Power density limits in different frequency bands?

I don't think that cell phone usage in airplanes is forbidden for interference problems anymore, in fact. I think it's almost entirely because no one wants to have to sit next to an annoying passenger who won't stop making annoying loud and telephone calls. I was pleased to see how the Japanese are perhaps the only polite people on the planet, when it comes to restraining themselves in making voice calls on cell phones.

(Of course, the more recent trend to text instead of using voice has largely eliminated one problem, and created a distraction problem instead.)

ok but what is the metrics should i use to measure the EMI effect@Aparna Sathya Murthy @ Albert Manfredi

I think mobile phones could disturbance a work other device(s). In my laboratory we can observe very often interesting phenomenon. Employee who test an electronic device and who is outside of anechoic chamber, where a test device is placed (the employee stayes close to receiving test system), his mobile phone has to be swiched off. If not, the mobile phone influances negative on test results. It disturbances a receiver.

The general EMI model is an EM source with an EM path to a victim.

There are four paths:

1. Conductive; current noise thru a shared impedance in a two conductor connections from source to victim. A conductive path does not depend on distance or frequency.

2. Inductive; a change in current (di/dt) inducing a voltage thru the mutual inductance of two current loops. An inductive path is near field (wavelength/2) and increases with frequency.

3. Capacitive; a change in voltage (dv/dt) induces a current thru a shared electric field of conductive areas. An capacitive path is near field (wavelength/2) and increases with frequency.

3. Capacitive; a change in voltage (dv/dt) induces a current thru a shared field of conductive areas.

4. Radiated; a source antenna radiates EM power that induces a current in a victim that contains an "accidental" antenna. An radiated path is far field (greater than wavelength/2) and increases with frequency.

The source path victim model provides a reliable means of visualizing EMI issues in signal circuits and systems.

Visualizing EMI requires to consider electronic signals of at any frequency as four dimensional (x, y, z and time) EM waves.

From an engineer's viewpoint, the relevant effects and applicable metrics depend upon the context of requirements and use cases. Requirements can include not only user-oriented requirements but also government and industry-imposed requirements. When developing products for multiple markets one often encounters a complex matrix of requirements as illustrated by the many approval marks seen on consumer products. And governments can vary in the primary philosophy behind their requirements - in the US it seems to be "Thou shall not interfere with other equipment" while in other locales "Thou shall not let other equipment interfere with your operation" seems more important. Industries will then add further requirements. As an example I recently was talking with an EMC (Electromagnetic Compatibility) engineer working for a company that produces DOCSIS-compliant TV set-top boxes. He had plenty of job security keeping up with ever-increasing DOCSIS EMI susceptibility requirements. DOCSIS is trying to address the user-visible issues that can arise from placing low-voltage high-density FPGA signal processing in the same box as the wideband RF front end (often fed with a high-power signal) as well as sitting next to large flat-screen TVs, Wi-Fi systems, Blu-Ray players, etc. Many of the DOCSIS requirements extend orders-of-magnitude beyond US FCC and CE regulations in both frequency and field strength.

Additionally there are the extreme examples of context-based requirements such as those for space systems, medical systems, etc.

A good starting point for viewing samples of metrics may be the CE (European) standards as mentioned here: https://en.wikipedia.org/wiki/List_of_common_EMC_test_standards#European_standards_concerning_unwanted_electrical_emissions