Part II: A Heuristic Approach to Classifying Different Multiple Urban Settings for Ambient RF Energy Harvesting Potential using FM Technology as an RF Energy Source

Part


The RF Harvester
Electromagnetic fields coming from cellular phones, Wi-Fi access points, and other transmitter technologies create disturbances that can be measured, which in turn potential energy that varies over time or distance. To be able to use the electromagnetic fields, RF energy harvesting technology is used (Nechibvute et al., 2017) to directly power up battery-less and remote battery recharging systems (Shrestha et al. 2013).
RF energy in power density can be harvested coming from different sources like the Global System for Mobile Communications (GSM), Television (TV), and Frequency Modulation (FM) radio frequency sources using RF harvester. RF harvester can be use to harvest FM signals from an ambient environment that can power up different wireless sensors (Urrehman et al., 2017). FM broadcast, GSM 900, GSM 1800, 3G and Wi-Fi has potentials for RF energy in a semi-urban area (Khalid et al., 2020). The paper aims to provide a guide model footprint for preliminary Potential energy harvesting capabilities coming from FM sources in different multiple urban settings because of the emerging technology of IoT peripherals and massive research and designs of RF energy harvesters.

Harvesting RF Energy and the Environment
A survey of measurements was carried out in the urban city of Tokyo, which showed that several wireless activities of TV broadcasts would give an average field intensity of .9 v/m (-17dbm) found. With the use of the electric field incident, a dipole RF harvester was designed. The performance of the designed RF harvester was evaluated through a field test at 6,500 meters from the Tokyo TV tower. The results showed that the RF harvester can be charged by a tank with a voltage of 2.9 volts and used for a battery-less wireless sensor operation in an urban-type environment (Vyas et al., 2012). Piñuela et al. (2013) conducted a study entitled "Ambient RF Energy Harvesting in Urban and Semi-urban Environments" which surveyed a city-wide RF in different points of London. The study showed that different power density results in urban and semi-urban places because of the number of stations, proximity distance, propagation characteristics, and line of sight scenario. The survey concluded that DTV, GSM 900, GSM 1800, and 3g were potentially useful ambient sources of RF energy and used these as the bases of the design of an RF harvester. The cell RF harvester they made demonstrated its operating level at Power Conversion Efficiencies (PCE) of 40% at -25 dBm in a semi-urban environment ( Figure 1). Parks et al. (2013) developed an RF harvester through characterized bench testing simulations. Their RF harvester was tested in a real-world urban environment which operated a sensor at 10.4 km and 200 meters with a 1 MW TV station and a cellular-based cell-site station. The study demonstrated that the minimal RF input power for an RF harvester to operate was -18dbm. In the study of Zhang et al. (2014), the group made simulations for the design of a wideband cross dipole RF energy harvester to optimize the urban environments' 200 uW / cm2 and below power density. This power density ranging from 5 to 200 uW/cm2 (-4dBm to 12dBm) was inputted to the RF harvester for design considerations and through simulations achieved a maximum Power Conversion Efficiency (PCE) of 57 to 120%. The result gave good qualities for RF energy scavenging/ harvesting. Nimo et al. (2015) in their works entitled "Analysis of Passive RF to DC Rectification and Wireless RF Energy for Micro-Watt Sensors" made an RF harvester through simulations and designs using different Power Received Level (PRL) inputs in an 868 MHz ISM band. Their work got a measurement of 1 volt at -14.4dBm and 6 volts at 5dBm. They tested the RF harvester in an ambient environment after simulations and with the power levels of -27dBm to -50dBm along with the distance of 110 meters a Direct Current (DC) voltage of 2.3 volts was obtained ( Figure 2).
In the works of Song et al. (2015) entitled "A Broadband Efficient Rectenna Array for wireless energy harvesting" the city of Liverpool was surveyed through the use of a spectrum analyzer, which showed a -30dbm to -10dbm power density/ power received levels. The results of the survey were used as an input for the design optimization of RF harvester for frequency bands 1.8 to 2.5 GHz, which was tested at a distance of 1 meter. The test showed that the RF rectenna harvester could give a maximum conversion efficiency of 55% at -10dbm and enable power up a wireless sensor. Borges et al. (2015) in their study entitled "Radio Frequency harvesting for Wearable Sensors" presented a small biomedical sensor that was fabricated and designed for urban and suburban areas with the use of computer simulations to determine the potential of cell sites' 900 and 1800 MHz as a source for RF energy harvesting in urban and suburban places. Through testing and designing by inputting different Power levels, the RF harvester got 3.61 Volts, 2.34 Volts, and 1.4 Volts for -2dBm, -5dBm, and -10dBm respectively along with the highest efficiency of 27 % attained at 6dBm. Gabrillo et al. (2015) in their study on "Enhanced RF to DC converter with LC resonant circuit" used a loop antenna via experimental testing in an urban setup and found out that at 77.84 meters from a TV signal output power of 1.80 MW / 2.55dbm can be measured. The harvester can power a device without battery maintenance which has better applications in wireless sensor setups. Shariati et al. (2015) made an RF harvester for urban environments that power electronic gadgets that require only a small amount of power. Through experimentations and simulations, the RF harvester gave a voltage of 772.8 mV at an input power of -10dBm. The D.C. output harvested at different RF energy sources increased by3.14 and 7.24 units? over the frequency of 490 and 860 MHz respectively. Shariati et. al tested the RF harvester in the suburban area of Melbourne, Australia, and perpetual RF harvesting for energizing low-powered systems was achieved. Khemar et al. (2017) exploited the use of 3G technology as an RF energy source for powering wireless sensor nodes. The group fabricated an RF harvester for a frequency band ranging from 917 MHz to 3 GHz and thru simulations obtained a 50% efficiency at -7dBm Power Received Levels (PRL). The field test measurements with an obtained voltage of 190 mV. A measured PRL of -10dBm to -7dBm was obtained for an urban environment which was relatively high. -ur-Rehman et al. (2017) made an RF harvester for an FM ambient environment. Through simulations with a power level of -10dBm to -30dBm, their experiment showed that at this power level, an RF harvester can harvest FM signals from an ambient environment that can power up different wireless sensors. The RF harvester was tested on a building for performance checking via a spectrum analyzer and demonstrated that at -14.2 dBm Power Received Level (PRL), the equivalent distance from an FM source under an ambient environment was 10 meters. Milanezi et al. (2017) In their study entitled "Radiofrequency Energy Harvesting System Based on a Rectenna Array in Urban Environments" Milanezi et al. (2017) conducted measurements with the use of a spectrum analyzer to evaluate the feasible energy power levels in Brasilia, Brazil at a reference distance of 300 to 1000 meters from the RF energy source. Their study showed an 11dBm average incident power from the urban environment at 91.7 MHz, which was used as an input for the design of their harvester. Their harvester acquired a converted energy efficiency of 18%. They stated that a 42 RF harvester at -3dBm or a 2 RF harvester at 10dBm can charge a cell phone for about an hour.

Mutee
In the paper of Khemar et al. (2018) entitled "Design and experiments of a dualband Rectenna for ambient RF energy harvesting in urban environments," a survey was made outside Paris. A -7dBm highest peak power was measured this peak power level was used to design an antenna. The test results of the proposed RF harvester got a Power Conversion Efficiency (PCE) of 33 % for a frequency of 1800 MHz along with an incident Power Received Level (PRL) of -7dBm in an urban zone. The authors concluded that a temperature sensor can be activated using the RF harvester. Takhedmit et al. (2018) in their work "RF Energy Harvesting in Urban Environments using Transparent Rectenna," showed their RF harvester antenna array design intended for an urban environment. The urban environment measurements used were based on the study of Pinuela et al. 2013. Their harvester for ISM band 2.45 GHz was designed thru simulations at the input of 1 to 10 uW / cm2 power density and regarding this power density, the output of voltages varies. The Rectenna / Antenna array in their studies showed a measured output voltage of .55, 1.4, and 2.13 V together with an energy conversion efficiency of 16, 28, and 36 % at 1, 5, and 10 uW/cm2, . They concluded that (DC) output properties are suitable in an urban area.
In the study of Waguaf et al. (2018) entitled "Energy Harvesting with 2.45 GHz Rectenna for urban application", they simulated electromagnetic experiments via computer software applications. They set up an emitting station and received RF harvesters at 2 meters apart and with this setup, developed an RF to DC boosting RF harvester. They concluded that the boosting antenna for RF harvester can give an output voltage of up to 140 mV for a 1uW/cm2 (-11dBm) power density. Zeng et al. (2018) presented a compact dual-band rectenna antenna rectifier RF harvester designed for urban areas via computer simulations to which measured a peak efficiency of 62% at 0.88 GHz along with 15.9 μW/cm2 power density (-1dBm) and 50% at 1.85 GHz along with 19.1 μW/cm2 power density (-10dBm). The energy harvester enabled activation of an LCD watch at 1.275 Volts output at a distance of 25 meters away from a cellular station. Caselli et al. (2019) in their study entitled "Analysis and design of an Integrated RF energy harvester for ultra-low power environments" showed that the surveyed measurements of urban and semi-urban environments got an average of -22 and -29dBm power levels with respect to RF energy source 700 and 1000 MHz band. The RF harvester through post simulations recharged a capacitor at 2 V about 950 MHz input frequency along with -18dBm input power. Khalid et al. (2020) in their study entitled "Quad Band 3D Rectenna Array for Ambient RF Energy Harvesting" surveyed a semi-urban area and found out that FM broadcast, GSM 900, GSM 1800, 3G, and Wi-Fi has potentials for RF energy. To harvest this energy, a multi-band antenna with an inclusion of a monopole antenna was made and designed via computer simulations concerning FM transmitter as a source along with the surveyed measurement results in a semi-urban area. The results about the RF harvester got a maximum efficiency of 80% at -6dBm and through field testing in an ambient environment at 300 meters reference distance from the nearest tower, the RF harvester got a conversion efficiency of 31.3% at -15dBm PRL. The study showed that RF harvester is acceptable for sensor applications. Garah et al. (2017) did a study on the calculation of the GSM path loss (908 to 957 Mhz). They assessed the area environments of Batna, Algeria. The data gathered path loss was compared to path loss measured by empirical models such as Cost 123, Hata, SUI, and Egli model. It showed that the model Cost 231 leads the Hata. SUI and Egli model via Root Mean Square Error calculation results. Onuu et al. (2017) made a comparative analysis between the Hata and Egli model via GSM path loss prediction calculations in some major towns in Akwa State. The Mean Square Error (MSE) results showed that the Egli model with the MSE value of 5.97 dB is suitable for path loss prediction in a nonuniform landscape. Fanan et al. (2017) looked into the finest amalgamation of an empirical model of data terrain for forecasting TV white space via investigating the performance of Extended-Hata, Davidson-Hata, and Egli model with the TV band ranging from 470 to 790 Mhz at 1000m, 100 m and 30 m respectively when compared to various locations in the United Kingdom. The findings showed that the Egli model manifested a good performance compared to other models. Akanni and Oliseloke (2020) scrutinized a radio signal propagation with respect to 88 Mhz frequency at different degrees of altitude. Using a digital distance meter, the LOS of several data points from the transmission source was established. The outcomes were verified and compared to LOS, Egli, and Longley-Rice Model. The results showed satisfactory acceptance of the Egli model with respect to different degrees of altitude.

SYNTHESIS
The Egli model outperforms other empirical models regarding path loss measurements in terms of accuracy. This is the reason why this formula was put into use with respect to the heuristic modeling. All authors concluded that there is a good potential for RF energy harvesting from different RF Transmitters located in different environments such as urban, semi-urban, and ambient environments. They optimized the power density/levels coming from these sources with their RF harvester designs the reason why the output of their harvester got a higher voltage for powering up an IoT and can harvest RF energy at a farther distance.
In the survey of Song et al. (2015), Adam et al. (2017), and Piñuela et al. (2013) the measured power densities differ per environment. According to the study of Pinuela et al. (2013), results vary because of proximity distance, propagation characteristics, and lineof-sight scenario. The study further showed that there is a corresponding relation between the surroundings and the RF station energy sources such as the propagation of waves. In this investigation, the behavior of wave and radio frequency propagations were assessed per multi-urban settings to classify the area along with an omnidirectional antenna. The antenna serves as an assumed device because it radiates in all directions perpendicular to an axis, which is very beneficial for a heuristic model as often used by radio network stations and base stations.
Power densities that are viable for energy harvesting can be as low as -30dBm up can be harvested from the ambient surroundings specifically, urban, semi-urban, urban high. Any frequencies that are given whether it be a TV, GSM, LTE, or any other frequency there is only one common factor which is the power density/ power level.

METHODOLOGY
This section gives a clear picture of how the study was implemented based on the fundamentals of radio propagation.

Link Budget for Power Received Level (dBm) of FM, TV, and Cell
The book of Alexander entitled "Optical Communication Receiver Design" published by SPIE Optical Engineering Press, 1997 positioned that the performance of an optical communications system is quantified using a Link Budget, which is also used in microwave links. All losses and gains in the transmitter power are all accounted for. To know the performance of the communications, the Budget Link formula according to the book of Elbert may be presented as: Power Received (dBm) = Power of the transmitter (dBm) plus all gains (dB) and minus all the losses (dB).
The formula is supported by Alexander's book on "Optical Communication Receiver Design" and Elbert's book on "Introduction to Satellite Communication" This formula can determine the RF potentials in terms of the PRL in dBm per different classified multi-settings.

Log path distance Loss in different Environments (dB) of FM, TV Cell
The difference between Received Signal Power (RSP) and Transmitted Power (TP) is called Path Loss. The received power falls as the distance between radios increases. The energy decreases based on the free Space Model according to Gutierrez in his book entitled "Selected Readings on Telecommunications and Networking" published by IGI Global Inc., 2009. Gutierrez also stated that the Free-space model is not accurate because it assumes a clear line of sight when it comes to different environments. The use of the Log distance path loss model would be the best formula to use instead of free space. harvesting techniques. They identified as one of the techniques for RF sources optimization with the power density of .01 uW / cm2 which is equal to a Power Received Level (PRL) at a minimum of -20dBm. These values were also postulated from the book of Khan and Yuce entitled "Internet of Things (IoT): Systems and Applications" published by Jenny Stanford Publishing Ltd., in 2019 and from the book of Gungor and Hancke entitled "Industrial Wireless Sensor Networks" published by CRC Press in 2013.
The framework gives a clear picture of how the study was implemented starting from the recognition of the type of multi-urban setting, determining the losses regarding the effects of the environment to propagations of waves, and the effects of radiofrequency propagation which were subtracted from the Power made by the transmitter resulting to a PRL in dBm. The PRL that can enable power harvesting systems and IoT and small electronic devices were all selected. All the selected power levels in dBm were converted into graphs and tables which can be a useful tool for having a heuristic prediction regarding RF energy potential (Figure 4).

Procedure
The goal is to classify different multiple urban settings such as Line of Sight (LOS), Rural (R), Suburban (S), Urban high (UH), Urban Very High (UVH), and Area with Obstructions or Non-line of Sight (NLOS) for ambient energy harvesting potential capabilities per wave propagations behavior of FM, TV, and Cell with their corresponding environment, and RF propagations. The procedure aims to find the distance radius range from the antenna tower that has the PRL -20dBm up to 0dBm way up to the nearest point of the antenna tower, labeled into scores with their corresponding Voltage (milli-Volts) and Power (Watts) that has good viability for energy harvesting (Figure 3 and Figure 4).

The -20dBm and up -The Power levels with the potential for RF energy harvesting and their scores
Power levels with potential viability for Energy harvesting such as -20dB to -15dBm, -15dBm to -10dBm, -10dBm to -5dBm, -5dBm to 0 dBm, and 0 dBm way up to the point to the nearest transmitter were labeled into scores as 1 (Fair with 22mV to 40mV initial voltage), 2 (Good with 400mV to 70mV initial voltage), 3 (Very Good with 70mV to 125 mV initial voltage), 4 (Satisfactory with 125 mV to 223 mV), and 5 (Excellent-Hot Zone with 223 mV and up) respectively (see Figure 5). Then the power densities are converted into their corresponding Powers (P) in Watts with the formula: and Voltages with the formula: . The results were put into a  where PRL is the Power Received Level in dBm, P is the Power in dBm, n is the path loss exponent, do is the reference distance in meters of the source transmitter, f is the frequency in Megahertz and ht is the height of the antenna in ft. With the use of MS Excel spreadsheet by inputting distance, the antenna height, power of transmitter resulting to a PRL in dBm. The PRL that ranges from -20dBm and up with their corresponding distance, antenna height, were selected and were put into tables and graphs.
By using the budget link with the assistance of the excel (Microsoft Excel TM) spreadsheet's What-if -Analysis -Goal seek function, the PRLs in dBm were computed with the following steps: (Note: by the use of Goal seek or via the Trial and Error method in which different distances were continuously inputted to the Excel formula until the results would be -20dBm and up ranges were spotted such that -20dBm to -15 dBm as 1, -15 to -10dBm as 2, -10dBm to -5dBm as 3,-5dBm to 0dBm as 4, and 0dBm up to the nearest point of the transmitter power as 5 as excellent for RF harvesting as hot zone radius to know the distance that has RF potentials) 1. Rural score from 1 to 5 of FM PRL in dBm for less than 25 Kilowatts of FM transmitter power at an antenna height of 500 ft and below. 5. Rural score from 1 to 5 of FM PRL in dBm for 50 to 75 Kilowatts of FM transmitter power at antenna height of fewer than 500 ft. 6. Rural score from 1 to 5 of FM PRL in dBm for 50 to 75 Kilowatts of FM transmitter power at antenna height of 500 ft. and up. 7. Rural score from 1 to 5 of FM PRL in dBm for 75 to 100 Kilowatts of FM transmitter power at antenna height of fewer than 500 ft. 8. Rural score from 1 to 5 FM PRL in dBm for 75 to 100 Kilowatts of FM transmitter at antenna height of 500 ft. and up. Note that with the results, the distance can be already seen with the corresponding scores such as Power in dBm, Voltages in Volts, Power in Watts for potential harvesting of An FM source located in a rural area. All of these results were put into a table and plotted on a graph. 9. Same procedures from 1 to 8 were used to the Line of sight (LOS), Suburban (S), Urban high (UH), Urban very high (UVH), and Area with obstructions multi-urban settings or Non-Line of Sight (NLOS).

Statistical Treatment
The statistical techniques used in analyzing and interpreting the results with the model were Percentage Range of Accuracy and Mean Squared Error (MSE) to measure the accuracy and predictive error for power levels used in the studies by different authors as a benchmark for energy RF energy harvesting to determine the acceptability of the model. The formula may be presented as follow (Equation 5): Representation of an MSE (see Figure 7) for measuring the predictive error of the model where MSE stands for Mean Square Error; SSE, Sum Squared of Error; n, Number

Equation 5
Equation 6 of the population (Figure 5), and in the addendum to determine if the models' scoring system can predict the RF potentials of different authors, or if the results of the authors were within the scope of the model's scoring system. Weighted mean was used with the formula (Equation 7): Where x̄ stands for the Weighted Mean; Σfx, summation of the weights times the sample; Σx, the sum of the weights with the use of the scale scoring of 5, 4, 3, 2,1 and 0, as Excellent Hot Zone (E) for RF potentials, Satisfactory (S) level for RF potentials, Very Good (VG) level for RF potentials, Good (G) level for RF potentials, Fair (F) level for RF potentials, and 0 levels for RF potentials respectively.
Standard Deviation and Mean were used to determine the model's average meters distance accuracy and measurements regarding how the number's variability in accordance to different authors' studies and works as a benchmark for the model.

The formulae were as follows (Equation 8):
Where SD stands for Standard Deviation; ∑, the sum of; x, value in the data set; µ, mean of the data Set; n, number of data points.

RESULTS
This section shows the results of the heuristic approach to classifying different multiple urban settings regarding RF harvesting capabilities such as Line of Sight (LOS), Rural (R), Suburban (S), Urban High (UH), Urban Very High (UVH), Non-Line of Sight (NLOS) in accordance to transmitter technologies like FM, TV, and Cell. Prediction of the distance radius from the transmitter for viable RF harvesting can be easily determined without tedious calculations.

Equation 8
Equation 7      Figures 8,9,10,11,12 and 13 show the RF Energy Graphical Footprints Model (GFM) of the distance and scores intensity for RF energy potentials for power and antenna heights in a classified LOS multiple urban settings. This can be used by RF installers and designers for RF harvesters.

Procedure (GFM and Table)
For example, the given Rural Multi Urban settings the farthest average distance radius for RF energy potentials from the sources such as TV, FM, and cell in classified Rural (R) multiple urban settings. A 75 Kilowatts to 100 Kilowatts of power with an antenna height ranging from 500 to 1000 ft., would give an RF energy potentials intensity score of 1 at a distance range of 1935 to 1620 meters from a TV source (See Part 1). In FM when used as an energy source, RF potentials can be found at1610 to 1350 meters with a score of 1 but has more distance coverage for energy harvesting due to the higher power and antenna height. Table 4 presents the distance of RF energy potentials of an FM Technology as a source in the classified rural area multiple urban settings. (Same procedure for Line-of-Sight, Suburban, Urban High, Urban Very High, and Non-Line-of-Sight Multi Urban Settings.

The RF Energy Graphical Footprints Model (GFM) / Researchers' scoring model as a Guide for RF Energy Harvester Installers and Designers
RF installers and designers can use the GFM as a guide for their RF harvesters, to determine the RF energy potentials, classify the area, then evaluate if RF sources and the harvester is within the radius scope of area for approximations. For example, an estimated distance range of 1620 meters away from the antenna tower of an FM station in classified rural multi-settings with a power of 75 kW to 100 KW at 500 ft. antenna tower, the researcher can distinguished the scores By simply looking at the RF energy GFM regarding viabilities for RF energy harvesting.
RF designers can also use the GFM for the design considerations of their RF harvesters because they can view the Power Received Levels (PRLs) for RF harvesting through intensity scores and temporal scores if operating hours were considered for the design.

Statistical results
The percentage acceptability range of the RF footprints graph model compared with different authors' past studies as a benchmark showed a percentage acceptability range of 84% with a mean squared prediction error of 14 along with an accuracy range of 1451 meters. In comparison with the work of different authors on RF harvesting, the grand weighted mean was a score of 3 which is within the boundaries of the model scores, reflecting a Very Good level for RF potentials. Therefore, the model RF energy GFM is acceptable.  84% acceptability. The remaining 16% is for the designing of the RF harvesters which is not within the scope of this paper. If the design of RF harvesters were to be included, the total would be 100%.

Comparison of Study with Other Works
Range of Accuracy and the Mean Squared Error computation to measure the predictive error for Power Received Levels (PRLs) used in the studies from different authors as a benchmark for RF energy harvesting to determine the acceptability of the model.

DISCUSSION
The RF Energy Graphical Footprints Model (GFM) intensity scores range from 1 to 5 in accordance with initial voltage, power, and distance. RF harvesting potential can be made heuristically with FM, TV, and cell technologies through a simple Graphical Footprints Model (GFM) per urban multiple settings. Designers of RF harvesters can see where they can put their systems and determine the PRL in dBm, power in watts, voltage in milli-Volts, and distance in meters through intensity scores. Due to attenuation for different multiple urban surroundings, the PRL for viable RF energy potentials coming from TV, cell, and FM vary based on the distances for this can be seen in the model Footprints for RF energy potentials.
Classified Line-of-Sight (LOS) multiple urban settings for RF Energy Potential categorized as transmitting source and the RF harvesting device are in clear view of each other without any obstacle LOS can be applied to rural areas, suburban areas, urban high areas, and urban very high areas if the RF harvester and the transmitter RF energy source are in clear view of each other.
The classified rural area multiple urban setting, categorized as an area that has small houses or hard to see one or two buildings and encompassing a big landscape of trees, mountains, hills, and valleys. The classified suburban multiple urban setting, categorized as a combined fraction of peopled district within a city where some structures are closely spaced and with hard-to-see buildings that may produce obstructions of signals to the RF harvester.
In a classified Rural multi-urban setting, the farthest average distance radius for RF energy potentials for RF harvesting from an FM source is 1610 meters but has a lesser distance for RF potentials if compared to the results in the classified LOS multi-urban setting. The farthest average distance that is fair for RF harvesting from an FM source in a classified Suburban multi-settings area is 1320 meters with an intensity score of 1 and the results regarding the RF potentials distances are lesser than to rural and LOS classified multi-settings.
RF designers and installers can harvest RF energy from an FM Transmitter source in a classified Urban High multi-setting at 1110 meters range with an intensity score of 1, but due to the area/environment path loss exponent, the results have a lesser distance compared to Rural, Suburban, Line of Sight Multi-settings. In a classified Urban Very High multi-setting, RF designers and installers can harvest RF energy as far as 950 meters from an FM RF energy source with an intensity score of 1. The distance of Urban Very High has a much lesser radius range than at Urban High due to the loss exponent environment.
The farthest average distance radius for RF energy potentials from an FM source in classified NLOS multi-settings is 610 meters with an intensity score of 1. The distance for RF potentials of an NLOS area has the lowest radius range because of the environment. RF designers as much as possible must take a deep precaution regarding designs in this classified area if they are able to extend their RF harvester antenna to make it as a Line of sight with the source is highly recommended.

CONCLUSIONS AND RECOMMENDATIONS
The RF Energy GFM has an acceptability percentage of 84% with a Mean Squared Prediction Error of 14 for the RF Energy Footprints Graph model. The model for Weighted Mean results perceived coming from the different authors regarding their RF harvesters received a score of 3 which means Very Good level for RF potentials. Therefore, the model is acceptable as well as the average range distance of the RF Energy Graph Model of1451 meters. Since the model is part of RF harvesting as a preliminary tool, the model will give information to the designers and installers regarding RF energy potentials before going to the area.
A classified Rural, Suburban, Urban High, and Urban Very High can turn into a classified NLOS due to mountains, hills, trees, electric posts, and building obstructions. Simple height adjustments regarding RF harvesters will do to make it into a classified LOS for a much better RF harvesting the same thing must be done with the classified NLOS area.
The model, along with the studies of different authors shows that Power density / received levels ranging from -20dBm and up, have the potential for RF energy and RF harvesting with an average percent accuracy of 84% with 14 Mean Squared Error and with Mean average accuracy for a distance of 1451 meters.
The RF Energy Graphical Footprints Model (GFM) for energy potential capabilities is a plot of Power Received Level thru scores based on a -20 dBm and up with corresponding scores, initial voltages, Power densities, and Power in accordance to classified multi-settings that are viable for energy capabilities can be determined without any computations and the only effort required is the reading and following of values of the Energy GFM.
The RF Energy GFM serves as a tool for easy preliminary assessments of selected urban settings for RF energy harvesting in place of too many communications engineering computations to get the Power Received Level. The RF GFM will assist the RF designers to determine the best locations for their harvesters even in an off scheme compared to tedious surveying on-site to get the Power Received Level. The GFM will act as a map to show the best place for RF harvesting capabilities which can be used for design considerations.

RECOMMENDATION FOR FUTURE WORK
The RF GFM is one major contribution to ambient RF energy scavenging technology. RF harvester designers may use the model for their design considerations because it entails easy preliminary assessments of the selected urban setting for RF energy harvesting capabilities and serves as a map to identify the best locations for harvesting.
RF Energy GFM can be programmed into an application software by inputting needed parameters to determine the score of certain urban settings or by taking pictures through an android mobile phone in a 360° of a given map with different transmitters area and it will give already a corresponding score and succeeding researchers may improve the model by having a much smaller Mean Squared Error at below 14.
The future paper works include a study on RF energy GFM with wifi sources with the inclusion of prototype RF harvesters to be used in accordance with the RF Energy Potentials Graph model.

IMPLICATIONS
The RF Energy GFM serves as a tool for easy preliminary assessments of selected urban settings for RF energy harvesting in place of too many communications engineering computations to get the Power Received Level.
The GFM will act as a map to show the best place for RF harvesting capabilities which can be used for design considerations even in an off scheme compared to tedious surveying on-site to get the Power Received Level and the only effort required is the reading and following of values of the Energy GFM.