--- concepts.qbk	2010-03-30 21:23:27.000000000 +0300
+++ concepts.new	2010-03-30 21:21:37.000000000 +0300
@@ -238,6 +238,46 @@
 
 [endsect]
 
+[section Quasi-Random Number Generator]
+
+A quasi-random number generator is a __PseudoRandomNumberGenerator which
+provides a deterministic sequence of quasi-random numbers, based on some
+algorithm and internal state. [classref boost::random::niederreiter_base2
+Niederreiter Base 2] generator is an example of such a [prng quasi-random
+number generator]. The "quasi" modifier is used to denote more clearly that the
+values produced by such a generator are neither random nor pseudo-random, but
+they form a low discrepancy sequence. Such sequences share some properties of 
+random variables and in certain applications such as the quasi-Monte Carlo method 
+their lower discrepancy is an important advantage.
+
+[note Quasi-random sequences are known to give efficient numerical integration 
+rules in many Bayesian statistical problems where the posterior distribution can be 
+transformed into periodic functions on the n-dimensional hypercube.]
+
+Harold Niederreiter gives an extensive overview on random number generation
+and quasi-Monte Carlo methods in his book "Random number generation and
+quasi-Monte Carlo methods, Society for Industrial and Applied Mathematics, 1992". 
+
+In addition to the __PseudoRandomNumberGenerator requirements,
+a quasi-random number generator has some additional requirements. In the
+following table, `X` denotes a quasi-random number generator class, and `v` is 
+a const value of `X`.
+
+[table QuasiRandomNumberGenerator requirements
+  [[expression] [return type] [pre/post-condition]]
+  [[`X::dimension`] [std::size_t] [compile-time constant. If it exists, it is equal to v.dim()]]
+  [[`X::dim()`] [std::size_t] [the dimension of quasi-random domain. It must be no less than 1.
+                        The return value of this function shall not change during the lifetime of the object.]]
+ ]
+
+[note The `operator()` returns a successive element of an n-dimensional (n = `X::dimension`) vector 
+at each invocation. When all elements are exhausted, `operator()` begins anew with the starting
+element of a subsequent n-dimensional vector.]
+
+The class __niederreiter_base2 is a model for a quasi-random number generator.
+
+[endsect]
+
 [section Random Distribution]
 
 A random distribution produces random numbers distributed according to some