Wireless System Performance Analysis

Analysis of a Failing Clinical Decision Support Wireless System

Input:

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Algorithm:

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Output:

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Problem Definition

Correlation, Measurement and Experimental studies. Random assignment of RF/Air interface environmental conditions to subjects of the study.

Analysis of a failing Clinical Decision Support Wireless link considering the human factors that consistently influence link performance in the clinical setting.

In this project, the Customer`s Medical Heart Rate Monitoring System comprises of the following components.

  • A heart rate variability monitor strapped to the  patient`s chest. Acute Patients are immobilized in the ICU.
  • A wireless receiver subsystem is installed in the  monitoring room 30ft away from the bedside. The equipment vendor supplied  the equipment 5years ago, tested and verified that the transmitted R-R  intervals would display on the monitoring station.
  • A mobile App enhanced to provide mobile monitoring  capabilities.

Inputs

Inputs: The simulated link is built to allow collection of open source  link logs.

  • Tx Controller: A  Dell Inspiron 7559 wireless  interface card.
  • A decommissioned  phone with an activated  host controller interface(ADB) feature set [The Android Debug Bridge (ADB) tool is part of the Android Open Source  Project.(AOSP) Specification Page] ( https://source.android.com/setup/build/adb )
  • Wireshark  network Protocol cross sectional data Analyzer

Algorithm

R, python, SQL, feature extraction and engineering of Wireshark capture log data, logs parsed to extract relevant RF air interface parameters. Extrapolate availability, integrity and confidentiality (AIC Triad) parameters for analyzing end to end network performance.

Average MUS level for typical wireless LAN operating at 11 Mb/s turns out to be -76 dBm, for 54Mb/s it averages out to -62 dBm; for Bluetooth it is roughly -70 dBm. Since wireless LANs adapt the data rate to local enviroment conditions, the assumption is that the average MUS level for this wireless LAN device is about -68 dBm.

The system is based on a 17 dBm IEEE 802.11a transmitter operating at 5.6 GHz and a pathloss model that includes free space up to 8 m and -10 dB/octave: attenuation beyond the breakpoint. Assumptions: The receiver has a bandwidth of 20 MHz and associated noise floor of −101 dBm; the noise figure is  6 dB. The SNR required is about 7 dB for OFDM BPSK, coding 1/2. For 64QAM, 2/3 coding, SNR = 21 dB is used. Fading is not included in these figures.

Additional Link parameters:

  • Operating  frequency: 5.6 Ghz - 5.8Ghz range within ISM band
  • Channel info: 19  Channels separated by 1/2 MHz
  • Antenna gain: 0dB
  • Network  Topology: Point to point
  • Modulation  coding scheme: OFDM BPSK-1/2,16QAM,64QAM-2/3 different for BLE or BR/EDR

Outputs

The business technical goals and tradeoffs were considered in characterizing the system. This is a venerable method used for analyzing end to end network performance.

Availability: High availability system desired

Integrity: Ensure R-R interbeat intervals transmitted without structural variation

Confidentiality: Secure Protected Health Information (PHI) transmitted over the wireless link.

     

Project schematic diagram

Cardiac Event Monitoring System

Path loss link budget

Path loss computation. Algorithm defined to compute rssi, Rx sensitivity and associated air interface parameters, link margin and signal to noise ratio at the input of the detector.

Distribution of RxLev as a function of time

The confidence interval shown for the system parameter bellow assumes normal distribution. 95% confidence that the range which contains the true value of the estimated parameter over the duration of transmission, will have a receive signal level of -88dBm (rssi / Rxlev) at the input of the radio detector.

Received Power Density grouped by availability

On the average the link Rxlev power was concentrated about -88dBm during downtime.

Device sensitivity

Outputs: During normal transmission, device receiver sensitivity was normal, but some sensitivity drop contributed to a 36% drop in availability of the link.

  • The results show that capacity utilization is less than 50%.
  • Average RxLev is about -90dBm.
  • In order for the wireless LAN device to suffer very little loss of range, an interfering signal must be less than −101 dBm.
  • The receiver experiences desensitization possibly due to hardware failure.
  • Currently the system availability can  only be guaranteed at 99.759% of the time.
  • Due to HIPAA confidentiality requirements, the recommendation is to discontinue the transmission of personal health data (PHI) over the link that might violate HIPPA regulations.
Other projects: