In this dissertation two methods for spatio-temporal encoding in medical ultrasound imaging are investigated. The first technique is based on a frequency division approach. Here, the available spectrum of the transducer is divided into a set of narrow bands. A waveform is designed for each band, making it possible to transmit with several transmitters simultaneously and separate the signals at the receiver using a simple filtering operation. The separated signals can therefore be processed independently of each other. This facilitates transmission of more acoustic energy, which increases the signal to noise ratio and simultaneously the penetration depth so that the medical doctor can image deeper lying structures. The method is tested both experimentally and in simulation and has also evaluated for the purpose of blood flow estimation. The work presented is based on four papers which are also included in the dissertation. The second method is based on encoding the transmitters with pseudo-random sequences. The signals can be separated after only one transmission using an estimation based technique. Broadband decoding can, therefore, be performed at the receiver after a single transmission. This increases the amount of data acquired in every transmission and can potentially reduce motion artifacts. The work presented utilizing this method is based on three papers which are included in the dissertation.