IS-GPS-705D (797936), страница 5
Текст из файла (страница 5)
The Q5-code is modulated with a 20-bit Neuman-Hofman code that is also clocked at 1kHz.11IS-GPS-705D24 Sep 20133.3.2.1 Code Structure.The I5i(t) pattern (I5-code) and the Q5i(t) pattern (Q5-code) are both generated by the modulo-2summation of two PRN codes, XA(t) and XBIi(nIi, t) or XBQi(nQi, t), where nIi and nQi areinitial states of XBIi and XBQi for satellite i.
There are over 4000 unique L5 codes generatedusing different initial states of which 128 are currently assigned and identified in Table 3-Ia andTable 3-Ib using the same basic code generator. Section 6.3.4 provides a selected subset ofadditional L5-code sequences with assigned PRN numbers.Exclusive ORReset to all 1s on next clock123456789 10 11 12 13Decode 1111111111101All 1'sXA(t)XA CoderXBI State for SV iCode Clock1 ms EpochXIi(t)XBI(t-niTC)Initial XBI StateResetXQi(t)123456789 10 11 12 13XBQ(t-niTC)Exclusive ORXBI CoderXBQ State for SV iInitial XBQ State123456789 10 11 12 13Exclusive ORXBQ CoderFigure 3-2.
Generation of Codes12IS-GPS-705D24 Sep 2013Figure 3-3. Modulation of Signals3.3.2.2 Code Generation.Each I5i(t) pattern (I5-code) and Q5i(t) pattern (Q5-code) are the modulo-2 sum of two extendedpatterns clocked at 10.23 Mbps (XA and XBIi or XBQi). XA is an 8190 length code, with aninitial condition of all 1s, that is short cycled 1-chip before its natural conclusion and restarted torun over a period of 1 millisecond (synchronized with the L1 frequency C/A-code) for a total of10,230 chips. The XBIi and XBQi, with initial conditions indicated in Table 3-I, are 8191 lengthcodes that are not short cycled. They are restarted at their natural completion and run over aperiod of 1 millisecond (synchronized with the XA code) for a total of 10,230 chips.
Thepolynomials for XA and XBIi or XBQi codes, as referenced to the shift register input, are:XA: 1+ x9 + x10 + x12 + x13, andXBIi or XBQi: 1 + x + x3 + x4 + x6 + x7 + x8 + x12 + x13.Samples of the relationship between shift register taps and the exponents of the correspondingpolynomial, referenced to the shift register input, are as shown in Figures 3-4 (XA code) and 3-5(XB code). In the case of the XB codes, the shift register can either be initialized with all 1s andadvanced ni states as specified in Table 3-I, or initialized with the state indicated in Table 3-I.The state of each generator can be expressed as a code vector word which specifies the binarysequence constant of each register as follows:(a) the vector consists of the binary state of each stage of the register,13IS-GPS-705D24 Sep 2013(b) the stage 13 value appears at the right followed by the values of the remaining states in orderof descending stage numbers, and(c) the shift direction is from lower to higher stage number with stage 13 providing the currentoutput.
This code vector convention represents the present output and 12 future outputs insequence. Using this convention, at each XA epoch (state 8190), the XA shift register isinitialized to the code vector 1111111111111, while at each XB epoch (state 8191), the XB shiftregister is initialized to a code vector peculiar to the PRN number and phase. The XB codevectors are as indicated in Table 3-I.
Alternatively, the XB shift register is initialized to the codevector 1111111111111 and advanced ni states as indicated in Table 3-I.The natural 8191 chips of the XA sequence is shortened to 8190 chips to cause precession of thesecond XA sequence with respect to the natural 8191 chip XB sequence, as shown in Figure 3-6.Re-initialization of the XA shift register produces a 10230-chip sequence by omitting the last6151 chips of the second natural XA sequence, or reinitializing to all 1s at the 1 ms epoch.
TheXB shift register is simply allowed to run its natural course until the next 1 ms epoch when it isreinitialized at its initial state, B0, based upon PRN number and phase. This results in the phaseof the XB sequence leading by one chip during the second XA sequence in the 1-millisecondperiod. Depending upon the initial state of the XB sequence, a third 8191-chip sequence may bestarted before the 10230-chip sequence is completed. Two different scenarios that may result areshown in Figure 3-6.In scenario a, the initial state of the XB sequence, B0, is less than State 6152. Thus, the secondnatural XB sequence does not run to completion prior to the next 1 ms epoch.
In scenario b, theinitial state of the XB sequence, B0, is greater than State 6151. Thus, the second natural XBsequence runs to completion and a third natural sequence starts (except when B0 is State 6152)prior to the next 1 ms epoch.14IS-GPS-705D24 Sep 2013POLYNOMIAL XA:1+x9+x10+x12+x13STAGENUMBERS110211312413514INITIALCONDITIONS6157181679181011111091211113112OUTPUT13TAPNUMBERSSHIFT DIRECTIONFigure 3-4. XA Shift Register Generator ConfigurationPOLYNOMIAL XB:1+x+x3+x4+x6+x7+x8+x12+x13STAGENUMBERS12435679810111213OUTPUT01234567INITIAL CONDITIONS ARE A FUNCTIONOF PRN AND PHASESHIFT DIRECTION8910111213TAPNUMBERSFigure 3-5. XB Shift Register Generator Configuration15IS-GPS-705D24 Sep 20131 ms = 102308190181XA CodeB091XB Code1 = 11111111111118 = 1111111111101219 = 11111111111102 = State 2040B0a) B0 = Initial State at 1 ms (less than State 6152)1 ms = 102308190181XA Code218191B091XB Code1 = 11111111111118 = 11111111111019 = 1111111111110912 = State 2040B0b) B0 = Initial State at 1 ms (greater than State 6151)Figure 3-6.
Relative Phases between the XA and XB Sequences3.3.2.3 Q5 Synchronization Sequence.Each of the 1 ms Q5-code blocks is further encoded with a 20-bit Neuman-Hofman code. The 20bits are modulo-2 added to the Q5 code chips at the PRN code epoch rate of 1 kHz. The code,nh20(t), starting coincident with the 20 ms data epoch on the I5 channel, is as follows:1stLastnh20(t) = 0 0 0 0 0 1 0 0 1 1 0 1 0 1 0 0 1 1 1 03.3.3Navigation Data.3.3.3.1 Navigation Data Modulation.The L5 CNAV bit train, D5(t), is rate 1/2 convolution encoded and, thus, clocked at 100 symbolsper second (sps).
In addition, the 100 sps symbols are modulated with a 10-bit Neuman-Hofmancode that is clocked at 1 kHz (reference paragraph 3.3.3.1.2). The resultant symbol sequence isthen modulo-2 added with I5 PRN code and used to modulate the L5 in-phase carrier.3.3.3.1.1 Forward Error Correction.The L5 CNAV bit train, D5(t), will always be rate 1/2 convolution encoded with a Forward ErrorCorrection (FEC) code. Therefore, the symbol rate is 100 sps. The convolution coding will be16IS-GPS-705D24 Sep 2013constraint length 7, with a convolution encoder logic arrangement as illustrated in Figure 3-7.The G1 symbol is selected on the output as the first half of a 20-millisecond data bit periodcoincident with the first bit of the 20-bit Q5 Neuman-Hofman code.Six-second navigation messages broadcast by the SV are synchronized with every fourth of theSV’s P(Y)-code X1 epochs. Although these epochs are not necessarily accessible to the L5 user,they are used within the SV to define GPS time.
However, message synchronization doesprovide the L5 user an access to the time of every 4th P(Y)-code X1 epoch. The navigationmessage is FEC encoded in a continuous process independent of message boundaries (i.e. at thebeginning of each new message, the encoder registers illustrated in Figure 3-7 contain the last sixbits of the previous message). Thus, herein, reference will continue to be made to these X1epochs.
See IS-GPS-200 for details.The FEC encoding convolves successive messages. It is necessary to define which transmittedsymbol is synchronized to SV time as follows. The beginning of the first symbol that containsany information about the first bit of a message will be synchronized to every fourth X1 epoch(referenced to end/start of week).
The users’ convolution decoders will introduce a fixed delaythat depends on their respective algorithms (usually 5 constraint lengths, or 35 bits), for whichthey must compensate to determine system time from the received signal. This convolutiondecoding delay and the various relationships with the start of the data block transmission and SVtiming are illustrated in Figure 3-8 for the L5 signal.G2 (133 OCTAL)OUTPUT SYMBOLS(100 SPS)DATA INPUT(50BPS)(ALTERNATING G1/G2)G1 (171 OCTAL)SYMBOLCLOCKFigure 3-7.
Convolution Encoder17IS-GPS-705D24 Sep 2013ENCODED DATA BLOCKTRANSMITTED ON L5DATA BLOCKDECODED BYUSERENCODEDDATA BLOCKRECEIVED BYUSERUSER’S DECODING DELAYDOWNLINK DELAYLATEREARLYSV 6 SECOND EPOCHSFigure 3-8. Convolution transmit/Decoding Timing Relationships3.3.3.1.2 Neuman-Hofman Code.Each of the 100 sps symbols are further encoded with a 10-bit Neuman-Hofman code. The 10-bitNeuman-Hofman code is defined to be 0000110101. The 10 bits are modulo-2 added to thesymbols at the PRN code epoch rate of 1 kHz starting at the 100 sps symbol transitions.
Theresult is that a "1" data symbol is replaced by 1111001010, and a "0" data symbol is replaced by0000110101.3.3.4 GPS Time and SV Z-Count.GPS time is established by the Operational Control System (OCS) and is referenced toCoordinated Universal Time (UTC) as maintained by the U.S. Naval Observatory (UTC(USNO)) zero time-point defined as midnight on the night of January 5, 1980/morning ofJanuary 6, 1980. GPS time is the ensemble of corrected composite L1/L2 P(Y) SV times,corrected via the clock corrections in the L1 and L2 NAV data and the relativity correction. Thelargest unit used in stating GPS time is one week defined as 604,800 seconds, concatenated withthe GPS week number.
GPS time may differ from UTC because GPS time is a continuous timescale, while UTC is corrected periodically with an integer number of leap seconds. There also isan inherent but bounded drift rate between the UTC and GPS time scales. The OCS controls theGPS time scale to be within one microsecond of UTC (modulo one second).The L5 CNAV data contains the requisite data for relating GPS time to UTC. The accuracy ofthis data during the transmission interval will be such that it relates GPS time to UTC (USNO) towithin 20.0 nanoseconds (one sigma).
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
