In this paper we introduce the quadratically constrained integer least-squares problem and show how the LAMBDA method can be used to solve it for the purpose of GNSS attitude determination. The integer least-squares principle with quadratic equality constraints (ILSQE) is used to formulate our cost function. The solution of the ILSQE problem is derived and it is shown how the solution can be computed efficiently and rigorously with a novel LAMBDA based method. Experimental results with various single frequency GPS receivers are given to show the effectiveness of the proposed method. The method is also compared with some current methods of GNSS attitude determination. Apart from its efficiency, the proposed method is shown to dramatically improve the success rates of integer ambiguity GNSS attitude resolution.