Sub‐optimal Internet of Thing devices deployment using branch and bound method

The Internet of Thing (IoT) network deployments are widely investigated in 4G and 5G systems and will still be key technical systems to drive massive connectivity of 6G systems. In 6G, IoT systems will operate with 6G new technologies such as Integrated Sensing and Communications. The IoT systems of 6G will be a platform to collect information in real world and create new use cases and business models. As the IoT devices and cellular networks are getting smarter, the IoT ecosystem allows us to bridge between human life and digital life and accelerate the transition towards a hyper‐connected world. Optimal and scalable IoT network design has been investigated in many research groups but key challenges in this topic still remain. An IoT devices deployment problem is investigated to minimise the transmission and computation cost among network nodes. The IoT devices deployment problem is formulated as Mixed‐Integer Nonlinear Programming problem. After relaxing the constraints and transforming the problem to a mixed integer linear programming (MILP) problem, the authors propose a new branch and bound (BB) method with a machine learning function and solve the MILP problem as a sub‐optimal solution. In the numerical analysis, the authors evaluate both conventional BB method and the proposed BB method with weighting factors and compare the objective function values, the number of explored nodes, and computational time. The performances of the proposed BB method are significantly improved under the given simulation configuration. The author finds the optimal mapping of IoT devices to fusion nodes.