COMUNICAÇÃO DE ÁUDIO E VÍDEO

INSTITUTO SUPERIOR TÉCNICO

Year 2011/2012 - 2º Semester, Responsible: Prof. Fernando Pereira

2^nd Exam – 29^th June 2012 (Friday)

MEEC: The marks should be out before 2^nd July (Monday) at 12am at the CAV Web page and the exam checking session will on the 2^nd July (Monday) at 5pm in room LT4.

MERC: The marks should be out before 2^nd July (Monday) at 12am at the CAV Web page and the exam checking session will on the 2^nd July (Monday) at 2pm in room 0.15.

The exam is 3 hours long. Answer all the questions in a detailed way, notably by including all the computations made and justifying in detail your answers.

Don’t get ‘trapped’ by any question; move forward to another question and return later. Good luck !

I (1 + 0.5 + 1 = 2.5 val.)

Consider an analogue TV system with the following characteristics:

• Number of ‘image elements’ per line:760; Kell factor: 0.7; Form factor: 16/9

• Total duration of a line: 64 μs; Frame rate: 25 Hz

• Useful duration of a line scanning: 0.92; Useful percentage of lines per image: 0.90

a) Using only the values above, and without making any specific further assumption, determine the minimum bandwidth (base band) needed for the luminance signal. (R: 6.454 MHz)

b) Determine the total number of lines in an image for the system above (still without further assumptions). (R: 625)

c) Provide and explain two main reasons justifying the adoption of the Vestigial Side Band modulation for the luminance signal.

II (1 + 1 + 1 = 3 val.)

Consider the JPEG standard to code photographic images.

a) Determine the compression factors that you would need to achieve for the luminance and for the chrominances to spend an average number of 0.64 bit/pixel (considering both the luminance and the chrominances) when coding a 4:2:0 image with 8 bit/sample, knowing that the average luminance compression factor is twice the average chrominances compression factor. (R: 25; 12.5)

b) Determine the total number of bits that have to be spent to code a 720×576, 4:2:2, 8 bit/sample image if an average number of 3 DCT coefficients are coded per block and each coefficient costs, on average, 4 bits; additionally consider that the EOB (End of Block) word costs 2 bits and all blocks in the image spend bits. (R: 181440 bit/image)

c) Consider a 720×576, 4:2:2, 8 bit/sample image coded with the hierarchical mode. How many layers can we use to code the image if the base layer global (luminance and chrominances) compression factor is 20, the global compression factor doubles for each new layer, each new layer has twice the resolution in both directions, and the total number of bits spent should be less than 10⁶ bits. (R: 2)

III (2 + 1.5 + 1 = 4.5 val.)

Suppose that you have been contacted by a company to design a videoconference solution to work between the various EURO’2012 stadiums using the lowest possible bitrate while guaranteeing the necessary minimum video quality. The company also requires that the initial visualization delay (measured as the maximum time difference between corresponding acquisition and visualization instants) is below 300 ms. The video resolution is CIF (352×288 luminance samples), 4:2:0 at 12.5 Hz with 8 bit/sample. Assume that you have available and providing the necessary minimum video quality two systems:

1. H.261 system with average compression factors of 18 and 22 for the luminance and chrominances, respectively, and critical compression factors (for the most difficult images) of 12 and 15 for the luminance and chrominances, respectively.

2. MPEG-1 Video system with N = M = 3 and average compression factors of

a. 18 and 22 for the luminance and chrominances, respectively, in the I frames

b. 25 and 35 for the luminance and chrominances, respectively, in the B and P frames

The critical compression factors are 75 % of the average compression factors for all frame types.

a) Determine which of the solutions above you would select to better satisfy the needs of your client assuming that the transmission rate for each solution correspond to its coding rate. (R: H.261)

b) Determine for which transmission bitrates the MPEG-1 solution would satisfy the initial visualization delay requirement, assuming that for this solution the transmission rate may be regulated (meaning that the coding and transmission rate would be different). (R: > 1.024 Mbit/s)

c) Explain 2 performance impacts of increasing the value of M, this means increasing the number of B frames between two anchor frames.

IV (1 + 1 + 0.5 + 1 = 3.5 val.)

Consider the MPEG-1 Audio standard to code audio content with 22 kHz bandwidth; assume reasonable compression factors and the most usual number of bits per sample.

a) How many complete stereo music pieces, with a duration of 4 minutes, can we store in a 900 MBytes disk using the Layer 3 of the MPEG-1 Audio standard to code the music content with a transparent quality regarding CD music content. (R: 255)

b) What is the maximum duration of each music piece that we can afford if we want to store 1000 musics in the same disk as above using a Layer 2 MPEG-1 Audio codec? (R: 40.91 s)

c) Explain how would the maximum number of stored musics vary if we increase the audio bandwidth three times but the audio becomes mono and not anymore stereo.

d) Describe two main technical differences between the MPEG-1 Audio Layer 2 and Layer 3 codecs and the corresponding advantages.

V (1 + 1 + 0.5 + 0.5 + 0.5 = 3.5 val.)

Consider a DVB digital TV system.

a) Knowing that a DVB solution may ‘insert’ 10 Mbit/s of total bitrate in a 8 MHz bandwidth channel, determine what would be the source bitrate that may be ‘inserted’ if all the system parameters stay the same with the exception of the channel coding ratio that goes from ½ to 1/3 and the modulation that goes from 8-PSK to 64-QAM. (R: 6.666 Mbit/s)

b) Why is it essential in a Single Frequency Network that the transmitters send the same data and do that well synchronized to transmit the same symbol at precisely the same time ? How do the transmitters obtain the necessary time reference ?

c) What are the two main components of the channel coding solution in DVB-x2 ?

d) What parameter can be used to tune the correction capability of the channel coding solution and what does this parameter express ?

e) What is the main reason justifying the availability of two channel coding block lengths in DVB-x2 ?

VI (1 + 0.5 + 0.5 + 0.5 + 0.5 = 3 val.)

Consider a 3D video system.

a) Explain what is a frame compatible stereo format. Also explain the difference between a spatial multiplexing and a time multiplexing frame compatible stereo format.

b) Explain why frame-compatible stereo video tends to have higher spatial frequency content characteristics.

c) What is the most important new feature/tool of the Multiview Video Coding (MVC) standard regarding the H.264/AVC standard ? How does it work ?

d) What is the implication of the ‘backward compatibility’ requirement for the MVC standard ?

e) If the backward compatible view in a MVC stereo pair spends 2 Mbit/s, what is the minimum rate that the second view has to spend if a perceptual quality similar to stereo simulcasting with 2+2 Mbit/s had to be achieved ? Why? (R: 500 kbit/s)