Part IV
Sorry, I got distracted, but I continue on the journey. Just so you know, along the way between parts Iv and this one, I typed this up: Why We Shelved TASMs. That post details much of what the falout of the report we generated, but also some dealing I had years later, as a result of this exercise 20 years ago this past August.
We arrived back in Norfolk, and returned to our digs on the ground floor of Building “Whiskey-5” (“W5” as the sign on the building said). As usual, our plate was full and I looked at the Ops Boss and asked when we were going to get to work on the tactical analyis. He said he had the next operation to plan (he did) and went back to sorting through the large stack of daily messages. Being the Ops Boss, he owned the two operations types, OSCS(SW) Koch and RMCS(SW) Rumbaugh, and they were put to work. Oh, well, off to the small shared conference room I went, hauling box after box, after box of raido logs, DRT traces, radio-telephone (R/T) logs, weather reports, intel messages and much more. For quite some time, I would sit in there, with charts, dividers, logs and notepads, piecing together engagements from first detection to simulated impact.
I managed to pull a total of 59 complete engagement sequences out of the piles of data. This data not only included the track of the intended virtual, constructive or cooperative target reckoned by the shooters and search aircraft, but the overlay of the actual tracks. While the times of position didn’t always match, I plotted the ded-reckoned tracks to allow some degree of checking apples to apples. For the flight of the simulated TASMs, I plotted the ded-reckoned tracks, based on the engagement plans, printed for final approval by the CO before the simulated firings. During this, I read the tactical signal logs between the shooters and reviewed any other available data.
I don’t recall how many total engagements were run. Some were disgarded for the fact that not all elements of the detection and tracking process were there. Some were not written down, some were not included in the records, most likely to not being packaged up properly. In any case, for 4 dedicated days, an average of almost 15 engagements a day wasn’t too bad for extracting some meaningful info.
In order to manage the data, I hauled in my new toy, a Macintosh 512K, to the office, and ran Excel as my database manager. The data was all kept on 800K 3.5″ floppies, as internal hard drives weren’t a common thing with home computers yet. I seem to recall the printed out data was a 8 page wide by 2 page high form, which I peeled apart (tractor feed paper in an ImageWriter ][ dot-matrix printer) and taped together to show the Commodore. Included in the spreadsheet (which wasn’t even available for the IBM PC yet), was my first serious work in making calculations using spherical geometry. It took a bit of reading in Bowditch, then dusting off the college level trig, and spending quite a bit of time making sure the parenthesis were placed properly to do computations between two fixes. I also did a lot of testing of the formulas, just to make sure I’d get the right answers. This was done so for every fix of the target, I would compare the “miss distance” between the apparent position (from the shooter’s track files) to “ground truth” (the target’s navigational files, which were assumed to be accurate in any case). The speadsheet would indicate the azimuth and distance of miss all along the sequence to engagement.
Read the rest of this entry »