Add whole decoding line to max_iter plot

src/thesis/chapters/4_decoding_under_dems.tex

@@ -1167,7 +1167,13 @@ reimplementation in Rust to achieve higher simulation speeds due to
 the compiled nature of the language.
 We reimplemented both the window splitting and the decoders themselves.
 
-% Simulation setup
+% Global experimental setup
+%   - Code
+%   - # SE rounds
+%   - Noise model
+%   - Per-round LER as figure of merit
+%   - Detector definition
+%   - # simulated error frames
 
 We chose to carry out our simulations on \ac{bb} codes, as they have
 recently emerged as particularly promising candidates for practical
@@ -1194,13 +1200,60 @@ generated by simulating at least $200$ logical error events.
 \subsection{Belief Propagation}
 \label{subsec:Belief Propagation}
 
-% Intro
+% Local experimental setup
+%   - BP variant
 
-We begin our investigation by using \ac{bp} with no further
+We began our investigation by using \ac{bp} with no further
 modifications as the inner decoder.
 We chose the min-sum variant of \ac{bp} due to its low computational complexity.
 
-% Whole decoding as a lower bound on the error rate
+% [Thread] Get impression for max gain
+%   - More global = better -> Compare windowed vs. whole
+
+% [Description] Figure 4.8
+%   - Parameters
+%     - # BP iterations
+%     - W,F
+%     - Physical error rates
+%     - Windowed (cold start) vs whole decoding
+%     - (?) Semilog y axis
+%   - Figure description
+%     - TODO:
+
+% [Interpretation] Figure 4.8
+%   - Larger window -> better, because more global decoding
+%   - Diminishing returns as the window becomes larger
+%   - As expected, whole works best
+
+% [Thread] First comparison with warm start
+%    - Compare performance of warm start to cold start
+
+% [Description] Figure 4.9
+%   - Parameters
+%     - # BP iterations
+%     - W,F
+%     - Physical error rates
+%     - Warm vs cold start
+%   - Figure description
+%     - TODO:
+
+% [Interpretation] Figure 4.9
+%   - Generally better performance with warm start, as expected
+%   - It is surprising that warm start performs better than whole
+
+% [Thread] Warm start is better than whole due to more effective iterations
+
+% [Description] Figure 4.10
+%   - Parameters
+%     - # BP iterations
+%     - W,F
+%     - Physical error rates
+%     - Warm vs cold start
+%   - Figure description
+%     - TODO:
+
+% [Interpretation] Figure 4.10
+%   -
 
 We initially wanted to gain an impression for the performance gain we could
 expect from a modification to the sliding-window decoding procedure.
@@ -1529,6 +1582,15 @@ though the process is less global.
 
                     \addlegendentryexpanded{$W = \W$}
                 }
+
+                \addplot+[mark=*, solid, mark options={fill=black}, black]
+                table[
+                    col sep=comma, x=max_iter,
+                    y=LER_per_round,
+                ]
+                {res/sim/max_iter/SyndromeMinSumDecoder/p_0.0025/LERs.csv};
+
+                \addlegendentry{Whole}
             \end{axis}
         \end{tikzpicture}