# Channel modelling for wireless body area network

According to IEEE 802.15.4a

In WBAN, pulses are transmitted from an antenna , diffract around the body and can reflect off of arms and shoulders. so there are always two clusters of multi path components due to the initial wave diffracting around the body, and a reflection off of the ground.

I need to work with ZigBee (Narrow band data rate), Can I ignore the effect of reflection components in describing narrowband channel in small scale Statistics?

And take only the effect of diffraction?

IEEE 802.15.4a defines multiple bands from 2.4 to 10 GHz for operation. You should focus on one. Not sure you're actually looking for diffraction and not more for backscatter from surrounding objects and the ground.

I'd recommend not trying to come up with a physical channel model first. Do measurements, find a statistical model that works reasonably well. That might give research of what you'd consider physically a direction (and is comparatively quickly done properly).

I don't know what literature you've been reading so far, but "WBAN channel model" certainly leads to a lot of things that you can find via google - I've randomly picked

"Channel Modeling for Wireless Body Area Networks" by David B. Smith and Leif W. Hanlen, in: P.P. Mercier, A.P. Chandrakasan (eds.), Ultra-Low-Power Short-Range Radios, Integrated Circuits and Systems, Springer International Publishing Switzerland 2015, DOI 10.1007/978-3-319-14714-7_2, available online

and that proposes a normal or lognormal receive power distribution.

Also, now is a good time to consider why you're using 802.15.4, which is a Machine-to-Machine standard for relatively wide area coverage instead of 802.15.6, which is a standard designed for WBANs. Narrow range becomes a feature as soon as you consider that WBAN devices would ideally be deployed in relatively dense environments (imagine a train full of people) and you get interference-limited very early on.

I provide you a quick note. IEEE 802.15.4 is for wireless personal area networking (WPAN). As also pointed out in Marcus's answer, the IEEE standard for WBAN is 802.15.6 which uses three different PHY technologies (including a narrow-band PHY) with a sophisticated MAC protocol. But notice that the default PHY mode in this standard is UWB.

Due to the short distance, ground reflection has a considerable contribution. However, it depends on the Tx-RX orientation. For instance, if they are on opposite sides of the body, then there would be a rather strong ground reflection. About fading, there are issues that are far more complicated than only diffraction. To summarize, there are refractive, dispersive, and reflective effects caused by factors such as randomness associated with different body parts (tissue structure, fat, water density, etc.), metal parts in cloths, ornaments, metal implants, and many other effects caused by close proximity to the body. Most importantly moving body parts cause fading effects.

There are some channel models that are usually used in literature such as in-body channel model CM-3, or on-body channel model CM-4 which includes four different angles of Tx-Rx orientation. There are also channel models specific to locations such as hospital or work environments. By a quick search you may find Matlab implementation of such models easily. It is worth noting that because of all mentioned practical issues, WBAN channel modeling is a major task and an enormous body of research on this topic exists.

## Paper say :section 5-b. Small Scale Statistics

The small scale statistics are characterized by fitting the maximum received energy of the first diffracted cluster around the body in the first bin at the 49 location measurements to Nakagami-m distribution. In this simulation the Nakagami distribution has two parameters, $m$ and $\omega$. The $m$ parameter is normally distributed with mean and variance value shown in Table II. Respectively, the $\omega$ is equal to the maximum energy of the first cluster. NAKAGAMI Probability Density function of the maximum received energy of the first reflected cluster around body in the first bin at the 49 location is depicted in Fig. 14. In this regard, Fig. 15 indicates the fading channel complex impulse response at 49 different location in an exemplary indoor environment simulated in this section. Furthermore, the Ricean K factors for 49 locations in the room are computed and demonstrated in Fig. 16.

link of paper : http://publish.uwo.ca/~edolatab/1.pdf