Coding of correlated sources with prescribed distortion by separated encoders

AUTOR(ES)

Wolfowitz, J.

RESUMO

Let (Xn,Yn) = (X1,Y1),...,(Xn,Yn) be n independent, identically distributed pairs of chance variables. Each of two separated encoders encodes a different sequence of (Xn,Yn) to within given respective distortions z1,z2. Call a region in the plane of the respective rates (R1,R2) M-convex if it is convex, closed, and, to every point (r1,r2), contains all points such that each of their coordinates is larger than or equal to the corresponding coordinate of (r1,r2). This paper determines the (M-convex) region of achievable rates by giving its boundary. For z1 = z2 = 0, the region specializes to that of Slepian and Wolf. For z2 = ∞ and R2 = H(Yi), the minimum R1 is that of Wyner and Ziv; for z2 = ∞ and general given R2, the minimum R1 is that of my generalization of the theorem of Wyner and Ziv.

Documentos Relacionados

• UNIVERSAL STRING MATCHING ENCODERS BY RECURRENCE OF STANDARDS FOR SOURCES WITH FINITE NUMBER OF STATES
• "Collective coding" of correlated cone signals in the retinal ganglion cell.
• Use of Alternative Medicines by Patients with OA that Adversely Interact with Commonly Prescribed Medications
• Genetic Dissection of Segregation Distortion II. Mechanism of Suppression of Distortion by Certain Inversions
• Propulsion of African trypanosomes is driven by bihelical waves with alternating chirality separated by kinks