Sleep is a vital part of every humans daily circadian rhythm. People can rest and recover their body and mind and live a more active and alert life with an appropriate amount of sleep. The current gold standard method for sleep analysis is polysomnography, but due to the complexity, it is not convenient to perform it regularly and it disrupts the normal sleep environment of the patient. This thesis presents a method of integrating two alternative measurements of sleep analysis for an improved analysis. Combining the motion detection of actigraphy and the cardiac parameters of ballistocardiography, a novel algorithm was developed to analyze sleep and wake states without interfering with the natural sleep cycle of the participant. Without interfering with the natural sleep environment, this system can be implemented for continuous monitoring and be used to evaluate daily sleep patterns to assess overall sleep quality and health over time. The experimental results demonstrate the effectiveness of the novel proposed algorithm in comparison with each device used separately in improving the sleep classification.