The spectrum for cell phones is typically in the 850 MHz range. E=hv yields around 3.515E-6 eV. The lowest natural state electron binding energy is on the order of 10 eV >> 3.515E-6. Is this enough to dispel the "correlation"? Are there any other particle interaction mechanisms that would allow for ionization to take place with more significant quantum mechanical probabilities? Granted, the exact mechanism for cancer propagation is not known.