Indoor Genetic Algorithm-Based 5G Network Planning Using a Machine Learning Model for Path Loss Estimation

Abstract

Accurate wireless network planning is crucial for the deployment of new wireless services. This usually requires the consecutive evaluation of many candidate solutions, which is only feasible for simple path loss models, such as one-slope models or multi-wall models. However, such path loss models are quite straightforward and often do not deliver satisfactory estimations, eventually impacting the quality of the proposed network deployment. More advanced models, such as Indoor Dominant Path Loss models, are usually more accurate, but as their path loss calculation is much more time-consuming, it is no longer possible to evaluate a large set of candidate deployment solutions. Out of necessity, a heuristic network planning algorithm is then typically used, but the outcomes heavily depend on the quality of the heuristic. Therefore, this paper investigates the use of Machine Learning to approximate a complex 5G path loss model. The much lower calculation time allows using this model in a Genetic Algorithm-based network planning algorithm. The Machine Learning model is trained for two buildings and is validated on three other buildings, with a Mean Absolute Error below 3 dB. It is shown that the new approach is able to find a wireless network deployment solution with an equal, or smaller, amount of access points, while still providing the required coverage for at least 99.4% of the receiver locations and it does this 15 times faster. Unlike a heuristic approach, the proposed one also allows accounting for additional design criteria, such as maximal average received power throughout the building, or minimal exposure to radiofrequency signals in certain rooms.

5G mmWave Network Planning Using Machine Learning for Path Loss Estimation

“…To find the optimal deployment with a minimal number of 5G APs and still provide the required quality of service, we use the approach of Figure 1, and the link budget parameters from Table 3. The core of the system is a Genetic Algorithm [49]. The inputs are the floor plan and the PL estimation made with the ML algorithm for every possible AP-Rx link.…”

“…with X i,j , j = 1...M , a gene of the solution X i , representing either switched off (set to 0) or switched on (set to 1) [49]. The GA starts with an "Initialization" step during which the scenario parameters are set.…”

White spaces pattern finding and inference based on machine learning for multi-frequency spectrum footprints

“…Therefore, the computational complexity of machine learning increase exponentially with the degrees of freedom. For improving the algorithms computational performance further advancements have been done combining supervised learning for Deep Neural Networks and Reinforcement Learning like Deep Q-Learning [28] and also on mixed algorithms like Genetic Algorithms with supervised machine learning [29]. Although, a higher accuracy and adaptability to different environment and scenarios is achieved, it still comes at a significant computational cost that might not be effective for some use cases considering the current state of the art regarding Edge Tensor Processing Unit [30,31,32].…”

Predicting Path Loss of an Indoor Environment Using Artificial Intelligence in the 28-GHz Band

“…However, the implementation of ML methods in outdoor/indoor environments for modern communication systems faces various challenges. The challenges in applying ML methods to outdoor/indoor environments in modern communication systems are still an area of research [2]. Thus, this study aims to apply ML to predict path loss for an indoor scenario.…”