Electromagnetic Wave Radar System Design For Water Detection
This paper presents a design for an Electromagnetic Wave Radar system as moisture detector using NI USRP 2932. The proposed system was to capture reflected signal for varying moisture from 0%, 20%, 40% up to 100% for different Soil types. The proposed design has been simulated in LabView along with the Horn antenna and a laptop to get the A-scan and B-scan for the two set-ups. A significant data from reflected signal was observed showing the effect of moisture contents on different soil types. Furthermore, reflection coefficient and dielectric coefficient relationships with moisture content were established and modeled.
Introduction
Detection radar system helps in detecting underground materials like water, loam soil, sand and ridge gravel by using electromagnetic energy pulse like radio wave which determine an object’s characteristic properties. The dielectric constant of each material will be used to characterize the reflected signal. Also, reflection coefficient of the reflected signal is also popular technique in modeling the underground soil layer. One of the most promising technologies that could detect and identify buried and hidden materials is the Ground Penetrating Radar (GPR). GPR has the ability to detect both metallic and non-metallic objects in the sub-surface. Ground Penetrating Radar is the cognomen for the family of radar systems that gives image beneath the ground. In this study, the researchers designed a system similar to GPR called the Electromagnetic Wave Radar (EWR). The project EWR uses the National Instruments – Universal Software Radio Peripheral 2932 (NI-USRP 2932) with a working HORN antenna that can detect moisture from underground surface materials such as soil, sand and rock. Each material was evaluated based on its moisture content ranging from 0%, 20%, 40%, up to 100%. To establish the effect of moisture contents on underground soil layers reflection coefficient and dielectric coefficient were measured. The results of the measurements are characterized and modeled through A-scan and B-scan.
System design
Electromagnetic Wave Radar System design
An EW radar system is primarily composed of a transmitter, a receiver and an antenna as shown in Figure 1. The transmitter sends signal to the ground from the antenna. Input signal is mixed with a RF carrier frequency from the oscillator circuit. The convolution of two signals resulted to a modulated signal that is further amplified by the power amplifier, through the antenna, and transmitted into the soil from a distance. The reflected signal from antenna is demodulated by passing into a low noise amplifier and mixer. The transmitted signal is then reflected by the different type of materials found in the ground. Each material has a varying dielectric coefficient, so each material reflects a unique signal. The reflected signal is received by the receiver and the signal is then shown in the receiver front panel.
SDR System
SDR is become popular platform for Radio Frequency (RF) system in the field of research. One of the reasons it is preferred is because of its flexibility on being implemented on different operations and expensive and bulky hardware components implementations are replaced by software.
The EW system composes of an antenna and the NI-USRP 2932. The NI-USRP 2932 acts as SDR responsible for converting physical radio waves into processed digital data as shown in Fig. 2. The SDR is implemented as a combination of RF circuits, specialized digital circuits, and software. It has onboard Digital Upconverter (DUC), Digital Down Converter, Analog to Digital Converter (ADC), and Digital to Analog Converter; all of which are required for the signal to be configured on the Host PC before transmission, and for the received signal to be analyzed on the Host PC.
Signal Transmission Theory and Principles
The basic concept of this design is just simple communication systems as shown in Figure 3. To model wave propagation of soil response is similar to transmission lines at single-frequency time harmonic signals under steady-state conditions. When transmission line encountered different characteristic impedances, the incident wave is partly reflected back toward the source and partly transmitted across the junction of other line.
In Figure 3, the vertical depth of a single point located at the boundary of two materials having different dielectric constants can be calculated with the aid of the electromagnetic wave propagation velocity equation as: where c is the propagation velocity in free space and εr1 is relative permittivity of the first material. The transient response behavior resulted to multiple reflections (echoes) because voltage pulse travel back and forth between the sending and receiving ends of the medium and it usually track using bounce diagram. The presence of incident and reflected waves in soil gives rise to a standing wave pattern. The transmission line standing wave ratio (SWR) in medium is defined as: The mismatch generates a reflected wave having reflection coefficient of far end typically soil layered (Γ=Γl ) and near field typically transmitted output signal (Γ=Γg). The reflection coefficient (Γ) shows the relationship of the medium n2, n1 having dielectric coefficient of ε1 and ε2 respectively. This relationship could be used as the basis that reflection coefficient and dielectric constants are related.
Prototype set-up of soil layered
To test different soil type, a prototype of soil layers was constructed inside MSU-IIT campus.
Software implementation
Figure 5 shows signal generation where the transmitter sends signal to the ground. Here, the input signal is not a sine wave since it consists of many harmonics. Instead, the researchers use a pulse width for this radar system, a digital signal that indicates one or zero.
Figure 6 shows the error block diagram of the EWR circuit. The result of this block diagram is shown in Figure 3.1.2.1a. In this system, three hardware components such as laptop, USRP device and antenna are used for displaying errors in connection, transmission and reception of signals. The receiver signal generator is shown in figure 7. It displays the parameters needed to be seen by the user. The front panel output for the TX and RX are shown in figure 8. It shows graph of the TX and RX output signals, along with the baseband power spectrum.
Results and discussions
Designed EW system using SDR technology
After designing the radar system in LabView, the researchers test its workability using the NI-USRP 2932 and the Horn antenna. A rectangular container for the loam soil with 0% moisture was used as basis for the material testing.
To test the optimum performance of the system, antenna height from soil surface is tested. The results from Figure 9 have five colored raw data representing various heights between the soil and the antenna. Test set-up 1,2,3,4 and 5 are sets such that the antenna is located each at 2, 4, 6, 8 and 10 inches from the soil surface. The graph shows that increasing the antenna height also increases the amplitude. Among the trials, trial one has the best desired output. The first trial produced the best output since it captured medium amplitude and was not distorted. This method confirms that the system is working properly. Thus, materials to be tested will use a 2-inch antenna height to get the best results.
Test of different soil by varying the soil moisture content
Each weights of different soil type were gathered and used as basis for getting the moisture content. The percent moisture content is calculated as the weight by parts of the material over the total weight of the material in water. Also, using a 2-inch antenna above each material is properly observed.
Characterizing each material with varying moisture content is performed by establishing its relationship with reflection coefficient and dielectric constants. Figures 9 and Figure 10 show plots of the measured reflection coefficient and dielectric constants respectively versus the moisture content θv. To show the dependence of transition moisture on soil types more clearly, smooth curves were drawn through the measured data points. For each soil sample, ten repeated measurements were made and average values of each test were plotted at given moisture content.
As observe for Soil A, as moisture content (θv) is increase reflection coefficient decreases, however, as moisture content is further increase at θv=0.7, the system doesn’t follow the trend where system could not predict its behavior. Soil A is a sandy soil type, in which it has larger loosely packed particles resulted to a poor holding capacity. Soil B has nonlinear effect behavior, when moisture content (θv) is increased reflection coefficient decreases. This type of soil is dominated by sand and a little silt and clay, which makes relationship variance negligible.
Similar to Soil B, Soil C has nonlinear behavior also. However, this kind of soil has greater indirect relationship. Soil C has highest water holding capacity because of inherited good aggregation that provides a large number of pores that hold water against gravity. On other hand Soil D as moisture content (θv) is increased reflection coefficient decreases, however, as moisture content is further increased to θv=0.8, moisture content reached its saturated condition. This type of soil has tiny pores that can withstand volume of water against gravity; however, increasing moisture content will makes soil elastic. The reflected signal is not clearly read when it reached its saturation point, which makes the system difficult to predict soil behavior. As observed in the figure, the signal strength is higher under wetter conditions and lower under drier soil conditions. These results agree with the theory of reflection coefficient that higher soil moisture values have higher reflection coefficient and therefore will reflect more electromagnetic wave energy.
The figure shows that at low moisture content, ε’ is increased slowly with moisture content. After reaching a breakpoint moisture value, ε’ increase steeply with moisture content. The breakpoint moisture is dependent on the soil types and usually occurs at larger moisture values for clayey soils (Soil D and Soil C). It can be clearly seen that the transition moistures of Soil A of low field capacity (FC) samples occurred at θv=0.15. On the other hand, high FC samples (Soil C and Soil D) show large transition moistures beyond θv=0.05. Moreover, as observed in Figure, soil moisture starting at θv=0.45 is difficult to predict dielectric constant, same as explained by Topp et al. (1980).
Modeling the A-scan and B-scan
Section B presents a uniform soil type to establish the relationship of the soil moisture. In this section, the effects of dielectric materials is observe by observing its reflected signal strengths. Figure 12 presents a bouncing wave-like pattern from received signal called a-scan, reflected from layered soil. Air has a dielectric value of 1 which indicated a small spike. Loam soil has a dielectric value of 10, and the ripple in this region is due to presence of medium-sized rocks, large particle clays and earthworms. Sand has a dielectric value of 12, and the ripple in this region indicated presence of bigger-sized rocks, and broken pieces of glass. Water has the highest peak which signifies the highest dielectric value of 80. Lastly, the ridge gravel with a value of 14 created higher voltage amplitude due to its presence of unabsorbed water. Other factors that caused spikes were the presence of glass in the boundaries, which has a dielectric value of 2.9, and scotch tape, plastic cover, and air gap as used for avoiding water spillage. Figure 13 shows the sample B-scan. B-scan is derived from a series of A-scan, measured at a uniform distance. The EWR system detected different layers of loam soil, sand, water and ridge gravel. These layers correspond to different shades of grey, in which darker shades indicate higher amplitude. With water having a dielectric value of 80, ridge gravel as 14, sand as12, and loam soil as 10, the dielectric constants confirms its directly proportionality to voltage. As the transmitted signal was released from antenna, it passed through the air and projected a lighter shade. All the layers have spikes in the region due to presence of impurities. Water, having the highest peak, signifies the highest dielectric value, and is shown as the darkest shade. The ridge gravel created higher voltage amplitude due to its presence of un absorbed water, thus having darker shade compared to the other materials.
Conclusion
The designed EWR system was simulated in LabVIEW and was successfully verified using NIUSRP-2932. It reflected a transmitted signal according to type of soil. The antenna at 2-inch away from the surface produced the best output. The data gathered by the researchers coincided with the projected results based on material dielectric property and reflection coefficient. Different moisture contents of each soil type were characterized and estimated model is successfully established. Furthermore, soil layered was plotted showing A-scan and B-scan of soil materials. The result shows that when different material was arranged in layers with the presence of water, water is visible. Using the EWR system, detection of water is possible in B-scan since water has the highest value as of dielectric constant.
