Performance Testng of the Rockwell PLGR+ 96 P/Y Code GPS recever By Santago Mancebo and Ken Chamberlan Introducton: The Rockwell PLGR (Precson Lghtweght GPS Recever) + 96 s a Precse Postonng Servce P/Y code GPS recever. Ths recever s avalable to Department of Agrculture and Department of Interor land management agences under a Memorandum of Understandng wth the Department of Defense. These GPS recever performance tests were made at two stes: one under Western Oregon forest canopy at the Clackamas Test Network and another at the open-sky Natonal Geodetc Survey control staton Estacada. All tests were made Usng the Precse Postonng Servce (PPS) P/Y code wth Wde Area GPS Enhancement (WAGE) turned on. The tests were desgned to examne the effect of tree canopy on the followng: The postonal accuracy of varous lengths of data loggng from 1 second to 120 seconds to determne what, f any, ncreases n accuracy occur through the averagng of ncreasng numbers of postons. The effcency of recever performance. That s the number of postons logged durng a specfc perod dvded by the number of possble postons expressed as a percentage. Equpment: Ths test was performed wth the Rockwell PLGR+ 96 usng the manufacturer standard equpment nternal antenna. The manufactures lst prce s $XXXX. Ths unt s manufactured by Rockwell Collns. The PLGR+ 96 SW = 613-9898-013, HW = 1E11 was used for all tests. Test Network and Survey Staton Data: Forested Ste: The Clackamas Test Network s located n Western Oregon on the Mt. Hood Natonal Forest. Clackamas GPS Test Network s a ste for testng P and C/A Code (resource grade) GPS recever performance under moderate to heavy western Oregon tmber canopy. The vegetaton at the ste conssts of Douglas Fr and Western Hemlock over story, these trees are approxmately 24-40" d.b.h., wth a Vne Maple and Red Alder under story. The terran at the ste s nearly flat wth no terran obstructons above 20
degrees. The test network s composed of twelve ponts wth known geographc postons. These twelve ponts were establshed by a conventonal Total Staton closed traverse survey whch was based on two GPS ponts adjacent to the ste whch had reasonable satellte horzon. These two ponts were establshed by statc carrer phase GPS survey connected to the Oregon Hgh Accuracy Reference Network (HARN). Open Ste: The staton Estacada was used as a control ste due to ts clear-sky nature wth no obstructons. Staton Estacada s an Oregon Hgh Accuracy Reference Network (HARN) Order A survey mark establshed by the Natonal Geodetc Survey n 1998, PID-AI1994. The staton s 487 KM, N 47 W of the Clackamas Test Network ste. Feld Data Loggng Procedures: All GPS observatons were made at approxmate antenna heght of 1.5 meters. The recever was pole mounted at the Clackamas ste and trpod mounted at the Estacada ste. All GPS data was logged at 1 poston / 2 seconds. GPS postons ported n Natonal Marne Electroncs Assocaton NMEA 183 format to a notebook personal computer runnng VsualGPS software by Appolocom. All data was logged wth the recever was set wth WAGE (Wde Area GPS Enhancements) on. The recever wth WAGE on processes enhanced clock correcton sgnals from the satelltes. All data was logged wth an elevaton mask of 15 degrees. Clackamas ste: Log approxmately 60 postons at each of the 12 ponts. Data was logged wth the SNR and PDOP set to factory default.. Data was logged wth the ntegral antenna. Estacada control pont: Data logged for 1 hour Data was logged wth the SNR and PDOP set to factory default. Data was logged wth the ntegral antenna. Offce Data Processng Procedures: The data from the NMEA stream was converted from decmal degree format to Degreemnute- second format n Mcrosoft Excel then exported to Geographc Calculator for converson to NAD 83 State Plane Coordnates (SPC), Oregon North Zone. The SPC coordnates were mported nto Mcrosoft Excel. Excel was used to compare the GPS wayponts to the true pont coordnates, calculate statstcs, and compute effcency.
All accuraces report at 2DRMS (see appendx 1 for formulas). The 2DRMS value closely approxmates the 95% error for a poston. In other words, 1 out of 20 measurements wll have errors larger than those publshed n ths report. All data was analyzed n the North Amercan Datum of 1983 (NAD 83). SITE CONDITIONS: These tests were made under dry condtons rangng. No measurements of canopy mosture were made. TEST RESULTS: Test 1. Examne the accuracy of varous lengths of data loggng at forested and open stes Accuracy Relatve to Tme 25 20 Meters 2dRMS 15 10 5 0 1 2 5 10 20 50 100 200 500 1000 Tme n Seconds Place Forest Open Fgure 1: The graph shows the GPS accuracy relatve to length of occupaton at both the open canopy Estacada and forested Clackamas Stes.
Test 1 Dscusson: Ths data reflects a runnng average of all postons at each locaton. Open ste accuracy wth 1 second of data was 13.3 meters 2dRMS. Forested ste (Clackamas) accuracy after 1 second of data was 21.8 meters 2dRMS Forested ste (Clackamas) accuracy after 50 postons of data was 13.4 meters 2dRMS. For open ste (Estacada) there was lttle or no ncrease n accuracy for longer occupaton tmes up to 100 seconds then a slow ncrease n accuracy. For the forested Clackamas Test Network accuracy ncreased when 50 postons were logged.. 802 postons were logged at forested Clackamas Network of these 48 were deleted due to very hgh errors. 1543 postons were logged at the open ste (Estacada), of these none were deleted. Loggng of 50 postons at forested stes appears to provde enough data to detect outler postons. Test 1 Recommendatons: Open stes: One second of data s suffcent; however three wll provde some redundancy. Forested stes: 50 postons of data are recommended.
Test 2: Examne the accuracy dfferences between Forested and Open stes. PLGR Effcency 120% 100% 80% Effcency (%) 60% 40% 20% 0% Forest Place Open Fgure 2: Ths graph presents the effcency of the PLGR at the forested Clackamas Test Network and the open-sky staton Estacada. Test 2 Dscusson: Open ste (Estacada) effcency s 100%. Forested ste (Clackamas) effcency s 60%. Recommendatons for data loggng wth PLGR: Open stes: 1 second of data s suffcent however 3 postons are recommended. Forested stes: 50 postons of data are recommended. Effcency s about 60%. The Natonal Map Accuracy Standard ( NMAS 1947) for pont postons on 1:24,000 Scale Quadrangle maps s 12.2 meters at 90% or to put t n the same confdence nterval as our study 14.8 meters at 95%. Therefore ths recever when used as recommended above meets natonal map accuracy standards when operatng under tree canopy.
Appendx: accuracy computaton formulas Ee Ee = east et = north n T ( 2 2 Ee En ) Eh = + 2 drms = 2 RMS