A Power Sensor Network for Identification of Heterogeneous Wireless Networks

We consider a scenario with multiple radio sources operating in the 2.4GHz ISM band and emitting signals that may overlap in time and frequency. Each source is characterized by an on/off signal, which represents its activity in time. Sources are modelled as statistically independent if they belong to different networks or statistically dependent if they belong to the same network, in which case they produce signals that do not overlap in time. A network of sensors performs measurements, where each sensor measures average received power with some time granularity. We show how source activity signals can be recovered from the power measurements using blind signal separation techniques. Recovered signals are partitioned into groups where each group is formed of non overlapping signals in time that belong to the same network.

[1]  Arogyaswami Paulraj,et al.  Blind separation of synchronous co-channel digital signals using an antenna array. I. Algorithms , 1996, IEEE Trans. Signal Process..

[2]  Eric Moulines,et al.  A blind source separation technique using second-order statistics , 1997, IEEE Trans. Signal Process..

[3]  Jaap C. Haartsen,et al.  The Bluetooth radio system , 2000, IEEE Personal Communications.

[4]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[5]  J. Cardoso,et al.  Blind beamforming for non-gaussian signals , 1993 .

[6]  Jaap Haartsen.  Bluetooth Radio System , 2003 .

[7]  Karim Abed-Meraim,et al.  Algorithms for joint block diagonalization , 2004, 2004 12th European Signal Processing Conference.