To investigate the performance of HOSVD in estimating the system error, we simulated the data according to model 2. Poisson random variables are generated with conditional mean: , , and . We generated the system error , simulating real data; thus, we fitted a bi-variate function on a real image as Display Formulawith . We also set to be the standard Gaussian distribution centered at the uniformly distributed random indices . The minimum distance between is set to be greater than 6. Finally, is a step function with dwelling time distributed as an exponential random variable in each state. Precisely, 100 fluorophores intensity were simulated with mean 1000 and the system error with mean 400. The performance of our algorithm is evaluated by maximum error and the mean squared error between and at . We compared our algorithm with the commonly used local subtraction5 ( LS, refers to the size of the background) at , which can be specified as Display Formula is defined in Algorithm t3. The result is shown in Table 1. With 100 replications of simulation, the maximum error obtained by coupling Algorithms t1 and t2 was evaluated to be . As contrasted with the signal intensity of 1000, this indicated an error of about 1.3%.