Long-distance anti-interference self-feedback regulating fire emergency speaker system

Abstract Modern building groups are getting larger and larger, the lines longer longer, which directly leads to two major problems. One is that volume of loudspeakers at end line cannot meet requirements normal use. If power increased, near will howl self-excite too much. Second, amplifier number be matched exactly. It urgent develop fire-fighting or emergency broadcasting systems can applied long-distance transmission, having strong anti-interference ability, realize automatic adjustment through feedback. This paper develops a set intelligent fire broadcast system based on distributed amplifier, following functions: it connect with other platforms big data platform, cross-platform, cross-system real-time playback system, solve problem terminal volume. By switching channels, control, etc. Using near-end strategy optimization algorithm tailoring, optimal joint interference avoidance obtained environmental noise interference. Based deep reinforcement learning algorithm, matching speakers solved.