Calculation of EM fields induced by an antenna in the near field

[DC2]

The title pretty much covers it. I'm having to calculate the field induced inside the human body by an antenna in the near field (essentially, a phone placed close to a user's head), and I'm drawing a blank on how to relate the field generated by the antenna to the field induced inside the tissue. A first instinct would be to quite simply divide it by the relative permittivity of the tissue, as if it were a static field, but a static field it is not. Additionally, this would not account for the fact that the spatial distribution of the field would be altered by the presence of the body. What am I missing here?

Thanks in advance.

 

[tech99]

The body tissues differ in electrical properties - permittivity and conductivity - so the absorbed dose varies greatly. I suggest looking up the many papers on this issue, in particular the international guidelines published by the International Committee for Non Ionising Radiation Protection (ICNIRP), https://www.icnirp.org/cms/upload/publications/ICNIRPrfgdl2020.pdf
Not only is the body exposed to the energy which is radiated, but when the antenna is very close to the body the tissues are also exposed to the energy stored in the reactive near fields of the antenna. These electric and magnetic fields are difficult to calculate. My general view is that it is a very specialised area of engineering, especially if human health and safety are involved.

 

[DC2]

The body tissues differ in electrical properties - permittivity and conductivity - so the absorbed dose varies greatly. I suggest looking up the many papers on this issue, in particular the international guidelines published by the International Committee for Non Ionising Radiation Protection (ICNIRP), https://www.icnirp.org/cms/upload/publications/ICNIRPrfgdl2020.pdf
Not only is the body exposed to the energy which is radiated, but when the antenna is very close to the body the tissues are also exposed to the energy stored in the reactive near fields of the antenna. These electric and magnetic fields are difficult to calculate. My general view is that it is a very specialised area of engineering, especially if human health and safety are involved.

My problem, if you can call it that, with the papers on the subject of RF dosimetry in general, is that the authors use professional softwares (E.G. Semcad X) to which I don't have access, to calculate these kinds of fields. The fields in the radiated and reactive near-fields con be calculated with matlab, but the problem of how to model them inside the tissue remains.

 

[tech99]

Maybe this is like a mutual impedance problem when we have two antennas closely spaced.

 

[DC2]

I'm not sure, because, supposing that we have only one antenna driven (let's call it 1), and another (2) which is open circuit, they are related by Z₂₁=V₂₁/I₁, where Z₂₁ is the mutual impedance and V₂₁ is the open circuit voltage induced on 2. Now, if I knew Z₂₁ that'd be fine, but I would not even know where to begin to calculate it.

 

[berkeman]

I'm not sure, because, supposing that we have only one antenna driven (let's call it 1), and another (2) which is open circuit, they are related by Z₂₁=V₂₁/I₁, where Z₂₁ is the mutual impedance and V₂₁ is the open circuit voltage induced on 2. Now, if I knew Z₂₁ that'd be fine, but I would not even know where to begin to calculate it.

If the second antenna is open circuit, who cares? How can it affect any EM propagation other than via parasitic reflection?

 

[tech99]

If the second antenna is open circuit, who cares? How can it affect any EM propagation other than via parasitic reflection?

The second antenna has losses because it has conductivity. It also has permittivity. I think the approach may be to consider a two planar layer model of bone and fat, then impose 1 Amp on the antenna. Treat the radiated and reactive components entering the material separately. Energy reflected by the material will then alter the antenna feedpoint voltage. Please don't ask me to do this!

 

[tech99]

If you are willing to space the antenna slightly from the head you are outside the reactive near field and then need only consider the radiated energy. For a spacing of lambda/ 2 pi at 900 MHz this requires 5cm.

 

[berkeman]

I'm having to calculate the field induced inside the human body by an antenna in the near field (essentially, a phone placed close to a user's head)

the authors use professional softwares (E.G. Semcad X) to which I don't have access, to calculate these kinds of fields.

Can you say more about why you "have" to do these calculations, and why you are not able to use COMSOL or other software packages to do these calculations? Do you just need an order of magnitude approximate answer? What is your assignment, and who gave it to you?

 

[DC2]

Sorry for the long absence, I have been quite busy recently.

Can you say more about why you "have" to do these calculations, and why you are not able to use COMSOL or other software packages to do these calculations? Do you just need an order of magnitude approximate answer? What is your assignment, and who gave it to you?

The short version is, I am doing a reserch internship / thesis, and my advisor insists that I use MATLAB for the calculations. Additionally, the university has not given me an institutional email address, so also using that to access the student versions of those professional software packages is out of the question.

If you are willing to space the antenna slightly from the head you are outside the reactive near field and then need only consider the radiated energy. For a spacing of lambda/ 2 pi at 900 MHz this requires 5cm.

That could be a sort of last option. Making it a far-field problem would certainly make things much easier, but it would not really be a realistic scenario. After all, no one makes a phone call with their phone three fingers away from their ear.

I should have been less stingy with the details in my original question, sorry about that.

 

[tech99]

If we place a dipole terminated with a 75 Ohm resistor closer then approx lambda/2pi to a transmitting dipole, it receives half the transmitted power, irrespective of exact distance. The other half is radiated away. Are we saying that brain tissue could, in principle, couple more power from the transmitter than this if placed within the reactive near field region?

 

[tech99]

If we place a dipole terminated with a 75 Ohm resistor closer then approx lambda/2pi to a transmitting dipole, it receives half the transmitted power, irrespective of exact distance. The other half is radiated away. Are we saying that brain tissue could, in principle, couple more power from the transmitter than this if placed within the reactive near field region?

After some thought, I have the answer to my own question. Suppose brain tissue is a lossy dielectric, then it will not couple as much power from a radiation field as a matched antenna. However, the reactive near field of an antenna can have an electric field which is much stronger than the radiated field, and this could make a lossy dielectric absorb more power. Typically this will be near the end of a monopole.