Working Through Timothy Pratt's Satellite Communications Problems
I keep running into people asking for the Timothy Pratt solution set online. There is a textbook by Timothy Pratt titled "Satellite Communications" that has been around since the late 90s, and it covers link budget calculations, satellite geometry, power and noise analysis, modulation schemes, and orbital mechanics. Students and engineers working through those problems often search for worked solutions because the book does not include answers in the back. The standard approach to solving these problems is methodical. You start with the fundamental equation for received power, which is P_r = P_t + G_t + G_r - L_p - L_a, where you account for free space path loss separately. The path loss alone is 20 log10(d) + 20 log10(f) + 92.45 when distance is in kilometers and frequency is in gigahertz. That constant 92.45 catches people out constantly. If you are working in miles instead of kilometers, that constant shifts to 36.6, and I have seen entire link budgets go sideways because someone mixed units without catching it. The noise temperature calculation is where most students stumble. You need the antenna noise temperature, the receiver noise temperature, and the cascaded noise figure. The system noise temperature T_sys = T_ant + T_receiver, but T_ant itself changes depending on elevation angle. At low elevation angles, the antenna sees more atmospheric noise and ground spillover, which can add 50 to 150 kelvins compared to a zenith observation. Pratt's problems usually assume 90 degree elevation for simplicity, but in real deployments this matters a lot.
One problem that trips people up repeatedly involves the G/T ratio and how it interacts with effective isotropic radiated power. The EIRP is just P_t times G_t expressed in dBW, and the figure of merit for a receive system is G/T in dB/K. When you combine these, the carrier-to-noise density C/N0 = EIRP + G/T - L_path - L_atm - k, where k is Boltzmann's constant at -228.6 dBW/K/Hz. This formula is clean on paper. In practice, I spent two days once trying to debug a link margin calculation that was off by 3 dB, and the issue turned out to be that I had used Boltzmann's constant in J/K instead of W/Hz/K without converting properly. The numerical value is the same but the dB representation is different, and it cost me an afternoon. For orbit determination problems, Pratt uses the Keplerian elements approach. The altitude determines the orbital period through T = 2*pi*sqrt(a^3/mu), where mu is Earth's gravitational parameter at 3.986e14 m^3/s^2. A common mistake is using diameter instead of radius when the problem gives planetary diameter. Another subtle point is that geostationary altitude is approximately 35,786 km above the equator, not round numbers. If you calculate 36,000 km, your velocity and period will drift enough to matter in exam conditions. The polarization isolation problems are another category that students find tricky. Cross-polarization discrimination typically runs 25 to 30 dB for good antennas, but if Pratt's problem states 20 dB and you default to 30, your interference calculation will be wrong. Always use the number given in the problem statement even if it looks conservative. Real satellite transponders often have worse polarization isolation than textbook examples, which is why link budgets include margin for exactly this reason.
When working on rain attenuation, the specific attenuation model gamma = k * R^alpha applies, where R is rain rate in mm/hr and k and alpha depend on frequency and polarization. At 12 GHz, rain is relatively benign. At 30 GHz and above, it becomes a serious design factor. I once designed a Ku-band uplink where the engineer had calculated clear-sky margin but neglected the 2 dB rain fade allowance. That link failed every afternoon during monsoon season because the fade margin was simply not there. If you are looking for solution resources, the textbook does not come with an official answer key. Most people who have worked through these problems share their approaches on engineering forums and academic sites. The solutions follow the same structure I described above: state your assumptions, show each substitution step, track units carefully, and verify that your final number sits in a physically reasonable range. If your C/N0 comes out to positive 60 dB-Hz, something went wrong. Typical values for a well-designed satellite link sit between 10 and 30 dB-Hz depending on the band and application. The biggest practical limitation of relying on textbook solutions is that real-world satellite engineering involves regulatory constraints, spectrum sharing, and equipment tolerances that Pratt's problems deliberately simplify. The math is the foundation, but the actual deployment adds things like polarization misalignment losses, pointing errors, and atmospheric turbulence that do not appear in end-of-chapter exercises. If you are studying for an exam, the textbook methods are sufficient. If you are building a system, plan for the messier version.
