• Unfreeze’s

  • The globetrotter sw/hw architect blog… (yet another geek …)

14th May 2008

UIRA source code back online

As I have been asked, I put back the UIRA svn online, check out the UIRA Source Code Page

posted in Linux | 0 Comments

14th May 2008

New pictures upload

It’s been a while since I had time to upload pictures to this blog, but I finally had 5 minutes for doing that. The new set includes:

More to come on next trips, right now, I am still in Dongguan (东莞), and definitely, this place has nothing nice to visit … I may take some pictures anyway, but not sure.

posted in Blog, China | 0 Comments

10th May 2008

Echelon PL3120, PL3150, PL3170 (Neuron Chip) Integration and Hints

Those days I am working on the Echelon PL3120 Chip (to make it integrate our products), I having some problems to integrate it properly, here is some hints for you guys if you are trying to integrate them as well (The following information is also applicable on the PL3150 and PL3170, and I guess PL31xx):

Read the rest of this entry »

posted in Electrical Engineering | 0 Comments

5th May 2008

Dongguan SMS Spam Problem

Those last weeks were quite busy on the business trip thing, been to SuZhou (苏州) and Dongguan (东莞)… but now I am facing a little problem in Dongguan … I am beeing spamed ! using SMS, tons of SMS.

About 20 to 30 Spam SMS an hour (!!!) issued by a non disclosed sender, Saying ‘GMCC’ … and ‘Dongguan’ … (what’s an interesting text, isn’t it ?).  Windows Mobile (installed on my pda phone), comes with a Black list application, which is fine but … which can’t handle non disclosed senders … The only cure I found was deactivating the sound notification of new incoming message … and I clear them up every once a while …

On the other end, SMS Outlook doesn’t help, as, as far as I know, it does not support filters or spam protection …

Windows Mobile: Yet another spam victory

posted in China | 0 Comments

11th April 2008

Cosinusoidal Regression

Here is the problem:

Given 3 points P_0\begin{pmatrix}x_0\\y_0\end{pmatrix}, P_1\begin{pmatrix}x_1\\y_1\end{pmatrix} and P_2\begin{pmatrix}x_2\\y_2\end{pmatrix} verifying x_0<x_1<x2 and y_0\ne0, y_1\ne0 and y_2\ne0. We want to know what cosinusoidal curve \mathcal C pass though those 3 points like : \mathcal C: y = A\cos(\omega x +\phi)

Read the rest of this entry »

posted in CNAM, Electrical Engineering | 0 Comments

30th March 2008

Debian package mini howto

As I am trying to put all my former docs online, here is a rough debian packaging howto

Read the rest of this entry »

posted in Linux | 0 Comments

30th March 2008

Upgrade to Wordpress 2.5

Here is the new version of wordpress…

Upgrade was pretty straight forward, and I didn’t loose any of my plugins… so far so good. but before you get too excited … I got to tell you at least the “visual editor that does not mess with your code” thing … well does not work at least for embedded php code …

posted in Blog | 0 Comments

22nd March 2008

Finding the minimal distance of a Hamming code

Still guessing about it but, I couldn’t find an elegant way to find the minimal hamming distance of a code … till I finally got it !, here it is:

Basically the question falls to find the minimum amount of difference between two different codewords, say you got your codeword which is n bit long, systematical (meaning you got m bit of message, and k bit of code), so n = k+m (off course (m,k,n)\in\mathbb N^3).

Obviously, the minimum distance can’t be more than k … But it can be less, for sure …

The first solution that comes to my mind is to actually check the distance of all codewords, it’s easy, got two codewords C_1=(c_{1_k}) and C_2=(c_{2_k}), the distance is just \sum_{i=1}^n(c_{1_i}+c_{2_i}\equiv 2). While this method sure works, it’s also quite overkill… it makes you acually calculate n! distance calculations, which means n\times n! modulo 2 additions, plus checks…

Well, let’s find a bit nicer way … Let be G the systemical matrix of your code, we have:

G_{m,n}=\begin{pmatrix} I_{m,m} & P_{m,k}\end{pmatrix}

Let C be a codeword, and M the message, then we have C=\begin{pmatrix}M &K\end{pmatrix}, so, we can say that the minimal distance d_{\mathrm{min}}, we can safely write d_{\mathrm{min}}(C)=d_{\mathrm{min}}(M)+d_{\mathrm{min}}(K).

The first part is easy to guess, basically d_{\mathrm{min}}(M)=1, so, d_{\mathrm{min}}\ge 1, now we need to find out d_{\mathrm{min}}(K), and that can be guessed easily from P, look, it is just the maximum distance of P lines …

Now you get it…

posted in CNAM, Electrical Engineering | 0 Comments

20th March 2008

Hamming Code, and easy endianess mistake

Working on my next practical class (ELE109), I had the impression of something wrong in that text:

Let be g(p)=p^3+p+1 the generator polynom of the code, the non systematical matrix G_a can be defined by calculating g(p), p\cdot g(p), p^2\cdot g(p) and p^3\cdot g(p):


\displaystyle G_a=\begin{pmatrix}1&1&0&1&0&0&0\\ 0&1&1&0&1&0&0\\ 0&0&1&1&0&1&0\\ 0&0&0&1&1&0&1\\ \end{pmatrix}

Well, that last one sound wrong … let’s check, it out, we can obtain the systematic matrix out of this one farily easily:


\displaystyle G=\begin{pmatrix}1&0&0&0&1&1&0\\ 0&1&0&0&0&1&1\\ 0&0&1&0&1&1&1\\ 0&0&0&1&1&0&1\\\end{pmatrix}

So say, for instance you got the data \begin{pmatrix}1&0&0&1\end{pmatrix}, then you get the code \begin{pmatrix}1&0&0&1&0&1&1\end{pmatrix}, so far, so good … That code can be written as p^6+p^3+p+1. Now let’s see what the division says:


\displaystyle\frac{p^6+p^3+p+1}{p^3+p+1} = (p^3+p) +\frac{p^2+1}{p^3+p+1}

Hell, now that’s wrong!, but what? The thing is, with that polynom we should have a code of p^6+p^3+p^2+1 …

Actually, if you read again the initial text, you get it… G_a is wrong! it should have been


\displaystyle G_a=\begin{pmatrix}0&0&0&1&0&1&1\\ 0&0&1&0&1&1&0\\ 0&1&0&1&1&0&0\\ 1&0&1&1&0&0&0\\ \end{pmatrix}

The polynom that the actual G_a represents is actually g(p)=p^3+p^2+1 … and now:


\displaystyle\frac{p^6+p^3+p+1}{p^3+p^2+1} = p^3+p^2+p+1

And now, that makes sense … Now guys, don’t trust too much texts, and also, remember a vector is little endian when writing (MSB comes first)!

posted in CNAM, Electrical Engineering | 0 Comments

20th March 2008

Why are Chinese (Asian?) more attractive ?

Well, it’s been some time I go from France to China and back, some colleague pointed at me that French/European guys think Chinese (Asian?) girls are more attractive than European … usually …

I didn’t realize that before but actually, it’s not really only about physical traits, I finally figured it out why (or at least part of it).

  1. in China, usually, people do smile … while French people usually do not (or less at least)
  2. Chinese people really are more welcoming and caring than we are …
  3. Chinese are usually slimmer … not sure it comes into consideration though …
  4. And finally, well, physically, we like differences right ?

now that I just got it, it makes sense to me, I wish that could be the same in France, you know, just that makes you feel fine in here…

posted in China | 1 Comment

« Previous Entries

  •  

  • May 2008
    M T W T F S S
    « Apr    
     1234
    567891011
    12131415161718
    19202122232425
    262728293031