Influence of Training Set Selection in Artificial Neural Network-Based Propagation Path Loss Predictions

This paper analyzes the use of artificial neural networks (ANNs) for predicting the received power/path loss in both outdoor and indoor links. The approach followed has been a combined use of ANNs and ray-tracing, the latter allowing the identification and parameterization of the so-called dominant path. A complete description of the process for creating and training an ANN-based model is presented with special emphasis on the training process. More specifically, we will be discussing various techniques to arrive at valid predictions focusing on an optimum selection of the training set. A quantitative analysis based on results from two narrowband measurement campaigns, one outdoors and the other indoors, is also presented.

Synthesis and structural characterisation of [Cu2[(PDC)2(bpa)(H2O)2]•nH2O•mDMF (n=3 and m=1 for 1, n=7 and m=0 for 2, PDC=2,5-pyrindinedicarboxylate and bpa= 1,2-di(4-pyridil)ethane))

[EN] Combination of polycarboxylate anions and dipyridyl ligands is an effective strategy to produce solid coordination frameworks (SCF) which are crystalline materials based on connections between metal ions through organic ligands. In this sense, combination of polycarboxylate anions and dipyridyl ligands is an effective strategy to produce extended structures. In this context, this work is focused on two novel CuII-based SCFs exhibiting PDC (2,5-pyridinedicarboxylate) and bpa (1,2-di(4-pyridyl)ethane), being the first structures reported in literature containing both ligands. Chemical formula are [Cu2[(PDC)2(bpa)(H2O)2]•3H2O•DMF (1), and [Cu2(PDC)2(bpa)(H2O)2]•7H2O (2), where DMF is dimethylformamide. Compounds 1 and 2 have been characterized by means of XRD, IR, TG/DTG, and DTA analysis.