Issues working with complex math

**Mitul** · Mar 3 '06, 08:45 AM

Re: Issues working with complex math

Hello friend,

I hope that following function will helpful to your. This function will
use to calculate circular distance

function Calculate_dista nce($lat1, $lon1, $lat2, $lon2) {
$theta = $lon1 - $lon2."<br>";
$dist = sin(deg2rad($la t1)) * sin(deg2rad($la t2)) +
cos(deg2rad($la t1)) * cos(deg2rad($la t2)) * cos(deg2rad($th eta));
$dist = acos($dist);
$dist = rad2deg($dist);

return $miles = round($dist * 60 * 1.1515);

}

With Regards,
Mitul Patel
mitul [at] siliconinfo [dot] com

**Sjoerd** · Mar 3 '06, 10:55 AM

Re: Issues working with complex math

Michael Trausch wrote:[color=blue]
> The problem as best as I can tell resides in the latlong_box($la t,
> $long, $miles) function that I've written in PHP, but I'm not seeing
> anything different, computationally , from what I've implemented in bc;[/color]

This could be a case of rounding errors, try the BC functions in PHP:

PHP: BC Math Functions - Manual

http://www.php.net/manual/en/ref.bc.php

BC Math Functions

**Michael Trausch** · Mar 3 '06, 07:45 PM

Re: Issues working with complex math

Sjoerd wrote:[color=blue]
> Michael Trausch wrote:[color=green]
>> The problem as best as I can tell resides in the latlong_box($la t,
>> $long, $miles) function that I've written in PHP, but I'm not seeing
>> anything different, computationally , from what I've implemented in bc;[/color]
>
> This could be a case of rounding errors, try the BC functions in PHP:
> http://www.php.net/manual/en/ref.bc.php
>[/color]

Interesting. Is there any way to decrease the rounding error, or, to
have the functions asin, sin, cos, and friends all work with the bcmath
thing?

Thanks,
Mike

**noone** · Mar 5 '06, 01:35 AM

Re: Issues working with complex math

Michael Trausch wrote:[color=blue]
> Hello everyone,
>
> I'm having a little bit of trouble trying to implement some arithmetic
> logic into an application that I'm working on, and I'm hoping that
> somebody can possibly point me in the right direction. I am working
> with a database with ZIP codes, latitudes, and longitudes, and am
> working to implement the Haversine formula alongside with another
> formula for creating a "box" that I can use to get locations from within
> a square (or as close as you can get with the Earth, anyway).
>
> While I was reading up on the math and working through the problem, I
> wrote a function (well, two) in the bc calculator language to process
> this information manually while initially working with the concepts.
> Now that I have a semi-functional understanding of the problem, I
> rewrote the logic in PHP, and am having a problem actually computing
> longitude portions of coordinates for the "box" that I want to use to
> pull data about a given region.
>
> The problem as best as I can tell resides in the latlong_box($la t,
> $long, $miles) function that I've written in PHP, but I'm not seeing
> anything different, computationally , from what I've implemented in bc;
> of course, I could just be suffering from staring at it too long, too.
> :-) I'm not sure if I am expecting something that PHP won't provide, or
> if I'm handling something not quite right in PHP's arithmetic eyes, or
> what, really. The latitudes being computed are correct, it's just the
> longitudes that are way off.
>
> I will go ahead and give a link to the code and include a sample run of
> the script. I appreciate any help/pointers!
>
> Thanks in advance,
> Mike
>
> The source:
> http://www.staffasap.com/zip_test.phps
>
> The sample run:
>
>[color=green]
>>./zip_test.php[/color]
>
> Array
> (
> [zipcode] => 30034
> [latitude] => +33.6907570
> [longitude] => -084.2511710
> [city] => DECATUR
> [state] => GEORGIA
> [abbr] => GA
> )
>
> Query SELECT * FROM dataZipCode WHERE latitude <= 33.980220566111 AND
> latitude >= 33.401293433889 AND longitude <= -122.70422048692 AND
> longitude >= -122.70422048692[/color]
[color=blue]
> Returned 0 rows[/color]

if Decatur GA is at [longitude] => -084.2511710 where is the range in
your query .. should it not be somwhere between -70 and -122

[longitude <= -122.70422048692 AND longitude >= -122.70422048692] =
-122.70422048692

**Michael Trausch** · Mar 5 '06, 10:05 AM

Re: Issues working with complex math

noone wrote:[color=blue]
>
> if Decatur GA is at [longitude] => -084.2511710 where is the range in
> your query .. should it not be somwhere between -70 and -122
>
> [longitude <= -122.70422048692 AND longitude >= -122.70422048692] =
> -122.70422048692
>[/color]

Yeah, the problem was something along the lines of how I implemented the
hunting algorithm to deduce the valid range to look in. for a 20 mile
radius, it works out differently - vastly, actually:

for a five mi radius it works:
[color=blue]
> ./zip_test.php 30034 5[/color]

Query SELECT * FROM dataZipCode WHERE latitude <= 33.763122891528 AND
latitude >= 33.618391108472 AND longitude <= -84.211503845018 AND
longitude >= -84.290838154982 ORDER BY abbr,city
Returned 3 rows
City: Atlanta State: GA ZIP: 31132 Dist: 3.1496162946772
City: Decatur State: GA ZIP: 30032 Dist: 4.2562533102956
City: Decatur State: GA ZIP: 30034 Dist: 0[color=blue]
>[/color]

As well as a 20 mile radius:

Query SELECT * FROM dataZipCode WHERE latitude <= 33.980220566111 AND
latitude >= 33.401293433889 AND longitude <= -84.092501398957 AND
longitude >= -84.409840601043 ORDER BY abbr,city
Returned 65 rows
City: Atlanta State: GA ZIP: 30307 Dist: 7.8779054595022
City: Atlanta State: GA ZIP: 30340 Dist: 13.988388086945
City: Atlanta State: GA ZIP: 30338 Dist: 19.540119063232
City: Atlanta State: GA ZIP: 30335 Dist: 9.0276794496919
City: Atlanta State: GA ZIP: 30315 Dist: 7.6332861615796
City: Atlanta State: GA ZIP: 30332 Dist: 10.300183017737
City: Atlanta State: GA ZIP: 30329 Dist: 10.293572217837
City: Atlanta State: GA ZIP: 30328 Dist: 18.716589259417
City: Atlanta State: GA ZIP: 30326 Dist: 12.656176679345
City: Atlanta State: GA ZIP: 30309 Dist: 10.784558977204
City: Atlanta State: GA ZIP: 30324 Dist: 11.123997032003
City: Atlanta State: GA ZIP: 30322 Dist: 8.3657588395676
City: Atlanta State: GA ZIP: 30319 Dist: 13.608382045463
City: Atlanta State: GA ZIP: 30317 Dist: 7.1986920468663
City: Atlanta State: GA ZIP: 30342 Dist: 15.071788907405
City: Atlanta State: GA ZIP: 30345 Dist: 10.838995754932
City: Atlanta State: GA ZIP: 30306 Dist: 9.1595992448634

[... cutting more of the stuff out]

It doesn't do a perfect radius yet, but I don't think I need it to
unless I start looking at distances > 30 miles... using a square can get
to be covering a bit of a too broad area when dealing with things larger
then that.

thanks for the help guys,
] mike

**Jerry Stuckle** · Mar 5 '06, 12:35 PM

Re: Issues working with complex math

Michael Trausch wrote:[color=blue]
> noone wrote:
>[color=green]
>>if Decatur GA is at [longitude] => -084.2511710 where is the range in
>>your query .. should it not be somwhere between -70 and -122
>>
>>[longitude <= -122.70422048692 AND longitude >= -122.70422048692] =
>>-122.70422048692
>>[/color]
>
>
> Yeah, the problem was something along the lines of how I implemented the
> hunting algorithm to deduce the valid range to look in. for a 20 mile
> radius, it works out differently - vastly, actually:
>
> for a five mi radius it works:
>
>[color=green]
>>./zip_test.php 30034 5[/color]
>
>
> Query SELECT * FROM dataZipCode WHERE latitude <= 33.763122891528 AND
> latitude >= 33.618391108472 AND longitude <= -84.211503845018 AND
> longitude >= -84.290838154982 ORDER BY abbr,city
> Returned 3 rows
> City: Atlanta State: GA ZIP: 31132 Dist: 3.1496162946772
> City: Decatur State: GA ZIP: 30032 Dist: 4.2562533102956
> City: Decatur State: GA ZIP: 30034 Dist: 0
>
>
> As well as a 20 mile radius:
>
> Query SELECT * FROM dataZipCode WHERE latitude <= 33.980220566111 AND
> latitude >= 33.401293433889 AND longitude <= -84.092501398957 AND
> longitude >= -84.409840601043 ORDER BY abbr,city
> Returned 65 rows
> City: Atlanta State: GA ZIP: 30307 Dist: 7.8779054595022
> City: Atlanta State: GA ZIP: 30340 Dist: 13.988388086945
> City: Atlanta State: GA ZIP: 30338 Dist: 19.540119063232
> City: Atlanta State: GA ZIP: 30335 Dist: 9.0276794496919
> City: Atlanta State: GA ZIP: 30315 Dist: 7.6332861615796
> City: Atlanta State: GA ZIP: 30332 Dist: 10.300183017737
> City: Atlanta State: GA ZIP: 30329 Dist: 10.293572217837
> City: Atlanta State: GA ZIP: 30328 Dist: 18.716589259417
> City: Atlanta State: GA ZIP: 30326 Dist: 12.656176679345
> City: Atlanta State: GA ZIP: 30309 Dist: 10.784558977204
> City: Atlanta State: GA ZIP: 30324 Dist: 11.123997032003
> City: Atlanta State: GA ZIP: 30322 Dist: 8.3657588395676
> City: Atlanta State: GA ZIP: 30319 Dist: 13.608382045463
> City: Atlanta State: GA ZIP: 30317 Dist: 7.1986920468663
> City: Atlanta State: GA ZIP: 30342 Dist: 15.071788907405
> City: Atlanta State: GA ZIP: 30345 Dist: 10.838995754932
> City: Atlanta State: GA ZIP: 30306 Dist: 9.1595992448634
>
> [... cutting more of the stuff out]
>
> It doesn't do a perfect radius yet, but I don't think I need it to
> unless I start looking at distances > 30 miles... using a square can get
> to be covering a bit of a too broad area when dealing with things larger
> then that.
>
> thanks for the help guys,
> ] mike[/color]

I did something like this not too long ago.

It would be nice if you could search the database for something within a
circle :-). Unfortunately, it just doesn't work that way.

I did it similar to how you did - created a square and got everything
within it. Then I recalculated the distance and manually filtered out
anything not within the required radius. Worked fine and was quite quick.

--
=============== ===
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstucklex@attgl obal.net
=============== ===

Issues working with complex math

Issues working with complex math

Comment

Comment

Comment

Comment

Comment

Comment