The invention discloses a channel state selection method and device based on ternary coding. The method includes the steps that a sound source signal is modulated into (2L+1) electric-level-grade quantitative signals x, mapping transformation is performed on the quantitative signals x so that control signals p, m and z can be generated, selection processing is performed on the control signals p, m and z and an M channel feedback signal b, then, an M channel state vector signal y is output, shape-correction processing is performed on the state vector signal y, then, the M channel feedback signal b is generated, and the state vector signal y completes electric-acoustic conversion through a multi-channel power amplifier and an energy converter. The device comprises a modulator module, a mapping module, a selection module, a shape-correction module, the multi-channel digital power amplifier and a loudspeaker array or a multi-voice-coil loudspeaker unit which are all connected in sequence. Through the method and device, hardware resource expenditure can be effectively reduced, hardware electric power consumption can be reduced, system stability is enhanced, the amplitude of an output signal can further be improved, conversion efficiency is improved, and meanwhile high signal-to-noise ratio output capacity is achieved.