The basic technique is to track a projectile for sufficient time to record a segment of the trajectory. This is usually done automatically, but some early and not so early radars required the operator to manually track the projectile. Once a trajectory segment is captured it can then be processed to determine its point of origin on the ground. Before digital terrain databases this involved manual iteration with a paper map to check the altitude at the coordinates, change the location altitude and recompute the coordinates until a satisfactory location was found. An additional problem was detecting the projectile in flight in the first place. The conical shaped beam of a traditional radar had to be pointing in the right direction, but in order to have sufficient power and accuracy for this the beam's angle was limited, typically to about 25°, which made finding a projectile quite difficult. One technique was to deploy listening posts that told the radar operator roughly where to point the beam; in some cases the radar was not switched on until this point to make it less vulnerable to electronic counter-measures (ECM). However, conventional radar beams were not notably effective. Since a parabola is defined by just three points, tracking a long segment of the trajectory was not notably efficient. The
Royal Radar Establishment in the UK developed a different approach for their
Green Archer system. Instead of a conical beam, the radar signal was produced in the form of a fan, about 40° wide and 1° high. A
Foster scanner modified the signal to cause it to focus on a horizontal location that rapidly scanned back and forth. This allowed it to comprehensively scan a small
slice of the sky. The operator would watch for mortar bombs to pass through the slice, locating its range with pulse timing, its horizontal location by the location of the Foster scanner at that instant, and its vertical location from the known angle of the thin beam. The operator would then flick the antenna to a second angle facing higher into the air, and wait for the signal to appear there. This produced the necessary two points that could be processed by an analogue computer. A similar system was the US
AN/MPQ-4, although this was a somewhat later design and somewhat more automated as a result. Once
phased array radars compact enough for field use and with reasonable digital computing power appeared, they offered a better solution. A phased array radar has many transmitter/receiver modules which use differential tuning to rapidly scan up to a 90° arc without moving the antenna. They can detect and track anything in their field of view, providing they have sufficient computing power. They can filter out the targets of no interest (
e.g., aircraft) and depending on their capability track a useful proportion of the rest. Counter-battery radars used to be mostly
X band because this offers the greatest accuracy for the small radar targets. However, in the radars produced today,
C band and
S band are common. The
Ku band has also been used. Projectile detection ranges are governed by the
radar cross section (RCS) of the projectiles. Typical RCS are: The best modern radars can detect howitzer shells at around , and rockets/mortars at . The trajectory has to be high enough to be seen by the radar at these ranges, and since the best locating results for guns and rockets are achieved with a reasonable length of trajectory segment close to the gun, long range detection does not guarantee good locating results. The accuracy of location is typically given by a
circular error probable (CEP), the circle around the target in which 50% of locations will fall, expressed as a percentage of range. Modern radars typically give CEPs around 0.3–0.4% of range. However, with these figures, long range accuracy may be insufficient to satisfy the rules of engagement for counter-battery fire in counter insurgency operations. Radars typically have a crew of 4–8 soldiers. Only one is needed to actually operate the radar. Older types were mostly trailer-mounted with a separate generator, so it took 15–30 minutes to bring into action and needed a larger crew. Self-propelled ones have been used since the 1960s. To produce accurate locations, radars have to know their own precise coordinates and be precisely oriented. Until about 1980 this relied upon conventional artillery survey, although gyroscopic orientation from the mid-1960s helped. Modern radars have an integral
inertial navigation system, often aided by GPS. Radars can detect projectiles at considerable distances. Larger projectiles give stronger reflected signals (RCS). Detection ranges depend upon capturing at least several seconds of a trajectory and can be limited by the radar horizon and the height of the trajectory. For non-parabolic trajectories, it is important to capture a trajectory as close as possible to its source in order to obtain the necessary accuracy. Action on locating hostile artillery depends on policy and circumstances. In some armies, radars may have the authority to send target details to counter-battery fire units and order them to fire. In others they may merely report data to an HQ that then takes action. Modern radars usually record the target as well as the firing position of hostile artillery. This is usually for intelligence purposes because it is seldom possible to give the target sufficient warning time in a battlefield environment, even with data communications. There are exceptions. The new lightweight counter-mortar radar (LCMR – AN/TPQ 48) is crewed by two soldiers and designed to be deployed inside forward positions. In these circumstances it can immediately alert adjacent troops as well as pass target data to mortars close by for counter-fire. Similarly the new GA10 (Ground Alerter 10) radar was qualified and successfully deployed by the French land forces in 2020 in several different FOBs worldwide. ==Threats==