Testing & Training
 By Fran Ridge


     UFOlogy students are aware that many times radar would detect an unknown object and the Air Defense Command would sometimes scramble jet fighters to intercept and identify them. This was good training since these objects would invade our airspace with no warning.
guidelines I have initiated allow for one credited MADAR hit * per month per node so that we know the device is working correctly and it also provides monthly training so that ops don't become complacent, even if the "hit" is a false alarm. After all, at the end of the day what determines a valuable correlation is a good PSOI (Potential Sighting Of Interest) which we are looking for all the time.

     In an effort to improve on the MADAR system, there are a couple of things that I am directing that are new
ideas. Some DataProbes had thresholds that were set too high and were not picking up any anomalies the last few months. Others were getting too many "hits" because their threshold shields were set too low. I knew that most ops wouldn't know how and when to adjust their equipment so I went to work on it.

     The MATCH dBase I have here at MADAR Operations can be sorted in many ways by columns. Especially valuable is the sort by by node and date, all of the credited hits for each device. Some devices were having major "hiccups" because of bad locations, meaning they were in close proximity to some local E-M interference. So I had the ops move their devices a few inches, and that solved a lot of those issues, but raising the threshold slightly fixed others. In some cases, lowering some threshold values was required and some units came to life.

     Each day  I get a report from the server which shows hits from the previous day. 
After an adjustment, this list helps me to decide which ones have responded properly and others which have not.  The online NodeReports folder is a constantly-updated source of data on all the nodes.

So the whole system is operating much better and we are getting the hit numbers down somewhat so that the server isn't overloaded and our signal to noise ratio has been improved.
For example, one week in early March we logged 155 hits! (That's with about 90 units running, BTW). That's too many hits to be real and a lot of "noise" in the system. The new guidelines insure that the "hit" list will be getting smaller and much more accurate. It also lowers the burden of inputting many credited hits to the NUFORC. If a good sighting is in the area, we are looking. In fact we are working on a means to have our computers do a lot of the work for us.

To view specific node data you can access at this folder:
To view data concerning all nodes listed by node and date go to:

* Credited "hits" are those that were obtained by the device and uploaded to the server that were not documented as a false alarm, or had no problematic data.