|Tests & Reviews
"Filtering" is a term that is widely used when discussing radar detectors, but the concept of filtering is often misunderstood. Sometimes, a more sensitive detector is said to have "poor filtering" while a less sensitive detector is said to have "better filtering".
X and K Band False Alert Filtering
Microwave motion sensors operate on the same frequencies as X and K band police radar. Because of this, radar detectors are plagued with false alerts, especially in urban areas. Unfortunately, the radar signals from these motions sensors are (for all practical purposes) electronically identical to real police radar. Since there's no way for the detector to tell the difference between a motion sensor and police radar, there is no way to reduce false alerts from these sources without potentially damaging your protection against real threats. Any filtering method which suppresses alerts to signals which could be police radar obviously carries some statistical level of risk. It is up to each individual driver's personal preference just how much protection they are willing to trade for a quieter drive.
Radar detector manufacturers have taken a number of steps to try and minimize X and K band false alerts, while attempting to preserve as much protection as possible against real police radar threats. Here are some of the known filtering techniques ("city modes") for X and K band:
- Less sensitivity: a less sensitive detector will naturally be a lot "quieter", since it cannot find the radar signals from as far away as a more sensitive detector.
- Suppress alerts unless the signal is stronger than a certain threshold. Or only report the alert visually but not audibly, or at a lower volume. In this case, the detector actually "sees" the radar signal, but doesn't report it.
- Suppress alerts shorter than a certain duration. For example, a detector might report signals that last longer than 300ms in duration, but not report signals that are shorter than 300ms in duration. In some low-end detectors, the signal must be present for two seconds or more to trigger an alert. These filters are set up in the detector's software, and ultimately determine the detector's "response time", and generally have much more of an effect than such things as sweep rate and processor speed.
- Suppress alerts if there are multiple radar sources. Since motion sensors are frequently encountered in multiples, this might be a good indication that the sources aren't as likely to be police radar.
- Some detectors automatically adjust the alert threshold dynamically during operation, as the detector encounters multiple X or K-Band radar sources. If a detector sees multiple weak X or K band radar sources, it might assume it is in an urban area and raise the alert threshold so that a stronger signal is necessary in order to trigger an alert. When the detector no longer sees these multiple weak radar sources, it will lower the threshold again so that even weak signals trigger an alert.
- More advanced detectors might use algorithms combining the techniques above, based on a statistical model of what is likely to be a false alert based on signal strength, number of sources, frequency, or any other information the detector has at it's disposal.
- Some detectors have other novel approaches to reduce falsing. For example, some Cobra models sense if the vehicle is stopped, and don't alert, since if you're stopped you probably aren't speeding. The Escort 9500i uses GPS to eliminate false alerts when you are travelling under 15 MPH, or alerts from previously marked geographical locations.
Ka Band False Alert Filtering
Ka-band false alerts are often caused by harmonics of local oscillator leakage from other radar detectors. Fortunately, detector manufacturers DO have techniques to reduce or eliminate these type of Ka false alerts without affecting protection against real police radar. Here are some of the techniques used to suppress Ka false alerts from other detectors:
- Look for signals at 1/3 or 2/3 of the Ka frequency detected, which represent the 1st LO and 2nd harmonic of an interfering detector. If one or both of these signals is present, then the Ka alert is suppressed as a false.
For example, many old detectors had a local oscillator of 11.55 GHz, and the third harmonic of this is in the Ka band (11.55 GHz x 3=34.65 GHz). This caused Ka false alerts. So, in order to suppress alerts from this, detectors look for the leaked LO frequency of 11.55 GHz, or the second harmonic at ~23.1 GHz (11.55x2), which would also normally be present with an interfering detector, but not police radar. But, because of X and K bandpass filtering in some of these detectors, the harmonics might be too weak to detect, so in some cases detectors must err on the side of caution and report an alert anyway. Detector makers have various ways to approach this, and some simply compare the harmonic relationship between different carriers detected in these ranges, and if there IS a relationship then the alert is suppressed. The interfering signals don't always have to be present at the same time as the harmonic for the alerts to be suppressed, they only have to be detected within a specified timeframe or number of sweeps. This is only one example, there are many different frequency schemes detector makers must take into account when filtering Ka falses.
- Watch to see if a Ka signal changes frequency. If it does, it is most likely caused by the sweeping local oscillator of another radar detector, not police radar.
- A weak, recurring "blip" of radar at a specific frequency for a specific duration can also be indicative of the sweeping oscillator of another detector. If a radar detector sees such a signal, it can take steps to prevent a Ka false alert from being caused by this interference.