Tail-biting Codes for Lattice Wiretap Coding

Abstract

The secrecy gain of Construction A lattices obtained by tail-biting rate 1/2 convolutional codes is studied to evaluate the secrecy performance of a lattice in a wiretap channel communication. The higher the secrecy gain, the more difficult for an eavesdropper to correctly guess the transmitted message. To explore the best potential secrecy gain, an exhaustive search is conducted, considering various constraint lengths and even code lengths up to 100. The methods' validity is verified by evaluating the minimum free distance and comparing the results with known findings. This investigation identifies outcomes close to the upper bound of unimodular lattices.