Scribd
Upload
7 views10 pages

Convolutional Network Coding Explained

Convolutional network coding is utilized for data transmission in cyclic networks, ensuring no information loops despite cycles. It employs local and global encoding kernels to model the effects of input on output over time, with a unique convolutional code guaranteed under certain conditions. The Viterbi algorithm is highlighted as an effective method for decoding the transmitted signals in these networks.

Uploaded by

emaihai01
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
7 views10 pages

Convolutional Network Coding Explained

Convolutional network coding is utilized for data transmission in cyclic networks, ensuring no information loops despite cycles. It employs local and global encoding kernels to model the effects of input on output over time, with a unique convolutional code guaranteed under certain conditions. The Viterbi algorithm is highlighted as an effective method for decoding the transmitted signals in these networks.

Uploaded by

emaihai01
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Convolutional Network Coding -

Key Concepts
Introduction to Convolutional Network
Coding

• Used for data transmission in cyclic networks with unit-delay model.


• Each channel transmits one symbol per time unit (discrete time).
• Ensures no information loops despite network cycles.
• Modeled using z-transform to capture time delay as power series.
• Key constructs: Local Encoding Kernels (k), Global Encoding Kernels (f).
Definition of Convolutional Network Code

• Local kernel k_{d,e}(z): Defines how input d affects output e over time.
• Global kernel f_e(z): Represents total effect from source to channel e.
• Equation: f_e(z) = z * Σ k_{d,e}(z) * f_d(z)
• Rational power series ensures causality and finite memory.
Example 20.7 - Network Setup

• Nodes: s (source), x, y, w; Channels: s→x, s→y, x→y, y→w, w→x.


• Local Kernels: K_s = I, K_x = K_y = [1 1]^T, K_w = [1].
• Feedback loop: x → y → w → x creates a cycle.
Example 20.7 - Solving the Loop

• Recursive equations built due to cycle.


• f_{x,y}(z) = z * (f_{s,x}(z) + f_{w,x}(z))
• f_{w,x}(z) = z * f_{y,w}(z), f_{y,w}(z) = z * (f_{s,y}(z) + f_{x,y}(z))
• Solving: f_{x,y}(z) = (z^2 * a(z) + z^4 * b(z)) / (1 - z^3)
The Convolutional Network Code Existence
Theorem

• Let F be the base field and kd,e(z) ∈ Fhzi be given for every adjacent pair
of channels (d, e) on a unit-delay network. Then there exists a unique ω-
dimensional convolutional network code over F with kd,e(z) as the local
encoding kernel for every (d, e).
• Given local kernels, there exists a unique global convolutional code.
• Matrix form: [f(z)] = z * H_s(z) * (I - zK(z))^-1
• Guarantees causal, rational, and implementable network behavior.
• Captures full system behavior using local node rules only.
Decoding in Convolutional Coding
• • Each node receives a linear combination of
transmitted signals.
• • Decoding possible when sufficient maxflow
exists to the node.
• • There exists a decoding matrix Dₜ(z) for such
nodes.
• • Goal: Recover original source message x(z)
from received signals.
Viterbi Algorithm for Decoding
• Purpose: Find the most likely input bit sequence given
a received noisy encoded bitstream.
• Steps:
• 1. Build a trellis diagram representing all possible
state transitions.
• 2. Compute branch metrics using Hamming distances.
• 3. Maintain path metrics for each state.
• 4. Use traceback to find the most probable input
sequence.
Viterbi Algorithm – Visual Example
Summary
• • Convolutional network coding enhances
communication in dynamic and cyclic
networks.
• • Decoding relies on sufficient maxflow and
decoding matrices.
• • Viterbi algorithm provides a reliable method
to decode convolutional codes.

You might also like