Blog

Consider a VoIP phone that uses digital voice encoding with an audio CODEC that has a sampling rate of 20kHz with 16 bits per sample. How many distinct sound intensity levels can be represented with this CODEC?

Consider the following characteristics of digital audio when answering the following question: Digital representations of voice audio have some redundancy, and Listeners can often use context to guess what a speaker is saying even if some of the speaker’s message is not heard.   For example, upon hearing “H___, Alice, this is Bob,” Alice is likely to guess that the first word was “Hi” or “Hello.” Would lossy compression be an acceptable technique to use given these characteristics of digital audio?  If so, how would its performance compare to that of lossless compression?

When preparing to transmit a image file, _________ compression is/are both acceptable and useful because images typically have lots of redundancy and they often contain details that can be omitted without the user noticing.

You can represent 8 on a digital computer in several different ways. One way is to use 7-bit ASCII encoding: __________. Another is to convert the number to its binary equivalent: __________. 

Calculate message latency for a 25 Kbit message with a bitrate of 100 Kbits/second and a propagation delay of .25 seconds.  Show your calculations and remember to indicate the unit of measurement for all quantities.

Suppose parity(bitstring) is equal to 1 if there are an odd number of ones, otherwise it is 0.  If a host receives the bitstring 101011011, what does it mean, and should happen next?

Suppose you wanted to use public-key cryptography to submit a final paper via email so that only your instructor could read it. Which key would your instructor use to decrypt your submission?

How would recording (i) with a sampling rate of 20kHz recorded and using 8 bits/sample differ from recording (ii) with a sampling rate of 20kHz and using 16 bits/sample?

[Optional] You may add notes about the prior question here.

Suppose you took a digital picture of a poster that used only sixteen distinct colors, and you want to compress it.  If your digital camera stores pictures using a color depth of 64 thousand colors (that is, 16 bits per pixel) you could save a lot of file space by reducing the number of bits used for each pixel. What would be the optimal number of bits to use?  Show your calculations.