At　present,　how　to　extract　fetal　ECG　from　maternal　abdominal　signal　is　a　hot　issue.　This　study　is　based　on　wavelet　transform　modulus　maximum　algorithm　for　extracting　fetal　ECG.　Maternal　ECG　and　fetal　ECG　have　different　characteristics　in　the　wavelet　transform.　First,　the　wavelet　decomposition　and　modulus　maximum　detection　was　applied　to　the　maternal　abdominal　signal.　Detection　results　were　processed　to　inhibit　some　of　the　maximum　points　associated　with　the　fetal　ECG　and　remove　the　corresponding　signal　singularity.　The　wavelet　coefficients　were　reconstructed　from　the　processed　modulus　maxima　using　Mallat　alternative　projection　algorithm,　and　then　the　signal　was　reconstructed.　Maternal　ECG　and　fetal　ECG　were　distinguished　from　maternal　abdominal　signal.　Clinical　data　were　used　to　test　the　method.　The　results　indicate　that　the　wavelet　transform　modulus　maximum　algorithm　based　on　single-channel　signal　can　identify　fetal　ECG　from　maternal　abdominal　signal.　Compared　with　the　traditional　extraction　methods　based　on　multi-channel　signal,　this　method　is　not　affected　by　correlation　of　multi-channel　signals.　But　it　also　has　some　disadvantages　waiting　for　further　research.