Focusing on the national demand for optimizing communication and computing network infrastructure, and addressing the lack of advanced mathematical modeling and solution methods in the field of large-scale network optimization, this project leverages big data analysis, artificial intelligence, and operations optimization technologies to develop a unified framework and technical system for information network optimization algorithms. This will provide core support for the next generation of integrated communication-computation-sensing network optimization. Centered on the data-driven real-world network statistical simulation framework SRCON pioneered by Academician Luo Zhiquan's team, we aim to establish network performance modeling methods guided by physical mechanisms and driven by measured data. This includes proposing a large-scale network parameter tuning algorithm that integrates zeroth-order and distributed optimization. This approach revolutionizes traditional network optimization, which has relied on manual expertise and iterative trial and error, significantly enhancing network optimization efficiency and supporting the realization of intelligent autonomy for information networks.

Focusing on the demands and application scenarios of low-altitude sensing, this project leverages distributed signal processing and artificial intelligence technologies to develop a distributed low-altitude sensing prototype system and products. Common technologies driven by market demand will be iteratively updated, empowering the ongoing development of this industry in Shenzhen. Key technologies can be applied to areas including low-altitude sensing and national defense, thus supporting and enabling Shenzhen-based enterprises involved in low-altitude sensing.

Focusing on major standardized applications in sectors such as rail transit, high-energy-consuming factories, and power in the energy field, this project leverages operations research and artificial intelligence technologies. Starting from the perspective of adjustable loads and closely integrating with engineering practice, it follows the approach of "offline optimal planning - real-time re-optimization - rapid emergency recovery - full-system simulation" to develop specialized optimization solvers for the energy sector.