Navigation with Cellular CDMA Signals J OE K HALIFE ,K IMIA S HAMAEI , AND Z AK K ASSAS M OTIVATION Global navigation satellite system (GNSS) is at the heart of autonomous vehicles navigation systems. However, GNSS signals are unreliable due to: • Severe attenuation in deep urban canyons • Intentional and/or unintentional jamming • Spoofing! A PPROACH :E XPLOIT SOP S Ambient signals of opportunity (SOPs) may en- hance and assist conventional navigation tech- niques. C HALLENGES • Unavailability of SOP models for navigation purposes • Unavailability of receiver architectures for navigation observables extraction • Unknown SOP emitters’ states (position and clock) • Less stable clocks than GNSS satellite vehicles C ELLULAR CDMA AS SOP S • Uses code division multiple access (CDMA), which is suitable for ranging • Abundant and free to use • Higher received power and bandwidth than GNSS C ELLULAR CDMA N AVIGATION S OFTWARE -D EFINED R ECEIVER We implemented a cellular CDMA software-defined radio in LabVIEW to optimally extract relevant timing and positioning information. Although these signals were intended for communications, we were able to model them in terms of navigation observables. The receiver has three stages: (a) Acquisition: Signals from different BTSs are identified and a coarse estimate of their corresponding code delay and Doppler frequency is obtained. (b) Tracking: These estimates are maintained and refined via tracking loops. (c) Decoding: The message transmitted by the BTS is decoded and relevant information is extracted. (a) (b) (c) (a) (b) (c) LabVIEW Virtual Instruments N AVIGATION F RAMEWORK The pseudorange measurement ρ is given by ρ = ||r r - r BTS || 2 + c · (δt r - δt BTS )+ v, r r : receiver position r BTS : BTS position δt r : receiver clock bias δt BTS : BTS clock bias v : measurement noise c: speed-of-light Central Navigating Mapping Receiver BTS i Receiver Database BTS 2 BTS 1 r BTS δt BTS E XPERIMENTAL D EMO GRID 1 2 3 4 5 6 7 Estimated Estimated BTS 3 BTS 1 BTS 2 Mapping Receiver Navigating Receiver Trajectory Trajectory Cellular CDMA GPS Mean Error: 5.51 m Standard Deviation: 4.01 m A CKNOWLEDGEMENT This work was supported in part by the Office of Naval Re- search (ONR) under Grant N00014-16-1-2305, and part by a grant from the National Center for Sustainable Transportation (NCST), supported by the U.S. Department of Transportation (USDOT) through the University Transportation Centers Program. R EFERENCES [1] J. J. Khalife, K. Shamaei, Z. M. Kassas. “A Software-Defined Receiver Architecture for Cellular CDMA-Based Naviga- tion,” IEEE/ION PLANS 2016, April 2016. Best student paper award.

JOE KHALIFE, KIMIA SHAMAEI, AND ZAK KASSAS MOTIVATION

Download PDF Report

Upload
vodan
View
280
Download
2

Embed Size (px)

Citation preview

Navigation with Cellular CDMA SignalsJOE KHALIFE, KIMIA SHAMAEI, AND ZAK KASSAS

MOTIVATION

Global navigation satellite system (GNSS) is at theheart of autonomous vehicles navigation systems.However, GNSS signals are unreliable due to:

• Severe attenuation in deep urban canyons

• Intentional and/or unintentional jamming

• Spoofing!

APPROACH: EXPLOIT SOPS

Ambient signals of opportunity (SOPs) may en-hance and assist conventional navigation tech-niques.

CHALLENGES

• Unavailability of SOP models for naviga-tion purposes

• Unavailability of receiver architectures fornavigation observables extraction

• Unknown SOP emitters’ states (position andclock)

• Less stable clocks than GNSS satellite vehicles

CELLULAR CDMA AS SOPS

• Uses code division multiple access (CDMA),which is suitable for ranging

• Abundant and free to use

• Higher received power and bandwidth thanGNSS

CELLULAR CDMA NAVIGATION SOFTWARE-DEFINED RECEIVER

We implemented a cellular CDMA software-defined radio in LabVIEW to optimally extract relevanttiming and positioning information. Although these signals were intended for communications, we wereable to model them in terms of navigation observables. The receiver has three stages:

(a) Acquisition: Signals from different BTSs are identified and a coarse estimate of their correspondingcode delay and Doppler frequency is obtained.(b) Tracking: These estimates are maintained and refined via tracking loops.(c) Decoding: The message transmitted by the BTS is decoded and relevant information is extracted.

(a) (b)

(c)

(a) (b) (c)

LabVIEW Virtual Instruments

NAVIGATION FRAMEWORK

The pseudorange measurement ρ is given by

ρ = ||rr − rBTS ||2 + c · (δtr − δtBTS) + v,

rr: receiver position rBTS : BTS position

δtr: receiver clock bias δtBTS : BTS clock bias

v: measurement noise c: speed-of-light

Central

Navigating

Mapping

Receiver

BTSi

Receiver

Database

BTS2

BTS1

rBTS

δtBTS

EXPERIMENTAL DEMO

GRID

2 3 4

5 6

Estimated Estimated

BTS3

BTS1BTS2

Mapping

Receiver

Navigating

Receiver

Trajectory Trajectory

Cellular CDMAGPS

Mean Error: 5.51 m Standard Deviation: 4.01 m

ACKNOWLEDGEMENT

This work was supported in part by the Office of Naval Re-

search (ONR) under Grant N00014-16-1-2305, and part by a grant

from the National Center for Sustainable Transportation (NCST),

supported by the U.S. Department of Transportation (USDOT)

through the University Transportation Centers Program.

REFERENCES

[1] J. J. Khalife, K. Shamaei, Z. M. Kassas. “A Software-DefinedReceiver Architecture for Cellular CDMA-Based Naviga-tion,” IEEE/ION PLANS 2016, April 2016.Best student paper award.