Optimization of Mixed Numerology Profiles for 5G Wireless Communication Scenarios.

The management of 5G resources is a demanding task, requiring proper planning of operating numerology indexes and spectrum allocation according to current traffic needs. In addition, any reconfigurations to adapt to the current traffic pattern should be minimized to reduce signaling overhead. In this article, the pre-planning of numerology profiles is proposed to address this problem, and a mathematical optimization model for their planning is developed. The idea is to explore requirements and impairments usually present in a given wireless communication scenario to build numerology profiles and then adopt one of the profiles according to the current users/traffic pattern. The model allows the optimization of mixed numerologies in future 5G systems under any wireless communication scenario, with specific service requirements and impairments, and under any traffic scenario. Results show that, depending on the granularity of the profiles, the proposed optimization model is able to provide satisfaction levels of 60-100%, whereas a non-optimized approach provides 40-65%, while minimizing the total number of numerology indexes in operation.

Analysis of Machine Learning Techniques Applied to Sensory Detection of Vehicles in Intelligent Crosswalks.

Improving road safety through artificial intelligence-based systems is now crucial turning smart cities into a reality. Under this highly relevant and extensive heading, an approach is proposed to improve vehicle detection in smart crosswalks using machine learning models. Contrarily to classic fuzzy classifiers, machine learning models do not require the readjustment of labels that depend on the location of the system and the road conditions. Several machine learning models were trained and tested using real traffic data taken from urban scenarios in both Portugal and Spain. These include random forest, time-series forecasting, multi-layer perceptron, support vector machine, and logistic regression models. A deep reinforcement learning agent, based on a state-of-the-art double-deep recurrent Q-network, is also designed and compared with the machine learning models just mentioned. Results show that the machine learning models can efficiently replace the classic fuzzy classifier.