Blind Identification of Convolutional Encoder Parameters

<table>Recognition performances on different size of <svg height="15.8375" id="M412" style="vertical-align:-3.22282pt" version="1.1" viewbox="0 0 16.225 15.8375" width="16.225" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<g transform="matrix(.017,-0,0,-.017,.062,11.55)"><path d="M716 25v-25h-205l-207 313h-28v-196q0 -53 14 -69q16 -16 74 -23v-25h-338v25q60 8 73 22q15 15 15 73v431q0 58 -17 77q-16 19 -71 23v25h308q296 0 296 -172q0 -76 -45 -118q-42 -38 -112 -57l199 -282q15 -22 44 -22zM276 597v-252q103 1 143 31t40 118
q0 79 -30 113.5t-100 34.5q-29 0 -41 -10t-12 -35z" id="x1D411"></path></g>
<g transform="matrix(.012,-0,0,-.012,12.337,15.637)"><path d="M238 681l-124 -585q-7 -31 4 -31q10 0 37.5 18.5t49.5 41.5l16 -22q-40 -48 -89.5 -81.5t-76.5 -33.5q-42 0 -16 122l105 488q7 32 0 41t-39 9h-35l5 26q35 3 71 13t58 17.5t26 7.5q14 0 8 -31z" id="x1D459"></path></g>
</svg> for soft and hard decisions.</table>

The Scientific World Journal

Blind Identification of Convolutional Encoder Parameters

Figure 6