H. Sanneck. "Packet Loss Recovery and Control for Voice Transmission over the Internet". Ph.D. thesis, Technical Univ. of Berlin. Oct. 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... The basic idea of this booster is based on the observation that not all the packets have the same importance for objective speech quality [9] Considering the speech signal properties and low bit rate codecs features, it is possible to distinguish between important and unimportant packets [10]. We have developed three algorithms to deal with important packets: ##Selective packet loss recovery protects packets by using different configurations of the physical and data link layer and switching between them on a per packet basis. We altered for our measurements the maximum number of ....

....a kind of turbulence and causing sounds with broad frequency spectrum. In general voiced sounds have a higher energy than unvoiced sounds. 2.2.Voice codec We limited our studies to one codec, G.729. The reason for our restraint was that the G.729 is well investigated. Rosenberg [14] Sanneck [10] and Long Le [9] used it for similar measurements. Considering one codec we were able to improve the accuracy by conducting more measurements with the same parameter sets. If the measurements are done using other codecs too, they will likely lead to comparable or similar results [15] 2 The next ....

[Article contains additional citation context not shown here]

H. Sanneck, "Packet loss recovery and control for voice transmission over the internet," PhD Thesis, GMD Fokus, Berlin, Germany, (Oktober 2000).

....used to reconstruct the original speech signal. The reconstructed speech signals are then filtered through a postprocessing filter that reduces the perceived noise by emphasizing the spectral peaks (formants) and attenuating the spectral valleys ( 8] The experiment which has been carried out ([9, 10]) is to 1 We use the term frame for the unit of the encoding decoding operation and packet for the unit of transmission. One packet carries typically several frames. measure the resynchronization time of the decoder after consecutive frames are lost. The G.729 decoder is said to have ....

....a single packet due to the tight real time constraint of voice. K an error model for a single wireless link: we employ a simple Bernouilli model for bit errors which lead to frame corruptions and finally packet losses (the effects of packet loss due to buffer overflow in routers are studied in [16, 10]) The packet loss probability can be computed as follows: LNM 5O PRQS PT UV XWVY Z [I] 1) UV is the Bit Error Rate and 9 is the voice packet size at the PHY level. For our simulations we have used a value of bit for 9 : the header sizes are 24, 34, 20, 8, 12 bytes at the PHY, ....

Henning Sanneck, Packet Loss Recovery and Control for Voice Transmission over the Internet, Ph.D. thesis, GMD Fokus / Telecommunication Networks Group, Technical University of Berlin, October 2000, http://sanneck.net/research/publications/thesis/ Sann0010 Loss.pdf.

....that such correlation is very weak. This implies that patterns of short loss bursts interspersed with short periods of successful packet arrivals occur rarely (note in this context that in Table 1 the pattern 101 has by far the lowest state probability) Thus, in the following we employ a model ([12]) which only considers the past loss events for the state transition probability. The number of states of the model can be reduced from 2 m to m 1. This means that we only consider the state transition probability P ( l(s) j l(s k) l(s 1) with l(s k i) 1 8 i 2 [0; k 1] where k ....

....run length occurences larger or equal to m: Eq. 1) 6 We only discuss the results qualitatively here to give an example how an intra ow loss control algorithm performs and to show how loss models can capture this performance. Details on the simulation scenario and parameters can be found in [12]. 1 2 3 4 5 6 7 8 9 10 10 6 10 5 10 4 10 3 10 2 10 1 10 0 burst length k run length model (m=1) Gilbert model Drop Tail DiffRED Drop Tail (estimated) DiffRED (estimated) Figure 12: Comparison of actual and estimated burst loss length rates as a function of burst length ....

[Article contains additional citation context not shown here]

H. Sanneck. Packet Loss Recovery and Control for Voice Transmission over the Internet. PhD thesis, GMD Fokus / Telecommunication Networks Group, Technical University of Berlin, October 2000. http://sanneck.net/research/publications/thesis/ Sann0010 Loss.pdf.

....used to reconstruct the original speech signal. The reconstructed speech signals are then filtered through a postprocessing filter that reduces the perceived noise by emphasizing the spectral peaks (formants) and attenuating the spectral valleys ( 8] The experiment which has been carried out ([9, 10]) is to 1 We use the term frame for the unit of the encoding decoding operation and packet for the unit of transmission. One packet carries typically several frames. measure the resynchronization time of the decoder after k consecutive frames are lost. The G.729 decoder is said to have ....

....of a single packet due to the tight real time constraint of voice. an error model for a single wireless link: we employ a simple Bernouilli model for bit errors which lead to frame corruptions and finally packet losses (the effects of packet loss due to buffer overflow in routers are studied in [16, 10]) The packet loss probability can be computed as follows: pL = 1 (1 BER) s bit (1) BER is the Bit Error Rate and s is the voice packet size at the PHY level. For our simulations we have used a value of 944 bit for s: the header sizes are 24, 34, 20, 8, 12 bytes at the PHY, MAC, IP, UDP and ....

Henning Sanneck, Packet Loss Recovery and Control for Voice Transmission over the Internet, Ph.D. thesis, GMD Fokus / Telecommunication Networks Group, Technical University of Berlin, October 2000, http://sanneck.net/research/publications/thesis/ Sann0010 Loss.pdf.

No context found.

H. Sanneck. "Packet Loss Recovery and Control for Voice Transmission over the Internet". Ph.D. thesis, Technical Univ. of Berlin. Oct. 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC