Our small gu found enemy scouts technology is sending the wireless signal s [n], he use the receiver to receive signals for [n] y=0.3 s [n], [n] + w w [n] is white noise, the distribution to the mean to 0, the variance of 1 gaussian distribution (also called a normal distribution), assuming small gu has right to know the following information:
A signal sent by a (1) the enemy s [n] value can only be + 10 or 10;
(2) sent a total of 10000 s [n], n=1, 2,... , 10000;
(3) sent 8000 of them + 10, sent 2000-10,
But little don't know n=1, n=2 times,,, n=10000 moment send s [n] what,
Please help design algorithm small gu from received the 10000 y [n] in the detection of send s [n] value, require minimum error probability, the design algorithm to help him to confirm each time (n=1, n=2,... Moment, n=10000) send s [n] is + or - 10 10, makes the probability of misjudgment minimum at the same time, send to 10, wrong judgment to + 10 is; Or send + 10, judge into - 10 is misjudgment,
The input contains two parts (can be read from the file);
1) 10000 s detection [n], namely 10000 values, each value is only two possible, or + 10-10.
2) to detect the s [n] and send the s [n] is not the same as the number of
This is a stochastic optimization problem, please write the algorithm design ideas, and please write a program,

CodePudding user response:

, this is our homework, ask the great spirit guide, give thoughts to ideas can, of course program is given better!! Thank you very much!

CodePudding user response:

I'm going back to school to remake it for sure

CodePudding user response:

Baidu search relevant keywords,

CodePudding user response:

Didn't understand, the mean to 0, variance 1
The results
10 - & gt; (2, 4)
- 10 - & gt; (4, 2)
The two didn't overlap

CodePudding user response:

Numbers for only two possible inside, not + 10-10, and occurrences are given, to compare not too low efficiency