Fishing sonar utilizes sound waves to detect the distribution and activities of underwater fish schools. It typically employs either omni-directional transmission or directional scanning transmission methods for preliminary detection of the position of fish schools. However, there are several limitations with these traditional methods, including poor signal directivity in omni-directional transmission and uneven beam coverage in directional scanning transmission. To address these challenges, this study proposed an improved method based on rotating sector scanning, which aims to optimize the detection performance and meet the specific requirements of fishery sonar in practical fishing scenarios. This study used acoustic simulation methods to systematically analyze the factors that influence the scanning transmission performance, such as the number of sectors and the boundary angles between sectors. The performance of the proposed method was further validated through pool tests conducted under controlled laboratory conditions. The simulation results showed that the use of a four-sector rotating sector scanning transmission method effectively improves the signal directivity while maintaining a full 360° horizontal scan coverage. Specifically, the main lobe intensity was increased by 8 dB when compared to omni-directional transmission, and the maximum intensity difference within a single sector’s transmission beam was reduced to less than 1.5 dB, with the beamwidth difference kept within 2°. These results demonstrated the advantages of the proposed method in enhancing the detection capabilities of the sonar system. To further evaluate the practical performance of the system, laboratory pool tests were carried out to measure the transmission directivity between two adjacent sectors. The results from these tests showed that the proposed method improves the directivity by approximately 8 dB on average over a 64° range in front of the two sectors compared to the omnidirectional transmission. This finding suggests that the proposed design not only enhances the efficiency and accuracy of fish school detection but also helps to obtain accurate fish information in a wider range.