This action might not be possible to undo. Are you sure you want to continue?
Although it is one of the simpler equations, ironically, it is an equation that few radar analysts understand and many radar analysts misuse. The problem lies not with the equation itself but with the various terms that make-up the equation. It is my belief that if one really understands the radar range equation one will have a very solid foundation in the fundamentals of radar theory. Because of the difficulties associated with using and understanding the radar range equation we will devote considerable class time to it and to the things it impacts, like detection theory, matched filters and the ambiguity function. One form of the basic radar range equation is
PS PT GT GR λ 2σ = PN ( 4π ) 3 R 4kT0 BFn L
SNR is termed the signal-to-noise ratio and has the units of
watts/watt, or w/w.
PS is the signal power at some point in the radar receiver – usually
at the output of the matched filter or the signal processor. It has the units of watts (w).
PN is the noise power at the same point that PS is specified and has
the units of watts.
PT is termed the peak transmit power and is the average power when the radar is transmitting a signal. PT can be specified at the
output of the transmitter or at some other point like the output of the antenna feed. It has the units of watts
• • • • •
GT is the power gain of the transmit antenna and has the units of
GR is the power gain of the receive antenna and has the units of w/w. Usually, GT = GR for monostatic radars.
λ is the radar wavelength (see (21) of the Radar Basics section) and had the units of meters (m).
square meters or m2.
σ is the target radar cross-section or RCS and has the units of
R is the range from the radar to the target and has the units of meters.
©2005 M. C. Budge, Jr
we assume that the radar is in free space.• • • 23 o k is Boltzman’s constant and is equal to 1. Jr 2 . L accounts for a multitude of factors that degrade radar performance. The reason we term this power the peak transmit power is that we will later want to consider the transmit power averaged over many pulses. often. or the signal power component and follow this by a derivation PN . The average power in the signal over the duration of the pulse is termed the peak transmit power and is denoted as PT . o o We take T0 = 293 K −21 and usually use the approximation kT0 = 4 × 10 w/Hz . see Figure 1. • • Fn is the radar noise figure and is dimensionless. the environment in which the radar operates. ©2005 M. the ignorance of the radar analyst. These factors include those related to the radar itself. or has the units of w/w. We can account for the effects of the atmosphere at a later date. C. For now. It accounts for losses that apply to the signal and not the noise. B is the effective noise bandwidth of the radar and has the units of Hz. Budge. We assume that the transmitter generates a single. the radar operators and. We will spend the next several pages deriving the radar range equation and attempting to carefully explain its various terms and their origins. T0 denotes room temperature in Kelvins ( K) . or the noise component. I emphasized the word effective because this point is extremely important and seldom understood by radar analysts.38 × 10 w ( Hz K ) . f c . DERIVATION OF PS We will start at the transmitter output and go through the waveguide and antenna and out into space. A sketch of the pulse (the terminology we use) is contained in Figure 2. We will start by deriving PS . In the process we will present other forms of the radar range equation that are used in different applications. L has the units of w/w. L is a term included to account for all losses that must be considered when using the radar range equation. rectangular pulse (a standard assumption) at some carrier frequency.
We assume that the feed and the antenna are ideal and thus introduce no additional losses to the radiated power. In actuality. The only feature of the waveguide that is of interest in the radar range equation is the fact that it is a lossy device that attenuates the signal. the power radiated into space is Prad . the losses are incorporated in Lt and in other cases they are incorporated in the antenna gain. Since the waveguide is a lossy device we characterize it in terms of its loss which we denote as Lt and term transmit loss. Budge. the antenna assembly (antenna and feed) will have losses associated with it. In some instances. in a practical radar there are several devices between the transmitter and antenna feed. With this. ©2005 M. which will be discussed shortly. When using the radar range equation. the power at the feed is Prad = PT w Lt (2) and is termed the radiated power. C. We lump all of these into a conceptual waveguide. Although we only refer to the “waveguide” here. Jr 3 . Since Lt is a loss it is greater than unity.Figure 1 – Transmit Section of a Radar Figure 2 – Depiction of a Transmit Pulse The waveguide of Figure 1 carries the signal from the transmitter to the antenna feed. one must be sure that the antenna losses are accounted for in one place or the other. Because of the above assumption.
we will account for the fact that the focusing isn’t perfect by a scaling term. C.The purpose of the radar antenna is to concentrate or focus the radiated power in a small angular sector of space. as indicated in Figure 3. the radar antenna works much as the reflector in a flashlight. We will then include a scaling factor to account for the fact that the area of the rectangle is greater than the area of the ellipse. a radar antenna doesn’t perfectly focus the beam. Figure 3 – Radiation Sphere with Antenna Beam ©2005 M. Finding the area of the ellipse of Figure 3 is not easy. Abeam . Budge. In this fashion. the power density over Abeam is SR = Prad P L = T t w/m 2 . As with a flashlight. we assume that all of the radiated power is concentrated in an area. To get around this problem we find the area of the rectangle that contains the elliptical beam of Figure 3. Later. Jr 4 . However. Abeam Abeam (3) To carry (3) to the next step we need an equation for Abeam . This scaling factor will also account for losses in the feed and antenna. With the above. Therefore. for now we will assume it does.
and the area of the rectangle is Arect = R 2θ AθB m2 . that we call the transmit antenna gain as GT = 4π w/w K Aθ Aθ B GT PT w/m 2 . it is a mathematical and conceptual concept that we often use in radar theory. GT . ©2005 M. The power at the output of the antenna is PT Lt . or ERP. We note that an isotropic radiator cannot exist in the “real world”. To do so we ask the question: What power would we need at the feed of an isotropic radiator to get a power density of S R at all points on a sphere of radius R ? An isotropic radiator is an antenna that does not focus energy. If we substitute (5) into (3) we get SR = PT Lt w/m 2 . K A R 2θ Aθ B (6) At this point we define a term. If we denote the effective radiated power as Peff and realize that the surface area of a sphere of radius R is 4π R 2 we can write the power density on the surface of the sphere as SR = Peff 4π R 2 w/m 2 . IT IS NOT.The length of the two sides of a rectangle that contains the ellipse of Figure 3 are Rθ A and Rθ B . However. Budge. All the antenna does if focus this power over a relatively small angular sector. We can think of it as a point source radiator. 4π R 2 Lt (7) and use it to rewrite (6) as SR = (8) We now want to discuss a quantity termed effective radiated power. From this we write Abeam as (4) Abeam = K A R 2θ AθB m2 (5) where K A is the aforementioned scaling factor. C. Many radar analysts think that the power at the output of a radar antenna is the ERP. Jr 5 . (9) If we equate (8) and (9) we obtain Peff = PT GT w = ERP Lt (10) as the effective radiated power.
Another note is that the development above makes the tacit assumption that the antenna is pointed exactly at the target. We define the beamwidth of an antenna as the distance between the 3-dB points1 of Figure 4. To visualize the concept of beamwidth we consider Figure 4 which is a plot of GT ( θ . The 3-dB points are the angles where GT ( θ . φ ) vs. 1. GT must be modified to account for this. φ ) is a means of saying that the antenna gain is a function of where the target is located relative to where the antenna is pointing. If the antenna is not pointed at the target. Specifically: • It accounts for the fact that the beam area is an ellipse rather than a rectangle.65θ Aθ B (11) In (11) the quantities θ A and θ B are termed the antenna beamwidths and have the units of radians. Some of it will “spill” out of the beam into what we term the antenna sidelobes. θ for φ = 0 . φ ) is 3 dB below its maximum value. or dB. We next want to address the factor K A in (7). The unit of measure on the vertical axis is decibels. C. In many applications. ©2005 M. With some thought. the maximum value of GT ( θ . Jr 6 .000 w/w o o θ Aθ B (12) where the two beamwidths in the denominator are in degrees. We do this by means of an antenna pattern which is a function that gives the value of GT at all possible angles of the target relative to where the antenna is pointing. φ ) is the antenna gain. With this we can write the antenna gain as GT = 4π w/w . • • It has been my experience that a good value for K A is 1. In this case we write the gain as GT = 25.65. you will realize that two angles are needed to specify any point on the sphere discussed earlier. It accounts for the fact that the antenna causes ohmic power losses. K A accounts for the properties of the antenna. With this we find that the antenna represented in Figure 4 has a 1 The concept of 3-dB points should be familiar from control and signal processing theory in that it is the standard measure used to characterize bandwidth. The expression GT ( θ . and is the common unit of measure for GT in radar applications. or GT . As a side note. θ A and θ B are specified in degrees. Budge. It accounts for the fact that not all of the power is concentrated in the antenna beam.
φ for θ = 0 and find distance between the 3-dB points was 2.5 degrees. C. We might call this θ A . If we have a device wherein the power into the device is Pin (in watts) and the power out of the device is Pout . We say that the gain of the device. in dB is ©2005 M. The standard use of dB that comes from control and signal processing theory relates to the gain or loss of a device.5 o degrees in the φ direction. We would then call this θ B . Jr 7 .000 = 5000 w/w or 37 dB . We would then say that the beamwidth was 2. 2 × 2. Suppose we were to plot GT ( θ . We would compute the antenna gain as GT = 25.o beamwidth of 2 degrees in the θ direction. φ ) vs. We want to digress to discuss this further. Budge.5 (13) BW 3-dB Figure 4 – Sample Antenna Pattern A TANGENT TO DISCUSS dB In the above paragraph we introduced the concept of dB.
Specifically. or dB relative to a square meter. Pin (14) If we were to relate input and output voltages or current. we often wish to express the power in the units of dB. or dB relative to a meter (note the potential for confusion in the double use of dBm). Jr 8 . we discuss the power relative to a power of 1 watt. 1 mw (18) We also want to express area measures and distance measures in their dB equivalents. We can write the loss through the device. Specifically we write P in w P in dBw = 10log . and the units for distance would be dBm. in dB. 1w (17) We also often use the term dBm or dB relative to a milliwatt (or mw). it would likely be incorrect to use dBKm (dB relative to a kilometer) and dBm in the same equation. As with standard dimensional analysis. The appropriate equations are A in m 2 A in dBsm = 10log 2 1m for area and (19) l in m l in dBm = 10log 1m (20) for length. the gain through the device would be V I G = 20log out = 20log out Vin Iin . Budge. Pout It should be obvious that L = −G . C. the units in dB must be consistent. To do so. (15) It should be noted that a tacit assumption of (15) is that the input and output impedances of the device are the same. Since we are referencing the power to 1 watt we use the units of dBw. as P L = 10log in . (16) When working with the radar range equation we often have need to express the various parameters in their dB equivalents. The units for area would be dBsm. be careful to properly combine units of measure in dB. ©2005 M. As a caution. For example. The equation for it would be P in w P in dBm = 10log = P in dBw + 30 .P G = 10log out .
this process is very complicated and beyond the scope of this course. Jr 9 . Again. That is. we can say that the power at the output of the antenna feed is Pant = Srec Ae (24) where Ae is the effective area of the antenna. In fact. Srec = ( 4π ) PT GT σ 2 R 4 Lt . A more common term for Ae is effective aperture of the antenna. The process of capturing and re-radiating the power is very complicated and the subject of much research. the power density at the radar is Srec = Ptgt 4π R 2 . C. It will be noted from this figure that Ptgt can vary by about 25 dB depending upon the orientation of the aircraft relative to the radar. Thus far we have an equation for S R . it is an area measure that describes the ability of the antenna to capture the returned electromagnetic energy and convert it into usable power. We will discuss it further shortly. Ptgt Given the above assumption that the power radiated by the target is and that it acts as an isotropic radiator. the radar antenna captures part of it and sends it to the radar receiver.15 of the text. Budge. if we were to multiply S R by an area we would convert it to a power. aperture ©2005 M. As the electromagnetic wave passes by the target. substituting (8) into (21) and the result into (22). Therefore. To get an ideal of the variation of Ptgt refer to Figure 2. We note that S R has the units of w/m2. With this we say that the effective radiated power of the target is Ptgt = σ SR w . According to the dictionary. This is what we do with RCS. which we denote by σ and ascribe the units of m2. (23) As the electromagnetic wave from the target passes the radar. the power density in the location of the target. (22) Or. If we follow the logic we used for the target. the target is much like an antenna and radiates the power with different amplitudes in different directions. (21) We term Ptgt an effective radiated power because we assume it is radiated uniformly in all directions. some of the power in it is captured by the target and reradiated back toward the radar. we represent the target as an isotropic radiator.BACK TO THE DERIVATION OF PS Now back to our derivation. For now we simplify the process by using the concept of radar cross-section or RCS.
we want to turn our attention to the noise term. λ2 PT Gt GR λ 2σ (27) If we substitute (27) into (26) we get Pant = ( 4π ) 3 R 4 Lt . (26) It turns out that this equation is not usually very easy to use because of the Ae term. In fact. For now. With this we say that the signal power in the radar is given by PS = PT Gt GR λ 2σ ( 4π ) 3 R4 L (29) which is Pant with the additional losses added in. signal processor. We make this clarification of area because we don’t want to confuse it with the actual surface area of the antenna. According to antenna theory we can relate antenna gain to effective aperture by the equation GR = 4π Ae . etc. we can think of the antenna as an orifice that funnels energy into the radar. They would also include any losses that we want to associate with the receiver. Jr 10 . It turns out that the effective aperture is related to the physical area of the antenna. For now we will lump all of these losses with Lt and denote them by L .means opening or orifice. A more convenient method of characterizing the antenna would be through the use of its gain. For now we comment that they would include the losses associated with propagation through the atmosphere. In the above paragraph we said that PS is the signal power “in the radar”. Budge. We will save this discussion until later. we need to account for losses that we have ignored thus far. PN . Thus. we will discuss losses in more detail later. However. If we substitute (23) into (24) we get Pant = ( 4π ) PT GT σ Ae 2 R 4 Lt . C. displays. as we did on transmit. we didn’t say where in the radar. human operator. ©2005 M. That is Ae = ρ Aant (25) where Aant is the area of the antenna projected onto a plane placed directly in front of the antenna. in this context. (28) As a final step in this part of the development. which is often a paraboloid.
(32) −21 It is interesting to note that kT0 = 4 × 10 w/Hz .16 ºK) or about 62 ºF which. Jr 11 . is white. Because of this. With this approach we start by assuming that the noise in the radar.84 ºC (0 ºC = 273. C. Although this may be confusing at present (it says that we need to know N 0 to compute Teff which we need to compute N 0 !). which makes one think that o the value of T0 = 290 K was chosen to make kT0 a “nice” number and not because it is room temperature. In electronic devices the main source of noise is termed thermal noise and is due to agitation of electrons caused by heat. we will use the approach commonly used in radar and communication theory. must operate in the presence of noise.38054 × 10 w/ Hz × K is Boltzman’s constant. ©2005 M. it is a temperature quantity that we use to compute the proper noise power spectral density in the radar. In fact. it will hopefully become clearer when we undertake a more detailed discussion of noise. The heat can be caused by the environment (the sun. Budge. to derive this power. before we need to represent it as power. which are the same in this context. Eventually.DERIVATION OF PN All radars. humans. (We can’t use power because white noise has infinite power. It turns out that Teff is not an actual temperature. etc. Rather. However. is room temperature. Teff is the (30) ( ) effective noise temperature of the radar in degrees Kelvin (ºK). In most radars the predominant source of heat is the electronic equipment.) We define the noise power spectral density in the radar by the equation N 0 = kTeff w/Hz or w-s or joule −23 o where k = 1. and provide a means of characterizing the effects of the receiver electronics. as with all electronic equipment. or energy. we start by characterizing the noise in terms of its power spectral density. T0 = 290 K or 16. With the above we get N 0 = kT0 Fn . the room.) and by the electronic equipment itself. As an alternate formulation we also write Teff = FnT0 where Fn is termed the noise figure of the radar and T0 is a reference (31) o temperature normally referred to as “room temperature”. by some standards. the earth. we want to characterize noise in terms of its power at some point in the receiver.
we need to multiply it by a frequency term to convert it to a power.Since N 0 has the units of w/Hz. and zero at all other times. B is the bandwidth of the matched filter. The energy in the noise is given by (32) and is (35) E N = N 0 = kT0 Fn joule . In fact. the pulse width. This means that the energy in the signal is ES = PSτ p w-s or joule . you will recall. will take the same form as (1) and (34). It will be left as a homework problem to determine if B is the 3-dB bandwidth of the matched filter. A very common mistake in the use of the radar range equation is to use the transmit waveform bandwidth for B . and we are trying to represent the power at the output of this matched filter then. Recall that (29) is the signal power in the radar (again. This means that at the point we measure the signal it has a power (peak power) of PS for a duration of τ p . Budge. AN ENERGY APPROACH TO SNR In this approach to SNR we define the SNR as the ratio of the signal energy to the noise energy (which. rectangular pulse (as in Figure 1). For modern. we use this to write the noise power in the radar as PN = kT0 Fn B (33) where we term B the effective noise bandwidth of the radar. in terms of the radar range equation. and if the receiver employs a filter that is matched to the transmit pulse. It will be noted that I placed a lot of caveats on our ability to tie B to a specific bandwidth. For now. we won’t say where yet). It turns out that if the radar transmits a single. I did this to emphasize that we must be very careful in how we define the noise in a radar. we want to develop an alternate formulation for SNR which. With this we determine that the SNR is (36) ©2005 M. In fact. is the power spectral density). We will do this after we discuss some other topics. We want to emphasize the term effective. B may not be the actual bandwidth of any component of the radar. with one relation. We further assume that the shape of the originally transmitted pulse is preserved. C. pulse-compression radars this is incorrect! If we combine (33) and (29) with the relation SNR = PS PN we get (1) or SNR = PS PT GT GR λ 2σ = w/w PN ( 4π ) 3 R 4kT0 BFn L (34) What we have not done in (1) and (34) is state where in the radar we are characterizing the SNR. Jr 12 .
If we use (34). which are given above. λ and the total losses. C. The parameters that we will need to compute are the wavelength. PT Transmit Losses. The appropriate parameters are given in Table 2 in “dB units” and MKS units. E N kT0 Fn ( 4π ) R kT0 Fn L w-s (37) We note that (34) and (37) are the same equation if we let B =1 τp . Table 2 – Radar Range Equation Parameters RADAR RANGE EQUATION PARAMETER PT ©2005 M. (38) provides us with the definition of effective noise bandwidth as the reciprocal of the transmitted pulse width. LR Noise Figure Fn Other Losses. which we will. τ p Antenna Gain.SNR = Pτ ES P G G λ 2σ w-s = S p = T 3T 4R τ p joule/joule or or w/w . or can be derived from the parameters of Table 1. Lt Pulse Width. EXAMPLE At this point it will be instructive to consider a couple of examples.4 µs 38 dB 8 GHz 3 dB 8 dB 2 dB For the first example we wish to compute the SNR on a 6-dBsm target at a range of 60 Km. GT . Budge. GR Operating Frequency. B . we also need to compute the effective noise bandwidth. (38) In fact. Jr VALUE (MKS) 106 w VALUE (dB) 60 dBw 13 . For the examples we consider a monostatic radar with the parameters indicated in Table 1. Lother VALUE 1 Mw 2 dB 0. To perform the computation we need to find the parameters in the radar range equation ((34) or (37)) and be sure that they are in consistent units. f c Receive Losses. Most of the parameters are in Table 1. Table 1 – Radar Parameters RADAR PARAMETER Peak Transmit Power @ Power Tube. The two remaining parameters are the target range and the target RCS.
38 dB or 27. we need to solve the radar range equation for R .5×106 Hz 6. This means that if the SNR is a certain value at a given range. Doing so by using (34) as the starting point yields ©2005 M.26 dB(m) 6 dBsm 47. Jr 14 . Budge.38 dB As a double check.41 w/w or 14. The upshot of this discussion is that we define the detection range as the range at which we achieve a certain SNR. One of the important uses of the radar range equation is in the determination of detection range.6 w/w 6309.0375) ) ( 3.98) = ( 4π ) ( ( 60 ×10 ) ) ( 4 ×10 ) ( 2.01) 6 2 3 3 4 −21 6 ( 4π ) PT GT GR λ 2σ 3 R 4 kT0 BFn L (39) = 27. we note that SNR varies inversely with the fourth power of range. Substituting yields (40) SNR = ( 60 + 38 + 38 + 2 ( −14.6 ) ( 6309.6 w/w 0. The criterion for detecting a target is that the SNR be above some threshold value. we compute (34) using the dB values.0375 m 3.GT GR 6309.6 ) ( ( 0. C.98 m2 60×103 m 4×10-21 w-s 2. it will be greater than that value at shorter ranges.78) + ( −204 ) + 64 + 8 + 7 ) (41) which agrees with (39) (except for the last digit of the MKS value).78 dB(m) -204 dB(w-s) 64 dB(Hz) 8 dB 7 dB λ = c fc σ R kT0 B =1 τp Fn L = Lt Lr Lother If we substitute the MKS values from Table 2 into (34) we get SNR = ( 10 ) ( 6309.31 w/w 5. If we consider the above radar range equations.01 w/w 38 dB 38 dB -14.26 ) + 6 ) = 14.42 w/w − ( 32. In order to find detection range.31) ( 5.5 ×10 ) ( 6. or the maximum range at which a target has a high probability of being detected by the radar.98 + 4 ( 47. This gives SNR = ( PT + GT + GR + 2λ + σ ) − ( 30log ( 4π ) + 4 R + kT0 + B + Fn + L ) where all of the quantities are the “dB units” from Table 2.
Budge.98) = ( 4π ) 3 ( 19. PT GT GR λ 2σ R= 3 ( 4π ) ( SNR ) kT BF L 0 n 14 m. Jr 15 . (42) As an example.2 Using the Table 2 values in (42) yields PT GT GR λ 2σ R= ( 4π ) 3 ( SNR ) kT BF L 0 n 14 ( 106 ) ( 6309. suppose we want the range at which the SNR on a 6dBsm target is 13 dB. ©2005 M.5 on an aircraft type of target. we will show that a SNR threshold of 13 dB yields a detection probability of 0.0375) 2 ( 3.01) =64957 m or 65 Km ( ) ( )( ) (43) 2 The value of 13 dB is a standard detection threshold.95) ( 4 × 10−21 ) ( 2.5 × 106 ) ( 6. C. Later.57 ) 2 ( 0.31) ( 5.
This action might not be possible to undo. Are you sure you want to continue?