Phase encoding

The multisource technique with phase encoding (Romero et al., 2000) is used in all the numerical tests to increase computation efficiency. With a fixed-spread acquisition geometry, single shot gathers are encoded and blended together to form supergathers. For land surveys with a rolling spread of receivers, only these shots that share the same receivers should be encoded together. Then, all the forward modeling and migration operations are applied to the encoded supergathers instead of single shot gathers. The encoding functions used in this chapter are the combination of source-side random time shifts and random source polarities. There are several encoding strategies. The static encoding keeps the encoding function to be the same for all the iterations, while a purely dynamic strategy that changes the phase encoding function at each iteration. The hybrid strategy uses static encoding for a number of iterations to reduce I/O costs, and then dynamically changes the encoding function at every $N$ -th iteration, where $N$ is greater than 5.