RSS Models for Respiration Rate Monitoring

Abstract

Received signal strength based respiration rate monitoring is emerging as an alternative non-contact technology. These systems make use of the radio measurements of short-range commodity wireless devices, which vary due to the inhalation and exhalation motion of a person. The success of respiration rate estimation using such measurements depends on the signal-tonoise ratio, which alters with properties of the person and with the measurement system. To date, no model has been presented that allows evaluation of different deployments or system configurations for successful breathing rate estimation. In this paper, a received signal strength model for respiration rate monitoring is introduced. It is shown that measurements in linear and logarithmic scale have the same functional form, and the same estimation techniques can be used in both cases. The implications of the model are validated under varying signal-to-noise ratio conditions using the performances of three estimators: batch frequency estimator, recursive Bayesian estimator, and model based estimator. The results are in coherence with the findings, and they imply that different estimators are advantageous in different signal-to-noise ratio regimes.

Advancements in Methods and Camera-Based Sensors for the Quantification of Respiration

“…Respiratory rate seems to be one the most attractive parameters using the non-contact methods and has been addressed in a variety of conferences and journal papers divided as follows: [ 19 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 ] (over the period 2010–2015), [ 9 , 11 , 12 , 29 , 32 , 42 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 53 , 55 , 61 , 69 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 ,…”

Movement Detection Using A Reciprocal Received Signal Strength Model

“…This difference can be observed by examining σ 2 i values given in Table 1, and as shown, they are notably lower and they deviate less for the proposed model. The results imply that the modeling assumptions of EDEV are reasonable for σ 2 0 but not for σ 2 1 . The second notable difference between the systems is the .…”

“…In the following, performance of EDEV is evaluated using the receiver operating characteristic (ROC) which is defined by σ 2 0 , σ 2 1 and L. In the evaluation, σ 2 0 = 0.29 and σ 2 1 = 1.14 are fixed while L is varied. It is to be noted that as the ratio σ 2 1 /σ 2 0 increases, the detection performance improves and vice versa. The resulting ROC curves are illustrated in Fig.…”

“…These applications leverage human-induced perturbations to propagation patterns of radio signals which causes a measurable change in the received signal. Since the changes in the measured signal are a function of the person's position, electrical properties and velocity [2], RF sensing enables a wide range of applications including but not limited to: localization [3], breathing monitoring [4], occupancy assessment [5] and crowd size estimation [6].…”

Unsupervised Learning in RSS-Based DFLT Using an EM Algorithm

“…For DFLT, the state of the system in two-dimensional Euclidean space can be defined as where and are the x - and y -coordinates, and the velocity components are denoted as and . This state representation is particularly suitable for DFLT because the position and velocity define the temporal and spectral properties of the RSS [ 39 ]. This state evolves at time k in accordance with where is the state transition matrix of the dynamic model and is Gaussian process noise.…”