1: \begin{abstract}  

2: This work studies inference-based resource allocation 

3: in ultra low-power, large-scale backscatter

4: sensor networks (BSNs). Several  

5:    ultra-low cost and power sensor devices (tags) are 

6: illuminated by a carrier and reflect the measured information towards a wireless

7: core that uses conventional Marconi radio technology.

8: The development of multi-cell BSNs requires 

9:   few  multi-antenna cores  and several low-cost scatter radio devices, targeting

10:   at maximum possible coverage.

11:  The average 

12: signal-to-interference-plus-noise  ratio (SINR)

13: of   maximum-ratio combining (MRC) and zero-forcing (ZF)  linear detectors is found and harnessed

14: for frequency sub-channel allocation at tags, exploiting  long-term SINR information. 

15: The  resource allocation   problem is formulated as an integer programming optimization problem and solved through

16: the Max-Sum message-passing algorithm. The proposed algorithm is fully parallelizable and adheres to simple

17: message-passing  update rules, requiring mainly addition and comparison operations. 

18: In addition, the convergence to the optimal solution is attained within very few iteration steps.

19:  Judicious simulation 

20: study  reveals that ZF detector is more suitable for large scale BSNs, capable to 

21: cancel out the intra-cell interference. It is also found that the proposed  

22:  algorithm offers at least an order of magnitude decrease  in execution time

23:  compared to conventional convex optimization methods.

24: \end{abstract}