In recent years, algorithm based on Mathematical Morphology and wavelet transform has been proposed for ECG QRS Complex detection. However, its intensity of computation is high. This paper proposes the algorithm and hardware architecture for the whole system of QRS Complex detection based on Mathematical Morphology and Quadratic Spline wavelet transform, with implementation in Field Programmable Gate Array (FPGA). The system consists of Morphological filtering, Quadratic Spline wavelet transform and Modulus Maxima Pair Recognition modules. The parallel and pipelined architecture of system can operate in the maximum 35MHz with throughput of one sample per clock cycle. The QRS Complex detection accuracy for MIT/BIH arrhythmia database recordings and resource consumption are reported. The design is suitable for both batch processing of huge volume ECG data and real time applications for portable devices.