Story of missing hiker in Joshua Tree NP

[aeolian]

Thanks Richard. Here's the source discussing cell phone distance determination (Other Hands). See page 3.
https://www.otherhand.org/wp-content/uploads/2012/04/Ewasko-ping-discussion.pdf

I’m back into speculation mode here, but I think what the Verizon tech who examined the ping
did was to look at the RSSI levels and said, “That recorded tower signal strength corresponds to
a distance of 10.6 miles”. I then took that information and drew a 10.6 mile radius around the
tower’s location. But unless we were in flat Kansas, what I did was completely wrong.

Is there newer information indicating Bill's phone used a different distance determination? I've seen in various places that multiple methods are possible, though most require more than one tower, or GPS on the phone. I'm not a cell phone expert, though I've worked in technical fields and know that seemingly simple answers (like cell phone distance) are rarely simple in practice. Conversely, some simple appearing things are simple and very accurate.

I lean in the direction of the early SAR folks who dismissed the distance as unreliable (and searched primarily outside the 10.6 miles). Especially with a single ping far from the tower in rough terrain. I take the ping clue as Sunday morning Bill's phone was almost certainly on and made contact with Serin tower, maybe from 10 miles or so.

[bretpct]

This cell phone ping is certainly a confounding piece of data. And as far I can tell, it came from one deputy's conversation with a Verizon employee. That's all we really know. We're assuming that the ping distance was measured using a calculation of time near the speed light. As far as I know, this data was never reanalyzed by Verizon or the Sheriff's department, or even asked of the Verizon employee to verify the figure he/she provided.

Furthermore, if we look at the pings recorded from the Nguyen/Orbeso case, they are nowhere near where they ended up being found. And oddly enough, one of them was recorded near Smith Water Canyon. You could say, well they were using a different system to calculate their location, perhaps GPS, and the satellites were not getting a great lock on the phone. Seems kinda odd that the older phone would, in a way, be presumed to be more accurate. With regards to atmospheric ducting, I had a firsthand experience being in the mountains with zero bars of cell signal and watching a cloud float in the perfect spot to give me enough signal to make a phone call, so I think it's certainly possible he could get a brief ping from an area not in the RSSI coverage map.

I wanted to test a hypothesis that there was human error in the calculation or communication of the ping. I present the following scenario: The Verizon tech analyzes the ping and finds it be 17 miles from the Serin Dr. tower. The tech and the deputy have a phone conversation during which it is incorrectly communicated or inferred that the distance is 17 kilometers, so then the figure is converted to miles and we get 10.6 miles. Is this a likely scenario? Probably not, but hey, if NASA can make that mistake then I'm sure it's at least possible here. Which leads me to my most recent hike.

A 17 mile radius from Serin draws a line right over peak 4928, which is also near where searchers saw a light on a saddle the first night of search. They checked out the hill system according to the narrative, but there are only a few GPS tracks in that area, so I thought it could use a little more filling in.


Here's the hike in light blue. The black line is a 17 mile radius from Serin.


Near the peak I found a metal pipe sticking out of a rock.


Upon closer examination, it is a survey marker. First one I've seen on a pipe. It was not mounted to anything and you could pull it right out of the rock.


View toward Quail from near Peak 4298


Flat lowlands area to the North of the peak


Found this weird collapsed lean-to


Found this piece of trash in between CRHT and Juniper Flats Rd. Probably unrelated since that area gets a lot of activity, but posting it just in case. Seems like some kind of foam.


Gorgeous Day. Unfortunately I didn't beat the line of traffic leaving Keys View after sunset.

This hike made a lot of sense to me. It puts Bill in a spot near his car. Maybe he figured that Quail was too ambitious for his late start, the building heat, and his low water supply. So instead he decides to summit this nearby peak. Ultimately I found nothing, but I think this area may be worth one more trip to fill in a couple more spots. One thing I was thinking about is that it was noted Bill liked to stop and take pictures. Was a camera found in his car? If not, is it presumed that he was hiking with a camera?

Tracks: https://caltopo.com/m/LL01
Mylar balloons collected: 4

[aeolian]

I like your found lean-to, and balloon haul.

I'm reading more of the previous posts. A lot of theories. And while I'd like to poohpooh them and say mine is better, they all seem possible. The firmer pieces of evidence are so limited. Some I think are over-focused. E.g. the water bottle count. Seems pretty likely Bill brought a bunch (11) in the car, perhaps because that's what you do in the desert. And seems pretty certain he took two of them on his last hike. But that says nothing about other water he may have carried.

Looks like Tom switched his thinking from cell distance based on signal strength to time of travel. I haven't seen why the thinking shifted away from signal strength, but here's some discussion in earlier posts. If the distance is time-of-travel and correct, it's hard to justify most of the original SAR search area, or anything past 10ish miles unless he pinged from the far side of Quail Mountain, then returned toward his car.

It's too bad they didn't fly a Stingray (cellphone tracker) on Sunday morning when Bill's phone may have been alive. Also too bad Bill didn't walk out Thursday... (idea, can dead cellphones be detected? The antenna may still function. Shining a tuned RF source might produce a detectable RF signature, sorta like radar)

[jdclifford]

In the type of heat Bill was faced with, I could only see him attempting to self-rescue under the cover of night or in twilight. Of course, even with it being close to a full moon at the time, this could have ended up with him finally resting in a less predictable spot. I also try not to shy away from areas which the initial search and rescue operation covered because I believe (possibly wrongly) that they were solely looking for a live person who would be somewhat responsive to calls. Somebody please correct me if I am wrong on this assumption, as I would hate to be giving other people false ideas. Search and rescue was also searching for Bill in blazing temperatures, similar to a hike I did there in late June of 2020. I can only speak for myself, but the heat saps my mental acuity and I even found myself being less thorough than usual in my search. At least when it is cooler outside, you are not placing significant focus on self-preservation. I do not regret the hike because I wanted to be in similar conditions that Bill was faced with in on that fateful day, but I certainly would not do it again.

It is not totally uncommon that when a missing hiker's remains are found, they are found in areas already searched. I'm not leaning more toward this being the case, but it is something to consider. There are lot of nooks and crannies in some of those rockier areas of JT. I wouldn't suggest anybody overlap tracks, but I also wouldn't shy away from getting a stone's throw from them in rockier areas.

But like Tom and others have said here, at this point literally anything is possible!

[sk43]

The idea that the distance from the tower to Bill's phone was determined based on signal strength was merely a conjecture by Tom without any particular foundation. I am no cell phone expert either, but did find a good discussion of cell phone signal strength in patent US5056109A. Basically, signal strength is continuously bouncing around due to a number of effects (of which distance is only one) and the transmission power from a cell phone is adjusted continuously as well so as to make the signal strength received by the base station approximately constant. Distances inferred from signal strength would at best be quite rough (errors of order a factor 2). RichardK is correct, that the distance provided to the Sheriff's office was based on timing data.

[RichardK]

Here is Otherhand's discussion of using the speed of light to determine the tower to phone distance.

https://www.otherhand.org/home-page/sea ... y-12-2012/

Here is the field test he did with cell phone guru Mike Melson

https://www.otherhand.org/home-page/sea ... uary-2013/

But overall, Verizon’s path length measurements were very close, strongly suggesting their 10.6 mile measurement for Bill’s ping was in fact accurate. Validation of Verizon’s system also implies that any true distance for Bill beyond, say 11.1 miles (within 10% of 10.6 miles to quote the Verizon tech) would be a physical impossibility. This excludes all areas southerly of Smith Water Canyon, such as Quail Mountain and Lang Canyon. You know, those areas we spent so much time on.

[aeolian]

Thanks Richard. I've been paying particular attention to the cell phone in my reading of Tom, Adam, and this thread. I have less confidence in the 10.6 mile limit.

First, I have no new data, just a fresh set of eyes, and willingness to question assumptions (aka speculation).

I agree that if RTT is how distance was determined for Bill's brief ping, then his phone could not have been much outside of 10.6 miles (and unlikely much inside). However, I don't see where the Verizon technician, or any other expert, affirmed that RTT was used to calculate the distance (in fact there's no Verizon information besides the one scanty report of brief ping, 10.6 miles, 10% certainty - rehash of Tom, not a source document). Melson's work shows that Verizon's RTT system was quite accurate (in Jan 2013 report using a smartphone). It's entirely reasonable to me that Bill's phone was not "currently connected" on the network, and hence not yet enrolled in the RTT system - the ping was too short. If Verizon determined the distance from the phone's initial registration, the phone would have only signal strength to report. As I recall, cellular providers were not particularly interested in accurate location determination, and were dragged into it due to 911 needs. I'm saying that even if a high accuracy method like RTT were in use, it may not have been supported by the tower, or the phone, or may not be used in a brief ping.

In Some case clarifications from the Riverside County Sheriff’s Office, Tom says

– There is no detailed cell phone info in Ewasko’s file beyond what we already know and I’ve posted. If any such data ever did exist, it’s long gone.

Basically I don't see where RSSI vs RTT is discussed or why to prefer RTT as the distance algorithm used for Bill's ping. I do see RTT gives a narrow arc for Bill on Sunday morning and favors areas reachable from there for his resting place. RTT tends to rule out far away places like near the trailhead, or Stubbe Spring. RSSI makes them only quite unlikely, as it requires ducting (unlikely), and still favors areas near 10.6 miles. I'm also weighing the experience of the initial SAR folks (who basically ignored the ping distance) against analyses several years later (based on the sheriff's report of the Verizon tech's report). I want to assign 70-80% probability that Bill's phone was near 10.6 miles, but those SAR folks didn't.

[OtherHand]

I guess I'll jump in for some clarification, but be warned it's been a while and my memory may be fuzzy.

The reason the initial SAR people (and later myself) downplayed the 10.6 distance is that it was nuts. It didn't make sense based upon Ewasko's supposed destination and his car. But do note that as soon as the SAR HQ received the 10.6 mile figure they began sending teams into Smith Water and adjacent areas. I'm not familiar with current SAR ping accuracy, but in that era most SAR people were pretty skeptical of reported locations as they were often wrong....by a lot. But they were correct enough you had to check them out.

My early investigations into cell phone location reporting, way back then, is that signal strength (RSSI) was often used. With reflections and diffractions in rocky terrain, it's pretty certain tremendous errors could be possible in Joshua Tree. So in my initial searches I didn't pay much attention to the 10.6 distance, but rather looked primarily at locations of coverage based upon signal splash maps. I was as far out as the hills westerly of Ryan Mountain, which is well over 20 miles from Serin. Bill wasn't there. But almost all areas of cell coverage from the Serin tower, that made sense, were covered to some extent. And when you eliminate the possible, all that left is the.....Yeah, that.

It wasn't until I had multiple discussions with the aforementioned Mike Melson that I started seriously focusing on the 10.6 area. Melson worked in the cell industry and had contacts at Verizon. It was he who told me that the Serin tower used time of flight methodology to figure distance, info he got from Verizon. Foggy memory here, but he described a chip in the phone (or maybe the tower...can't remember) that had a series of "registers" that acted as a counter and was capable of measuring the nanosecond time differential. I recall being quite impressed at the technology for such small measurements, so his description stuck with me. Melson later did field testing with a hacked Verizon phone and verified the accuracy of what Verizon reported as distance from the Serin tower versus actual. So if Serin used time of flight measuring, the ping pretty much HAS to be near the 10.6 mile mark, unless the Verizon tech really screwed up.

My feeling is that Bill is in some nook somewhere, which searchers passed in close proximity to. People who haven't been out in that terrain really don't grasp how easy it would be to go by something even modestly shielded from view. Also, we assumed Bill was traveling with a logical destination in mind (i.e., get off the damn mountain) and thus focused on logical paths. If Bill didn't do that, but instead headed for one of the thousands of local high points in an attempt to get a cell signal, there's no way to search all that.

[jdclifford]

Thanks, Tom. Yeah, looking at pictures simply does not do justice to just how many hidden nooks there are. And I agree about if he climbed for cell service. A lot of those those areas are just rocks piled on rocks...seemingly endless places to look. A needle among many haystacks.

I'm thinking I might take a sabbatical year from searching for Bill and conduct more local searches, especially considering I can't bear full weight on my left ankle until late January. I don't visit stores when I travel (all my food and drinks are bought prior) and I'm fully-vaxxed, but I need to learn more about omicron, etc. before I head there again. We shall see! In the meantime, I'll be crossing my fingers for Bret and others who live closer to there.

[jdclifford]

Not sure of a better way to do this guys, but I'm dumping the hikes I've conducted in searching for Bill. The max attachments is 10, so there will be subsequent posts. The file name is my name, obviously, with the following sequence of numbers being the date hiked.