Posts

Showing posts from 2010

Neural network resemblance in biological and internet

Image
My work now is best disecribe as a mix of biology, optical analysis, and chemistry. And I had to thank a lot of people for providing the opportunity in this demanding task.

To think about neural network, it is more complicated than any internet routing algortihm. In fact, the routing mechanism derived teminologies from biological network of neurons, vice versa.


(c) Anton Rahmadi. title: neurite extensions as communication means of neurons. Image was recorded with optical maginification of 400x and fluorescent green emission/excitation.

Here are some resemblances: (1) burst signal (2) delay input and synchronization (3) spike.
Scientists are now developing algorithm to understand and mimick neuronal signal. Djiksra algorithm, a.k.a shortest path routing , being employed as an approach to biological neuron communication, other advanced mechanisms are also utilized.

To understand behaviour of healthy and impaired neuronal network (which is now a part of my task) is coincidently colliding with…

Coming up next: optimizing image analysis

Image
It is quite fortunate that I had an opportunity to work on a patented and very specialized software for cellular image analysis. The software was developed quite a long time ago and now being improved to provide a real time imaging analysis.

(c) Anton Rahmadi. title: Growing a healthy neuron to a high density ratio

From this picture, the software should be able to distinct each cell (represents by fluorescent green dots), hence producing a ratio based on cell size, intensity, and volume. Tweaking the software to meet my particular objective is not an easy task. A bit of mathematics and statistical background are needed, not only understanding the principle of image analysis.

I had to stop here, any advancement on this particular topic will be discussed later on.

Chromeleon: unofficial tweak

Image
Being offered to solve Chromeleon installation unofficially was a great opportunity to me. A hacker mind that is always hungry for challenge (in a good way of course!).

In our lab, we have eight Liquid Chromatography equipments: a pump, an auto-sampler, a thermostat, a UV detector, a photodiode array (PDA) detector, a fluorescence detector and a electrochemical detector, and a fraction collector. The problem was compatibility of those equipments to a particular version of Chromeleon (6.8SR9a).



Problems were developed when Chromeleon was upgraded to latest service pack; all equipments except the PDA went offline. But, the exact opposite situation was found with older Chromeleon. This had sparked unresolved drama between IT support and engineer from the company. One pointed out Chromeleon software bugs, while other blamed operating system incompatibility.

Well, then it came to my chance. I checked all connections using Linux. Yes, Linux! I found major connection uses telnet (port 2…

Yahoo group with new interface

Image
It becomes more beautiful, but rather uneasy for slow bandwith...

Anti-SSH Brute Force

(1) melakukan konfigurasi di /etc/ssh/sshd_config
MaxStartups 3:50:10
AllowUsers nama_user
Protocol 2
Port 222(2) tidak menggunakan nama user yang sering di target ataupun memiliki password dengan kombinasi yang sering menjadi target:
qwerty; admin; passwd; system; 123; 1234; 12345; 123456 ataupun kombinasi nama host, tahun, nama lembaga, maupun nama latin dari hewan/tumbuhan.

(3a) alternatif 1: menggunakan TCP Wrapper, mengedit file /etc/hosts.deny
sshd: ALL except blok_IP_1 blok_IP_2 blok_IP_3 contoh:
sshd: ALL except 192.168.10. <-- 192.168.0.0="" 192.168.10.0="" 192.168.10.254="" 192.168.254.254="" 192.168.="" all="" artiya="" bit="" blockquote="" except="" hingga="" sshd:="">(3b) alternatif 2: memblokir IP-IP yang suka melakukan SSH brute-force, masukkan sebagai cron:

crontab -e

59 * * * ssh-deny.sh isi dari ssh-deny.sh adalah
#!/bin/bash
wget -c http://w…