SBIR Phase I: 5G Network Performance and Demand Prediction for Smart Cities

小型企业创新研究(SBIR)第一阶段项目的更广泛影响/商业潜力将来自于启用自我发展的电信网络，以确保在大城市地区可以获得可靠的高速互联网接入，并且智能城市中的关键任务应用程序将获得有保证的服务质量，同时降低网络的资本和运营成本。此外，这项技术将通过在需要的地方和时间移动资源来降低电信网络的能源消耗，避免过度配置和闲置资源的浪费。这个小企业创新研究(SBIR)第一阶段项目开发了首个基于人工智能的网络性能和需求预测平台，以保证大都市地区以及最终智能城市的5G连接。到2050年，68%的世界人口居住在城市地区，不断的人口流动和5G应用网络需求的多样性，网络优化至关重要。然而，要实现最佳的网络配置，需要准确预测未来的网络需求。拟议的研究将利用大量人员流动及其活动的外部背景数据以及一系列机器学习方法来执行准确的网络需求预测，从而优化资源分配。该项目的主要目标是开发和部署一种基于云的自动化软件，最多可提前七天对城市地区的网络关键性能指标进行预测。该软件解决方案使网络运营商能够识别和预测准确的时间和空间需求和异常，了解将导致需求变化的因素，并根据这些信息确定未来的优化机会。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project will result from enabling self-evolving telecommunication networks to ensure that reliable and high-speed internet access is available in large metropolitan areas and mission critical applications in smart cities will receive guaranteed Quality of Service, while the capital and operational costs of the networks are reduced. Furthermore, this technology will reduce the energy consumption of telecommunication networks by moving resources where and when they are needed, avoiding over-provisioning and waste in idle resources.This Small Business Innovation Research (SBIR) Phase I project develops a first of its kind novel Artificial Intelligent-based network performance and demand prediction platform to guarantee 5G connectivity in metropolitan areas and eventually in smart cities. With 68% of the world population living in urban areas by 2050, constant movement of people and diversity in 5G applications network requirements, network optimization is critical. However, achieving optimal network configuration requires accurate prediction of future network demand. The proposed research will utilize the external contextual data of mass human movement and their activity along with a portfolio of machine learning methodologies to perform accurate network demand prediction and consequently optimal resource allocation. The main objective of this project is to develop and deploy an automated cloud-based software that performs prediction on network key performance indicators in urban areas up to seven days in advance. This software solution enables network operators to identify and anticipate accurate temporal and spatial demands and anomalies, understand the factors that will cause demand variations, and pinpoint future opportunities for optimization based on this information.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.