A pseudorandom number generator (PRNG) is an algorithm that produces seemingly random number sequences. They are employed in applications requiring randomness such as arbitrary sample selection in statistical sampling and secret key generation for ciphers. Where unpredictability is a concern, a cryptographically secure PRNG (CSPRNG) is the only type of PRNGs suitable for such applications. CSPRNGs are specially designed to withstand security attacks. In this paper, after describing a well-known lightweight stream cipher Trivium, we present Quadrivium, a PRNG inspired by the design of Trivium. We compare the statistical properties of Quadrivium by that of Trivium using NIST Statistical Test Suite and Dieharder: A Random Number Test Suite. The analyses show that Quadrivium performs as well as Trivium and has the advantage of producing longer sequences of random bits.