We review the status of bouncing cosmologies as alternatives to cosmological inflation for providing a description of the very early universe, and a source for the cosmological perturbations which are observed today. We focus on the motivation for considering bouncing cosmologies, the origin of fluctuations in these models, and the challenges which various implementations face.
I. MOTIVATIONThe inflationary scenario [1] is the current paradigm of early universe cosmology. Inflation solves several problems of Standard Big Bang cosmology, and it gives rise to a causal theory of structure formation [2] (see also [3]) which made a number of predictions for cosmological observations which were subsequently successfully verified. In spite of the phenomenological success, inflation faces a number of conceptual challenges (see e.g. [4] for a review of these problems) which motivate the exploration of alternative early universe scenarios. Before mentioning some of these challenges we must begin with a lightning review of inflationary Universe cosmology.According to the inflationary scenario, the universe underwent a period of almost exponential expansion at some very early time. As a consequence, the horizon expanded exponentially and became larger than our past light cone -both evaluated at the time of recombination -provided that the period of accelerated expansion was sufficiently long. During this period, spatial curvature was also diluted. Any wavelength of fluctuation was stretched quasi-exponentially during the period of inflation so that the wavelengths corresponding to scales which are being observed today in cosmological experiments were smaller than the Hubble radius H −1 (t) at the beginning of inflation, where H(t) is the expansion rate of space. The space-time geometry of inflationary cosmology is sketched in Fig. 1. In this figure, the vertical axis is time t, with t = t i denoting the beginning of the inflationary phase, and t = t R the end; the horizontal axis represents physical spatial distance. The Hubble radius is almost constant between t i and t R , and increases linearly before t i and after t R . The horizon is shown as the dashed curve which equals the Hubble radius at the beginning of the period of inflation but increases exponentially until t R . The curve labeled λ indicates the physical wavelength of a cosmological fluctuation mode. [5,6] for reviews of the theory of cosmological perturbations). One finds that the induced spectrum of fluctuations is approximately scaleinvariant and that the observed amplitude of fluctuations is achieved if the Hubble expansion rate during the period of inflation was of the order H ∼ 10 13 GeV, which corresponds to an energy density during the inflationary period which is of the order η ∼ 10 16 GeV, the scale of particle physics "Grand Unification". With this value of H, it turns out that the period of accelerated expansion has to last for at least 50 e-foldings in order for inflation to be able to solve the horizon and flatness problems.One key challenge fo...