Many desert expressways are affected by the deposition of the wind-blown sand, which might block the movement of vehicles or cause accidents. W-beam central guardrails, which are used to improve the safety of desert expressways, are thought to influence the deposition of the wind-blown sand, but this has yet not to be studied adequately. To address this issue, we conducted a wind tunnel test to simulate and explore how the W-beam central guardrails affect the airflow, the wind-blown sand flux and the deposition of the wind-blown sand on desert expressways in sandy regions. The subgrade model is 3.5 cm high and 80.0 cm wide, with a bank slope ratio of 1:3. The W-beam central guardrails model is 3.7 cm high, which included a 1.4-cm-high W-beam and a 2.3-cm-high stand column. The wind velocity was measured by using pitot-static tubes placed at nine different heights (1, 2, 3, 5, 7, 10, 15, 30 and 50 cm) above the floor of the chamber. The vertical distribution of the wind-blown sand flux in the wind tunnel was measured by using the sand sampler, which was sectioned into 20 intervals. In addition, we measured the wind-blown sand flux in the field at K50 of the Bachu-Shache desert expressway in the Taklimakan Desert on 11 May 2016, by using a customized 78-cm-high gradient sand sampler for the sand flux structure test. Obstruction by the subgrade leads to the formation of two weak wind zones located at the foot of the windward slope and at the leeward slope of the subgrade, and the wind velocity on the leeward side weakens significantly. The W-beam central guardrails decrease the leeward wind velocity, whereas the velocity increases through the bottom gaps and over the top of the W-beam central guardrails. The vertical distribution of the wind-blown sand flux measured by wind tunnel follows neither a power-law nor an exponential function when affected by either the subgrade or the W-beam central guardrails. At 0.0H and 0.5H (where H=3.5 cm, which is the height of the subgrade), the sand transport is less at the 3 cm height from the subgrade surface than at the 1 and 5 cm heights as a result of obstruction by the W-beam central guardrails, and the maximum sand transportation occurs at the 5 cm height affected by the subgrade surface. The average saltation height in the presence of the W-beam central guardrails is greater than the subgrade height. The field test shows that the sand deposits on the overtaking lane leeward of the W-beam central guardrails and that the thickness of the deposited sand is determined by the difference in the sand mass transported between the inlet and outlet points, which is consistent with the position of the minimum wind velocity in the wind tunnel test. The results of this study could help us to understand the hazards of the wind-blown sand onto subgrade with the W-beam central guardrails.