Efficient layered video delivery over multicarrier systems using optimized embedded modulation

We tackle the problem of efficient image transmission over multi-carrier modulation (MCM) systems, proposing the use of a layered or multiresolution (MR) framework. In this work, we treat the source as being characterized by multiple layers of importance, and therefore deserving of multiple levels o...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Conference on Image Processing Ročník 3; s. 452 - 455 vol.3
Hlavní autoři: Pradhan, S.S., Ramchandran, K.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 1997
Témata:
Algorithm design and analysis
Bit error rate
Bit rate
Constraint optimization
Design optimization
Gain
Image communication
Image resolution
Noise level
Throughput
ISBN:0818681837, 9780818681837
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:We tackle the problem of efficient image transmission over multi-carrier modulation (MCM) systems, proposing the use of a layered or multiresolution (MR) framework. In this work, we treat the source as being characterized by multiple layers of importance, and therefore deserving of multiple levels of noise immunity, i.e. having different BER requirements. We present the idea of embedded multi-carrier modulation (EMCM) as a very effective way of achieving this, and introduce a fast table-lookup based power allocation algorithm that optimizes the multicarrier constellation design in terms of maximizing the deliverable throughput bitrates for the different resolution layers, subject to a total power constraint. Simulation results of our EMCM system reveal substantial gains (up to about 25%) in deliverable bit rates over optimized TDM-based MCM designs. Further, in typical image transmission simulations using an embedded wavelet image coder, the EMCM approach yields almost 3 dB gains in delivered quality over conventional single-resolution MCM systems.
ISBN:0818681837
9780818681837
DOI:10.1109/ICIP.1997.632155