Deep Learning Data Privacy Protection Based on Homomorphic Encryption in AIoT

<table class="table-group" id="tab6"><tr><td><table class="table"><tr><td class="thead-hr" colspan="4"><hr/></td></tr><tr class="thead"><td class="align_left"> </td><td class="align_center">Our</td><td class="align_center">Cryptonet</td><td class="align_center">CryptoDL</td></tr><tr><td class="thead-hr" colspan="4"><hr/></td></tr><tr><td class="align_left">Single inference time</td><td class="align_center">11 s</td><td class="align_center">570 s</td><td class="align_center">320 s</td></tr><tr><td class="align_left">Activation function complexity</td><td class="align_center"><i>O</i>(1)</td><td class="align_center"><i>O</i>(<i>n</i>)</td><td class="align_center"><i>O</i>(<i>n</i>)</td></tr><tr class="table-tr"><td colspan="4"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Comparison of cloud single trip prediction time.</div>

Mobile Information Systems

tab6

Table 6

Table 6: Deep Learning Data Privacy Protection Based on Homomorphic Encryption in AIoT 