Cross-layer optimized routing with low duty cycle TDMA across multiple wireless body area networks

Abstract

In this paper, we study the performance of two cross-layer optimized dynamic routing techniques for radio interference mitigation across multiple coexisting wireless body area networks (BANs), based on real-life measurements. At the network layer, the best route is selected according to channel state information from the physical layer, associated with low duty cycle TDMA at the MAC layer. The routing techniques (i.e., shortest path routing (SPR), and novel cooperative multipath routing (CMR) incorporating 3-branch selection combining) perform real-time and reliable data transfer across BANs operating near the 2.4 GHz ISM band. An open-access experimental dataset of 'everyday' mixed-activities is used for analyzing the proposed cross-layer optimization. We show that CMR gains up to 14 dB improvement with 8.3% TDMA duty cycle, and even 10 dB improvement with 0.2% TDMA duty cycle over SPR, at 10% outage probability at a realistic signal-to-interferenceplus-noise ratio (SINR). Acceptable packet delivery ratios (PDR) and spectral efficiencies are obtained from SPR and CMR with reasonably sensitive receivers across a range of TDMA low duty cycles, with up to 9 dB improvement of CMR over SPR at 90% PDR. The distribution fits for received SINR through routing are also derived and validated with theoretical analysis.

Experimental Analysis of Cross-Layer Optimization for Distributed Wireless Body-to-Body Networks

“…Cooperative multi-path routing [43] yields better performance than single-path routing by providing simultaneous parallel transmissions with load balancing over available resources. We proposed a new CMR scheme in [44] for coordinated BANs, that uses two different paths (incorporating shortest path routing) which are combined at the destination. Also, it uses TDMA with low duty cycling to save energy consumption and avoid interference from surrounding noncoordinated BANs.…”

“…However, reducing duty cycle or active period often increases the overall delay and applying scheduling techniques without global coordination incurs extra overheads. In this paper, we investigate the performance of CMR [44] without any multiple access scheme or interference mitigation, by performing cross-layer optimization between PHY and Network layers for distributed BBN (or co-located BANs).…”

Cross-Layer Designs for Body-to-Body Networks: Adaptive CSMA/CA with Distributed Routing

“…An illustration of the two tiered architecture with 4 coexisting BANs (with fitted onbody hubs and sensors) is given in Fig.1. Dynamic routing is performed at the network layer in a cross-layered approach using the authors proposed routing techniques, i.e., SPR and CMR [12], that utilize and interact with the physical layer. Therefore, changes in channel states detected at the physical layer are passed on to the network layer, so routes with the most favorable channel conditions are chosen.…”

“…In Fig. 3, we also compare the CSMA/CA approach with a coordinated TDMA approach from [12], where the same setup is used with 4 coordinated BANs receiving interference from 6 non-coordinated nearby BANs. It is shown that adaptive CSMA/CA provides up to 20% and 30% improvement over TDMA (with a higher duty cycle of 8.3%), in terms of average throughput and packet arrival rate, respectively.…”

Wide-Sense-Stationarity of Everyday Wireless Channels for Body-to-Body Networks

“…Wireless body-to-body networks (BBNs) can enable coexistence of wireless body area networks (BANs) by exploiting body-to-body (B2B) communications using wearable on-body hub/sensor devices. While BANs are specifically designed to collect data from various sensors placed on/inside or around the human body, BBNs send data through closely located BANs to reach the intended destination/server in case of unavailable or out-of-range network infrastructure (in emergency indoor/outdoor situations) [1]. BBNs are envisioned to be selforganizing, smart and mobile networks that can create their own centralized/decentralized network connection without any external coordination.…”