap module, closely following the development of 5G technology and fully advancing the implementation of commercial plans.
In terms of application areas, Fibocom is focused on industrial intelligence and smart cities, primarily because these two major application scenarios have diverse demands for 5G technology. For instance, in machine vision, which integrates edge computing with 5G's high bandwidth and low latency capabilities, it has been widely used in industrial quality inspection, promoting the development of industrial automation and intelligence.
In response, Fibocom provides a complete product portfolio, such as the 5G smart module SC171 with computing power up to 12TOPS, to meet specific demands in application scenarios. In addition, endpoint devices with AI capabilities will also become a driving force in accelerating industrial intelligence processes.
Fibocom has invested significant research and development resources in these products, aiming to integrate 5G and on-device AI technologies, thus strengthening the company's core competitiveness in industrial intelligence and smart city domains.
Strengthening Edge AI Performance to Optimize Overall Cost-effectiveness
AI has become a recent global industrial focus, and Fibocom's AI strategy centers on edge computing and the AI endpoint device solutions mentioned above. Since AI computing power can significantly increase the cost of endpoint devices, leading to reluctance from enterprises with implementation needs, cost control is crucial.
Tiger Ying pointed out that the solution to this problem varies depending on the AI architecture. Devices using an offline AI architecture require precise control of edge computing power to balance cost and effectiveness, while devices using an online AI architecture require strong communication capabilities to avoid affecting user experience due to communication delays.
Tiger Ying further stated that Fibocom has integrated the above two solutions into one through long-term research and development. The company has invested in smart module development, leveraging the computational efficiency and heterogeneous algorithm capabilities of modules to assist clients in rapidly deploying AI-capable terminal devices and optimizing costs.
Its product portfolio includes 5G data modems, SoC 5G solutions based on Linux, and Android, and built-in AI computing capabilities. He mentioned that while AI may slightly increase device costs, its excellent performance leads to overall cost optimization considering macroscopic aspects such as user experience and value creation.
He also mentioned that Fibocom has constructed heterogeneous computing on the network, enabling the scheduling of CPU, GPU, NPU, and DSP processor performance as needed. In addition, Fibocom has developed its toolchain and integrated it with heterogeneous algorithms, RTK, high-precision positioning, and other functions into firmware.
Its products based on Linux, Android, and Windows architectures can meet the needs of different industry customers. Looking ahead, Fibocom will focus on areas such as robotic lawnmowers, low-speed autonomous vehicles, PCs, and robotic development platforms, providing deeper vertical domain solutions.
Enhancing π-shaped Capabilities to Meet Three Major IoT Demands
Regarding IoT, Tiger Ying pointed out that although the IoT architecture and concept have been around for over a decade and have become increasingly widespread in recent years, practical constraints mean that adoption varies across industries. To accelerate IoT adoption, several key factors are necessary.
First is the ability to apply AI capability to IoT devices, as AI significantly impacts data collection costs and processing quality. IoT leveraging AI will be more readily accepted by enterprises, thus speeding up adoption in specific fields.
Second is security design, which has become a major focus in IoT in recent years. Only IoT architectures that can ensure data integrity and privacy protection will be able to deploy in vertical markets.
Lastly, communication technology is crucial. Advanced technologies like 5G-Advanced (5.5G) and Non-terrestrial Networks (NTN) ensure that communication coverage is no longer a weak point, broadening the scope of IoT applications.
Fibocom has introduced corresponding products to address AI, data privacy, and satellite communications needs. In terms of AI, the company's 5G smart module SC171 and SC151 series are suitable for various 5G smart terminal devices.
For security, blockchain technology was deployed on 4G networks and commercialized. On the satellite communications front, Fibocom released NTN-supported communication modules in 2023.
Moving forward, the company plans to integrate these technologies and products, working with ecosystem partners to focus on specific vertical solutions, providing customers with high-quality and high-performance IoT solutions.
Fibocom consistently prioritizes intelligence in product innovation. Tiger Ying stated that this will continue to be the foundation for future development, with a market-driven approach to introducing cross-domain AI solutions.
He emphasized that while the market previously demanded "T-shaped" capabilities combining vertical telecommunication expertise to serve the horizontal industries, the AI era requires an additional vertical capability—AI specialization. Through "π-shaped" capabilities, which integrate horizontal and dual vertical expertise, Fibocom aims to create high-performance IoT architectures tailored to customers' needs.
The company is progressively enhancing these "π-shaped" capabilities. It will continue to collaborate with ecosystem partners, integrating 5G, AI, and IoT technologies to help clients seize smart business opportunities.
