Consider two complex numbers:
N 1 = 1 + j 0 N 2 = A + j B
Form a Cartesian coordinate system with A on one axis and B on the other. Within this A B plane, what is the area of the region described by the following inequality?
∣ N 1 ∣ + ∣ N 2 ∣ ∣ N 1 + N 2 ∣ ≤ 4 1
Notations:
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Thank you for sharing this. Very interesting
Engineer vs mathematician: Fight!
Log in to reply
Haha - I wondered if anyone would notice that subtle change...
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
|
This is a brute force approach to the problem. Looking forward to seeing more insightful solutions
Very nice problem and a very nice solution.
You can get an explicit formula for B in terms of A , and you can integrate it, but it's horrible. I'm trying to see if there's a neat way to do it but haven't found one yet.
It might help to know that this shape is a "Cartesian oval" - though I haven't found any area formulas online.
Log in to reply
That's interesting. I hadn't heard of the Cartesian oval before.
I spent some time trying to deduce a neat expression in terms of A and B . I did plot the region (plot not included), and just by eyeballing it, it did look like a circular shaped curve. Please do share your findings if you have any.
Log in to reply
I've posted them as a solution, such as they are. As I say, if I can get the final form any neater I'll post an update.
The fact it's so close to an ellipse is enticing. It would be nice to show that more formally - kind of approximating a quartic curve with a quadratic one - but I'm not sure if there's a good way to do that.
Problem Loading...
Note Loading...
Set Loading...
Putting the conditions together, we're interested in the region 4 ∣ 1 + A + B i ∣ ≤ 1 + ∣ A + B i ∣
Let's try to find an equation for the boundary of this region - ie where the two sides are equal. Expanding the absolute values, this is 4 ( 1 + A ) 2 + B 2 = 1 + A 2 + B 2
Note that B only appears in the form B 2 . This suggests looking for a biquadratic equation in B . With some unpleasant manipulations... 4 ( 1 + A ) 2 + B 2 1 6 ( ( 1 + A ) 2 + B 2 ) 1 6 + 3 2 A + 1 6 A 2 + 1 6 B 2 1 5 + 3 2 A + 1 5 A 2 + 1 5 B 2 ( 1 5 + 3 2 A + 1 5 A 2 + 1 5 B 2 ) 2 2 2 5 B 4 + ( 4 5 0 A 2 + 9 6 0 A + 4 4 6 ) B 2 + ( 2 2 5 A 4 + 9 6 0 A 3 + 1 4 7 0 A 2 + 9 6 0 A + 2 2 5 ) = 1 + A 2 + B 2 = 1 + 2 A 2 + B 2 + A 2 + B 2 = 1 + 2 A 2 + B 2 + A 2 + B 2 = 2 A 2 + B 2 = 4 ( A 2 + B 2 ) = 0
This is our hoped for-biquadratic (ie a quadratic equation in B 2 ). Plugging it in to the quadratic formula gives B 2 = 2 2 5 1 ( − 2 2 3 ± 8 2 − 7 − 1 5 A − 4 8 0 A − 2 2 5 A 2 )
or
B = ± 1 5 1 − 2 2 3 ± 8 2 − 7 − 1 5 A − 4 8 0 A − 2 2 5 A 2
To get real solutions, we need the second ± to be a + , so finally B = ± 1 5 1 − 2 2 3 + 8 2 − 7 − 1 5 A − 4 8 0 A − 2 2 5 A 2
Some quick points about this: the curve is roughly elliptical (in fact, this is a Cartesian oval). The region we're after extends in the A direction from − 3 5 to − 5 3 , and in the B direction from − 1 5 1 9 + 2 4 6 to 1 5 1 9 + 2 4 6 . If we assume the shape is an ellipse (it's not), and use these values to find its axes, the area we get is R ′ ≈ 0 . 9 1 9 6 7 , which is very close to the true answer.
However, Wolfram|Alpha can actually handle the integration we need it to do. Plugging in R = 2 ∫ − 3 5 − 5 3 1 5 1 − 2 2 3 + 8 2 − 7 − 1 5 A − 4 8 0 A − 2 2 5 A 2 d A
we find R = 0 . 9 2 1 5 3 4
Wolfram|Alpha can actually do the indefinite integral as well, but weirdly doesn't seem to be able to cope with the output. If I can get it into a manageable form, I'll post the exact value here.