A Hybrid Path Loss Prediction Model based on Artificial Neural Networks using Empirical Models for LTE And LTE-A at 800 MHz and 2600 MHz

Abstract

LTE's peak data rate for downlink and uplink can reach 326.4 and 86.4 Mbps, respectively [2], while LTE-A significantly enhances these specifications: it increases peak rates, achieving 3 Gbps for downlink and 1.5 Gbps for uplink; for such, it requires a bandwidth up to 100 MHz [3].A Hybrid Path Loss Prediction Model based on Artificial Neural Networks using Empirical Models for LTE And LTE-A at 800 MHz and 2600 MHz Bruno J. Cavalcanti, Gustavo A. CavalcanteFederal Institute of Education, Science and Technology Paraíba Campus Campina Grande, Brazil, bruno.cavalcanti@ifpb.edu.br, gustavo.cavalcante@ifpb.edu.br LaércioFederal University of Rio Grande do Norte, Caixa Postal 1655, CEP: 59078-970, Natal, RN, Brazil, laercio@ct.ufrn.br, gabrielmocan@bct.ect.ufrn.br, marcelo_medeiros_5@yahoo.com.br, adaildo@ct.ufrn.br, Journal of Microwaves, Optoelectronics and Electromagnetic Applications, Vol. 16, No. 3, September 2017 Path loss models are important for predicting coverage area, interference analysis, frequency assignments, and cell parameters -basic components for the network-planning process in the project of a mobile communications system [4]. Understanding the radio channel for the network deployment is utmost, being the modelling of the radio channel using the most appropriate path loss model, an essential factor.Propagation models can be classified [5,6] as: deterministic, empirical, and physical/statistical. The first ones can be considered the most accurate method. They are based on the behavior of radio waves propagated in space, calculating propagation losses mathematically, based on theoretical formulation.For such, accurate information is necessary, not only about buildings and terrains, but also about reflection and diffraction coefficients of the surfaces which are in the propagation path.Meanwhile, empirical models do not accurately predict the radio waves comportment, depending more on field strength from that specific environment to give an approximation based on measurements. Lastly, physical/statistical models combines empirical and statistic information about the environment, aiming to decrease computational cost.In order to make the communications systems more accurate -to have a more efficient planning, many efforts have been made towards the development of coverage prediction simulation methods and tools able to accurately estimate on measured data. In this sense, some techniques can help to provide more efficient simulation methods, reducing errors and providing more trustworthy results. The problem in path loss prediction between two points can be interpreted as a solution to obtain a function of several inputs and a single output, where the inputs contain information like locations of the transmitter and receiver, frequency and surrounding buildings.Thus, the prediction of path loss can be described as the transformation of an input vector containing topographical and morphological information about the environment to the desired output value [11]. Since neural networks can be effe...

Model-Aided Deep Learning Method for Path Loss Prediction in Mobile Communication Systems at 2.6 GHz

“…It has been shown in [25] that expert knowledge (modelaided learning) can be embedded into wireless systems for optimization. Additionally, such hybrid-based models have shown to be effective for path loss prediction as documented in [26]. In this work, the proposed DL model uses a simple path loss model for assisting in the learning process.…”

New Empirical Path Loss Model for 28 GHz and 38 GHz Millimeter Wave in Indoor Urban under Various Conditions

“…Fifth generation cellular networks (5G) offer unprecedented bandwidth and faster network speeds by tapping into a high band spectrum in the mmWave frequency range, typically from 30 to 300 GHz [12,13]. Propagation is more lossy in the mmWave bands when compared to today's microwave bands, thus the solution is to modify the standard path loss models (with a frequency between 28 and 38 GHz) to fit the real world measured propagation data for reliable mmWave channel planning [14,15]. However, the previous models (Okumura, COST-Hata) do not work with frequencies between 28 and 38 GHz, and thus they are not compatible with the 5G [16,17].…”

“…While some obstacles absorb and others reflect the radio signal; similarly, they all limit the transmission ability of signals ( Figure 2) [6]. 9.2° and 49.7° half-power band width (HPBW) antenna at receiver (RX) [14][15][16]. The measurement propagation [14][15][16] combines power delay profiles (PDPs) for single point angles using high directions (8,19).…”

“…The green dots in Figure 3 show the distances at the TX locations, which went up to 300 m. For RX locations, we used half-power Beam width antennas (HPBW) with 25.3 dB (11.2 • HPBW) and 16 dB (28.9 • HPBW) gain antennas at the TX and RX for the largest range of measurement of 184 dB incoming signals from different angles of arrivals. The data measurement was collected over distances of 25 to 500 m but was cut from 20 to 251 m based on the power failure [14]. For the frequency of 38 GHz propagation (Table 2), the signal was measured at 38 GHz RX at the University of Bridgeport, Connecticut, USA (see Figure 3 for different distances up to 300 m with six different handheld measurements, using two different transition locations).…”

“…Most TX-RX locations used 25.3 dBi (9.2 • HPBW) gain directional antennas, while others used 14 dBi gain (49.7 • HPBW) antennas at the RX. The data measurement was collected over distances of 25 to 500 m [14]. Tables 1 and 2 show path loss at a 3 m frequency for a distance signal, which was obtained with a high antenna gain at 28 and 38 GHz.…”