RSS-Based Cooperative Localization and Orientation Estimation Exploiting Antenna Directivity

Abstract

In this paper, we propose a factor-graph-based cooperative positioning algorithm that uses RSS radio measurements and accounts for the directivity of the antennas. This is achieved by modeling the directivity with a parametric antenna pattern and jointly estimating position and orientation of the agents. We propose two different approaches whereas the first one uses a continuous representation of the orientation state and the second one a discrete representation. We validate our proposed methods with simulations and measurements in a static sensor network with more than 900 agents in an indoor environment and show that the positioning accuracy can be improved significantly by considering the influence of orientations.

Indoor Positioning of Low-Cost Narrowband IoT Nodes: Evaluation of a TDoA Approach in a Retail Environment

“…Accompanying node-to-node RSS measurements were conducted simultaneously. Details are not covered in this work but are analogous to the measurement campaign described in [ 6 ].…”

“…However, due to severe multipath fading of the radio channel, the obtained results tend to be unreliable. Sophisticated algorithms and/or numerous reference nodes at known positions are required to allow for accurate positioning [ 6 ]. Furthermore, RSS quantization is often inaccurate and additional hardware features such as automatic gain control (AGC) result in non-monotonic readouts.…”

Target Localization in Wireless Sensor Networks Based on Received Signal Strength and Convex Relaxation

“…The predefined points are known as the anchors, while the target refers to the point whose position needs to be estimated. The common measurements are angle of arrival (AOA) [ 14 , 15 , 16 , 17 ], time of arrival (TOA) [ 18 , 19 , 20 ], time difference of arrival (TDOA) [ 21 , 22 ] and received signal strength (RSS) [ 23 , 24 , 25 , 26 , 27 , 28 , 29 ]. However, AOA requires antenna arrays.…”

Orientation Estimation using Wireless Device Radiation Patterns

“…The issues of radiation patterns have been explored extensively for wireless sensor networks (WSN) [18], where deliberately using a highly directional antenna is very advantageous in terms of reducing power consumption and RF interference. In these scenarios, it is often possible to model radiation patterns as a single Gaussian-shaped lobe or cone projecting outwards from the transmitter [19], [20], [21], [22], [17], [23]. However, when considering mobile consumer devices, such as smartphones, simple models for the pattern are no longer appropriate as they often use an omnidirectional antenna whose pattern is significantly affected by other hardware components packaged into the device.…”

“…There is also a particular type of cooperative localisation that is very popular in sensor networks [10], [23] where multiple devices being localised cooperate, the idea is that by communicating with other nodes that are also trying to determine their location, some additional information can be obtained. This type of localisation does have clear applications outside of sensor networks, for example, localisation in crowds or disaster recovery scenarios, but this is very distinct from our work where there is a single device to localise.…”