If you have found the mean value, μ, and standard deviation, s, of a sample data set, and the data is distributed like a normal (Gaussian) distribution, then the confidence interval can be expressed as the interval between μ-z*s/√n and μ+z*s/√n, where n is the number of data points in your sample and z is a number that depends on the degree of confidence you require. For example, if you want a 95% confidence interval then z is 1.96.
See http://en.wikipedia.org/wiki/Confidence_interval, for example, for more details.