Leif Hanlen

Propagation Models for Body-Area Networks: A Survey and New Outlook

This article is a review of wireless body-area network (BAN) channel models, with observations about the selection of the best channel model in terms of both first- and second-order statistics. Particular insight into the dominant factors that affect propagation for body-area networks is given. Important second-order statistical measures are discussed, where coherence times and fade durations are of particular interest. The IEEE 802.15.6 standard is used as a basis for the review, with observations and insights given about body-area networks. In this context, narrowband and ultra-wideband (UWB) models are summarized for different measurement environments and carrier frequencies. On-body, in-body, and off-body propagation models are discussed where appropriate. In general, lognormal fading or gamma fading models of the body-area network channel are most applicable. A goodness-of-fit criterion that directly trades off model error and complexity is presented, which gives a new outlook for channel modeling. By this new outlook it is demonstrated that through significant simplification of individual link propagation models for body-area networks, it is possible to combine link models with only a few parameters. Common misconceptions regarding the appropriateness of applying traditional path-loss measures to these short-range networks are then exposed. Finally, the use of relays, which is an option in IEEE 802.15.6, is shown to be important for maintaining reliability in various body-area-network propagation scenarios.

First- and second-order statistical characterizations of the dynamic body area propagation channel of various bandwidths

Comprehensive statistical characterizations of the dynamic narrowband on-body area and on-body to off-body area channels are presented. These characterizations are based on real-time measurements of the time domain channel response at carrier frequencies near the 900- and 2,400-MHz industrial, scientific, and medical bands and at a carrier frequency near the 402-MHz medical implant communications band. We consider varying amounts of body movement, numerous transmit–receive pair locations on the human body, and various bandwidths. We also consider long periods, i.e., hours of everyday activity (predominantly indoor scenarios), for on-body channel characterization. Various adult human test subjects are used. It is shown, by applying the Akaike information criterion, that the Weibull and Gamma distributions generally fit agglomerates of received signal amplitude data and that in various individual cases the Lognormal distribution provides a good fit. We also characterize fade duration and fade depth with direct matching to second-order temporal statistics. These first- and second-order characterizations have important utility in the design and evaluation of body area communications systems.

Open-source testbed for Body Area Networks: 200 sample/sec, 12 hrs continuous measurement

We present the design criteria and specifications of a novel Open-Source hardware channel sounder and Open-Source data sets for measurements of the Body Area Channel at the 2400MHz ISM band and 2360MHz band. We outline a need for open hardware and measurement data to facilitate robust standardization of the new Body Area Networks. We demonstrate typical analyses on a public data set, with reference to previous works, and show how complex network topologies may be simulated through simple real measurements using reciprocity.

Statistical characterization of the dynamic narrowband body area channel

A statistical characterization of the narrowband dynamic human on-body area channel, with application to biomedical/health information monitoring, is presented based on measured received signal amplitude. We consider varying amounts of body movement, and a variety of transmit-receive pair (Tx-Rx) locations on the human body. The characterization is presented for two different frequencies, near the 900 MHz and 2400 MHz Industrial, Scientific and Medical (ISM) bands. Common distributions used to describe fading statistics are compared to the received signal component for nine different Tx-Rx pair locations for the subjectpsilas body standing, walking and running. The Lognormal distribution provides a good fitting model, particularly when the subjectpsilas body is moving. The fit is independent of Tx-Rx pair locations. The Rayleigh distribution is a very poor fit to the received signal amplitude statistics.

Stability of Narrowband Dynamic Body Area Channel

The stability of a dynamic narrowband on-body area channel is characterized based on real-time measurements of the time domain channel impulse response (CIR) at frequencies near the 900- and 2400-MHz industrial, scientific, and medical (ISM) bands. A new parameter, channel variation factor, characterizes channel coherence time. Body movement is considered at various transmit-receive pair locations on the human body. Movement has considerable impact on the stability of the channel, a reasonable assumption for coherence time is approximately 10 ms, and there is greater temporal stability at the lower frequency.

Performance of Piconet Co-Existence Schemes in Wireless Body Area Networks

Coexistence of multiple wireless body area networks (WBAN) is a very challenging problem because each piconet can have a large number of sensors and their movement is unpredictable. Moreover, suitable global coordination schemes do not exist as there is no natural choice of coordinator between piconets. Adaptive schemes that work well with low-occupancy channels, such as listen before transmit, are not a wise global solution because of the potential for high levels of traffic in any one area [1]. In this paper we investigate the performance of three classic multiple-access schemes - namely TDMA, FDMA and CDMA - for (inter-network) piconet coexistence. We first consider a theoretical analysis of these schemes and then simulate each scheme using real-world interference measurements. It is found that co-channel interference could significantly degrade system performance if left unchecked, and that TDMA and FDMA are better choices than CDMA in terms of co-channel interference mitigation.

Simple Prediction-Based Power Control for the On-Body Area Communications Channel

Methods for transmit power control based on simple long-term channel prediction for the general body-area communications channel are presented. The power control methods are based on large sets of empirical every-day activity data. Numerous transmit-receive pair (Tx-Rx) locations on the human body, i.e. on-body, for a typical body-area-network (BAN) are considered. With the use of a simple prediction method based on held samples, and an enhanced held simple prediction method that uses short term mean path loss with the held sample, optimal power allocation for long-term transmit power control is described. When tested, according to the draft IEEE 802.15.6 BAN radio standard, on empirical data, both power allocation methods are shown to be more reliable, and also more energy efficient in terms of transmit circuit power consumption, than systems that use typical set Tx power levels for BAN.

Characterization of the Body-Area Propagation Channel for Monitoring a Subject Sleeping

A dynamic characterization of the wireless body-area communication channel for monitoring a sleeping person is presented. The characterization uses measurements near the 2.4 GHz ISM band with measurements of eight adult subjects each over a period of at least 2 hours. Numerous transmit-receive pair (Tx-Rx) locations on and off the body for a typical body-area-network (BAN) are used. Three issues are addressed: 1) modeling of channel gain, 2) outage probability, and 3) outage duration. It is shown that over very large durations (far in excess of a delay requirement of 125 ms that is typical for many IEEE 802.15.6 medical BAN applications) there is not a reliable communications channel for star-topology BAN. The best case outage probability, with 0 dBm Tx power and - 100 dBm Rx sensitivity, is in excess of a packet-error-rate of 10%. Following from these issues the feasibility of using alternate on-body or off-body links as relays is demonstrated.

Transmit power control for wireless body area networks using novel channel prediction

We present a predictor for real Body-Area-Network (BAN) channels that is accurate for up to 2 seconds, even with a nominal channel coherence time of 500 ms. The predictor utilizes the partial-periodicity of measured BAN channels using the previous 4 seconds of channel gain values. We demonstrate use of this predictor for power control with open-access and private channel measurements. When used under a realistic setting for IEEE 802.15.6, with packet loss less than 10%, we show that the accurate channel predictor does not translate into substantial reduction in packet loss or power usage over a simple sample-and-hold method, even though it is a more accurate predictor than sample-and-hold.

Capacity analysis of correlated mimo channels

This paper gives expressions for the capacity of ergodic multiple-input multiple-output channels with finite dimensions, in which the channel gains have a correlated complex normal distribution and receivers experience independent Gaussian noise. The particular correlated normal distribution considered corresponds to flat Rayleigh fading with arbitrary transmit and receive correlation. Knowledge of the correlation matrices is assumed at both the transmitter and receiver, while the receiver, but not the transmitter, has complete knowledge of the channel realization. The optimal input density is characterized via a necessary and sufficient condition for optimality, along with an iterative algorithm for its numerical computation. The resulting capacity is expressed in terms of hypergeometric functions of matrix argument, which depend on the channel correlation matrices only through their eigenvalues. Some closed-form expressions are also given in the case of single-sided correlation. Some consideration is given to high- and low-power asymptotics. Easily computable asymptotic expressions are also given for receive-side only correlation in the case that the number of transmitters is large. In that case, the capacity can be divided into two components: one arising from the dominant eigenvalues of the receiver-end correlation matrix, and the other from the remaining spherically distributed eigenvalues. Some numerical results are also presented.