Breaking Digital Prison: Deleting Illicit Material from the Internet

Here is the story about a digital undertaker company called Santa Cruise.
I liked their motto “The Right to be Forgotten”.
Further, read here http://mengnews.joins.com/view.aspx?aid=3039240

PhD Position in Information and Communication Technology.

Where: Sweden
Deadline: 2013-05-15.
Qualifications: The applicant is expected to hold before august 2013 an MSc degree in Optics, Photonics, Laser Physics, or similar. Experience with clean-room technology, experimental and programming skills (for numerical simulation) are an advantage.

Changing units from CM to inches in ruler in Ms-word 2007.

Sometimes when you change display option from Office Button >>Word Option>>Advance>>Display, you can’t see any changes in ruler. Mostly it is because of Asian language compatibility. 

If you have this problem, you can correct this by doing following steps. 

• Go to Office Button (Left top corner, round button) 
• Click on WORD OPTION (Near the “Exit word” button)
• Click on ADVANCED in the left panel.
• Go to the DISPLAY.
• Go to the Show measurement in unit of  drop down menu with Inches, Centimeters, Millimeters, Points, or Picas. Selcet one of them (e.g. Inches)
• Click on OK

Now,

• Press ALT+F11. You can see the macro editor. 
  
• Press CTRL+G (or Go to View >> Immediate Window).  
• In this window, type vba code as follows:

          Options.UseCharacterUnit =False

Then press Enter.

Now, it’s done. You can see your ruler in Inches.

Good luck. 

Why does Snoop suppress the duplicate TCP acknowledgements it receives after retransmitting lost packets?

Snoop suppresses the duplicate TCP acknowledgements to avoid triggering end-to-end TCP retransmissions and congestion control. A Snoop agent maintains state for each TCP connection traversing the wireless gateway. TCP data packets sent from the wired to the wireless host are cached locally, until TCP acknowledgments from the wireless host verify that they were received. When duplicate acknowledgments arrive, indicating that a packet was lost, the packet is retransmitted by the agent from its local cache. The duplicate acknowledgments are then suppressed. This avoids conflicting local and TCP retransmissions by suppressing duplicate TCP acknowledgments while local error recovery.

Why TCP without Selective Acknowledgments can only recover from a single packet loss per round trip time?

Even though multiple packets lost in the same transmission window, the sender considers the fist packet that was lost from the duplicate acknowledgement it received. After retransmission of the lost packet, the sender has to wait for the acknowledgements from the receiver for the next lost packet. So, if Selective Acknowledgements is not implemented TCP can only recover from a single loss per RTT. The Selective Acknowledgment option for TCP allows each acknowledgment to specify up to three contiguous blocks of data that have been received beyond the last packet in sequence. The sender can thus infer which packets have been lost and retransmit them without waiting for additional duplicate acknowledgments.

With some Split TCP schemes, a data packet can be acknowledged without first reaching its destination. Why is this a problem?

It is a problem with some split TCP scheme because it is a kind of violation of end-to-end TCP semantics. In fact, an acknowledgment originating from the wireless gateway may reach the sender before the corresponding data packet reaches its destination. In case, if the gateway crashes after the acknowledgment has been returned to the sender but before the data packet has reached the receiver, the sender will incorrectly assume that the packet has reached its destination safely. 
There is another issue with some split schemes is that wireless gateways face significant overhead as packets must undergo TCP processing twice.

Which are the main factors leading to TCP throughput degradations over wireless links?

One of the main reasons of throughput degradation over wireless links is transmission errors. Only few packets are corrected by frame error correction codes but more packet errors are assumed as a corrupted packet and discarded. These packets are considered as a lost packet by TCP and congestion window size is reduced as TCP takes packet loss as a sign of network congestion, in reality it may not be due to the congestion. This unnecessary reduction of congestion window may lead to the throughput degradation. 

Another reason of throughput degradation over wireless links is the Frame Error Rate (FER). It suffers from FER of 1.55% when transmitting 1400 byte frames over an 85 foot distance, with clustered losses. Reducing the frame size by 300 bytes halves the measured FER, but causes framing overhead to consume a larger fraction of the bandwidth. In shared medium WLANs, forward TCP traffic (data) contends with reverse traffic (acknowledgments). In the WaveLAN this can lead to undetected collisions that significantly increase the FER visible to higher layers. File transfer tests over a WaveLAN with a nominal bandwidth of 1.6 Mbps achieved a throughput of only 1.25 Mbps. This 22% throughput reduction due to a FER of only 1.55% is caused by the frequent invocations of congestion control mechanisms which repeatedly reduce TCP’s transmission rate. If errors were uniformly distributed rather than clustered, throughput would increase to 1.51 Mbps. This is consistent with other experiments showing that TCP performs worse with clustered losses.

Why is TCP performance more affected by losses over high-delay and high-speed network paths?

TCP recovers from errors through end to end retransmissions that is the reason why recovery is slower in high delay paths. Further, because it takes longer to reach the peak throughput supported by higher speed links TCP drastically reduces its throughput after each loss.

The CSMA/CA access protocol does not detect collision as CSMA/CD does. How does this limitation affect link-layer and transport-layer performance?

In wired network, when two transmitters start simultaneously collision occurs and that can be detected by transmitters. Then they abort their transmission and retry later. However, in wireless networks transmission and reception is done in the same band that makes collision detection difficult. Hence, the WLAN use CSMA/CA to avoid collision. CSMA/CA, is listen before talk mechanism, where transmitters wait for a random interval after the medium becomes idle before starting transmission. The transmitter that starts first seizes the medium and all others back off, there by reducing the change of collisions. However, there may be collision that is not detected by CSMA/CA and corrupted frames not retransmitted by CSMA/CA are considered by the link-layer and transport-layer. Link layer is responsible for flow control, acknowledgement and error recovery. It also species which mechanism are to be used for addressing stations over the transmission medium and for controlling the data exchanged between the sender and receiver. To recover from wireless losses, including undetected collisions, transmitters may need to optionally ask for acknowledgements to retransmit transparently unacknowledged frames at the link layer. Further, in some cases, wireless transmission errors are not related to network congestion but due to corrupted packets that not handed over to TCP. TCP assumes those packets are lost and reduces its congestion window size that leads to throughput degradation.

New Google

Thesedays, Google is a synonyms for search engine. People say googling for search in the internet. Many new innovative (not sure how innovative they are though) search engines are showing their existence in the market. 

Today, I received an email from SONSIK yahoo group member mentioning that the search engines that we are mostly using Google and Yahoo, there is a new searching engine promoting by Nepalese. The mail requested to try the search engine. I tried it and searched some stuffs too. I read the About Us column. It seems that basically the search engine is operating from US. It has offices in both country US and India. The promoters are seems to be Nepalese and Indian origin. The official blog is written in January 28, 2009, looks pretty new. 

The name of the search engine is given Sagoon. Although the GOO of Sagoon looks like taken from Google, it’s not true. It’s derived from Sanskrit which means Auspicious or ushering good results. In News section there are two language changing options – Nepali and Hindi. It can be assumed that this search engine is targeted to the Nepali and Indian users. History shows that Google and Yahoo are not big success in non-english speaking countries like China and Korea. So this search engine can be very popular as a regional search engine. 

The founder of the company Sagoon Inc. Govinda mentions in the blog that it’s the beginning of their journey, a journey with bigger challenge. Of course, as there are many Big Bosses in search engine industries occupying large market shares already, it might be challenging for Sagoon. 

Anyway, finger crossed to Sagoon team for their success. 

www.sagoon.com

Why TCP modeling?

There are certain things we can achieve with TCP modeling that cannot be achieved with simulation. Some of them are as follows. 

• We can model some uncertainties using stochastic process that cannot be determined by simply simulation. For example, number of current TCP connections, how much bandwidth is occupied by the non TCP transport protocols and what is the effect of those in TCP etc. 
  
• TCP operating scale is very large in terms of geographical scale and the number of inputs/outputs. To simulate the large number of inputs/outputs and geographical scale is not easy due to hardware constraints (some times budget constraints as well).
  
• Mathematical models are required to quantify metrics that define system performance and develop control strategies that result in the optimum being achieved.
• It also helps to determine how close it is to achieving a system optimum, as defined by a specific cost function, and subsequently identify issues that need to be addressed in future control protocol. 
  
• TCP is already mostly used transport layer protocol in the world. To test some optimization in it is not easy to implement in real world scenario and observe its performances. So, before implementation any optimization patches it is very much necessary to model it. Some examples of those patches are Fast Retransmission and Fast Recovery. These were not in the very first version of TCP.
  

The role of the "gateway" option of tcpdump

‘Gateway’ in tcpdump specifies a name of gate that transfers the packet. We can use gateway to monitor traffic from specific TCP/IP applications at strategic locations of a given TCP/IP network. For example to print all ftp traffic through snup gateway the command is as follows:
#tcpdump 'gateway snup and (port ftp or ftp-data)'

To print IP packets longer than 576 bytes sent through gateway snup the command is:
#tcpdump 'gateway snup and ip[2:2] > 576'

Disadvantages of IP fragmentation?

Some disadvantages of IP fragmentation are: 
A. Once packet is fragmented it can be reassembled only at another end. If there is larger bandwidth on the way, it cannot take an advantage of that. 
B. The overhead is large when IP fragmentation is used. IPv4 does not restrict fragmentation and can be performed at any link along the path. This can result in heavy processing demands on the routers in the network. (Not in IPv6).
C. There are some security issues with fragmentation. (e.g. Tiny Fragment Attack and Overlapping Fragment Attack)

Situations when RTO occurs even though packets reach receiver

When the ACK sent by the receiver lost and sender could not get the ACK then the RTO occurs. If there is a delay in arriving ACK to the sender due to some reasons (as described below) than RTO occurs even though packets reach receiver. For example, on a low-bandwidth link the arrival of competing traffic (possibly with higher priority), or some other change in available bandwidth, can cause a sudden increase of the RTT. This may cause a RTO. If the communication takes place from a low latency path to a high latency path, it may suddenly increase the RTT beyond the current RTO value. 

How cumulative acknowledgement can compensate for a lost ACK?

Cumulative ACK can compensate for a lost ACK with the subsequent ACKs of higher number. For example, A is a sender and B is a receiver. A sends packets to B from sequence number 101 to 201. After receiving all bytes from A, B sends a cumulative ACK of 202. Here, ACK Number 202 means, B received all the bytes up to 201. Let’s assume another case that there is no cumulative acknowledgement and all the packets are received by the receiver but one ACK is lost. In that case even if the packet is successfully received by the receiver, sender retransmits the packets, which it didn’t receive the ACK. In this case if there is cumulative acknowledgement, even one ACK is lost, if sender receives higher ACK number, sender determines that packets are received. The ACK with higher value compensate the previously lost ACK.

Why is the ISN (initial sequence number) of TCP selected randomly?

TCP layer on the TCP devices exchange Initial Sequence Number (ISN) that is assigned to any device during the startup of the TCP connection. The ISN is always random. There are several reasons to select ISN randomly. One of the reasons is that TCP segments may get mixed up with the different connections if the sequence number is started from the same number by all the devices. For example, a TCP connection is started with ISN 1. Let’s say there are 1 to 50 bytes. Suppose that there is some problem and the TCP connection is terminated. Now a new connection with the same ISN (i.e. 1) is started for the same number of bytes (i.e. 1 to 50 bytes). The device may think that these segments are the part of new connection. So, the data may get collapsed. 

The another reason to select ISN number randomly is any malicious person could write code to predict the ISN of subsequent TCP connection based on the ISNs used in earlier ones, which causes serious security risk. 

How an efficient TCP/IP implementation can extend the battery life of a mobile device?

Efficient TCP/IP implementation can reduce packet loss rate, data transfer time and resource consumption. Retransmissions of the packet eat up most of the battery power unnecessarily. If the packet loss can be reduced by the efficient flow control and congestion control mechanism then the packet retransmissions reduced significantly that conserve battery energy of a mobile device.

Which of the following applications can run satisfactorily ?

One-way delay measurement of a corporate TCP/IP network spanning two distant sites reveals that the mean one way delay is 500 ms. Which of the following applications can run satisfactorily over this network?
(a) Voice-over IP
(b) File Transfer
(c) Web Browsing
(d) E-mail
(e) Video on demand

Answer: 

Following applications can sun satisfactorily over given network: (a) File Transfer (b) Web Browsing (c) e-mail. 
The reason VOIP and VOD may not run satisfactorily is that the tolerable delay of these applications are several milliseconds only. e.g. <= 400 ms for VOIP. 

2D picture with 3D effect

Same picture with 3d effect. 

Original picture below

Picture taken in the lab with Nikon camera.

How to fill patterns in MS Excel 2007 Graph?

I am using MS-2007. Today I was making graph for my research paper. I wanted to fill patterns because I have to print it in B/W ink. I tried hard to find it hard in MS Excel 2007 but couldn’t find it. I don’t know why they excluded it. 
I found a link written how to make UI in this blog http://blogs.msdn.com/excel/archive/2007/11/16/chart-pattern-fills.aspx . It’s good. 

I found one another free UI addin to fill pattern. I tried it, it works fine. The link of that site is http://www.andypope.info/charts/patternfills.htm .