Abstract: ‍ ‍This paper investigated the secure communication for Integrated Sensing and Communication （ISAC） system assisted by Intelligent Reflecting Surface （IRS）. With the aid of the IRS， the base station desired to deliver messages to a legitimate user and detect a trusted target in the presence of an eavesdropping user. We aimed to jointly design the transmit beamforming vector of the base station and the reflection beamforming vector of IRS to maximize the secrecy rate， while satisfying the required signal-to-noise ratio （SNR） threshold for the sensing target， transmit power constraint and the unit-modulus constraint of the IRS reflection coefficient. However， the problem was non-convex due to the non-convex of the secrecy rate function. To tackle this issue， we proposed an effective algorithm based on the semidefinite relaxation （SDR） and an alternation optimization approach to solve this problem. The simulation results show that the system can not only provide effective target detection and ensure predefined SNR threshold， but also provide secure communication. At the same time， it proves that IRS can help to achieve higher secrecy rate compared with the scheme without IRS， which shows the advantages of deploying IRS in the Integrated Sensing and Communication system and the effectiveness of the proposed algorithm.