Appendix J
Selective Denial of Civilian GPS Signals by the Military

The recommended policy on GPS signal denial, in lieu of the use of SA, will force our military to take a variety of steps to deny local GPS access. Therefore, in considering signal structure enhancements of C/A-code emissions on or near L2 or transmissions at a new frequency in the L-band, the NRC committee was mindful of the need to maintain flexible selective denial options. In all cases, the issues of whether a C/A-code signal on L1 or L2 could be selectively denied without severely impacting the existing military receiver inventory, or whether a modified or enhanced military receiver would be needed were addressed. The preliminary assessment of these issues involved discussions with appropriate experts in the military GPS community (Captain Jay Purvis, National Air Intelligence Center; Mr. John Clark, the Aerospace Corporation; and Mr. William Delaney, MIT Lincoln Laboratory) as well as computer modeling of the selective denial jammer problem.

The following three questions were posed to the cognizant GPS experts:

(1)  

 Is it feasible to employ a noise jammer that covers a 2-Mhz C/A-code bandwidth,

  • -  

    without unduly impacting local friendly Y-code users;

  • -  

    without modification to existing military receivers;

  • -  

    with modification to existing military receivers?

  • (2)  

     Are there more sophisticated jamming signals that could render C/A-code receivers ineffective without unduly impacting friendly Y-code (and C/A-code) reception. For example, encrypted pseudo-noise jamming, which could be removed by friendly receivers?

    (3)  

    Are there high-confidence deceptive spoofing techniques which, over operationally useful areas, could render a sophisticated aided C/A-code receiver inoperative?

    It was clear from discussions on these and related questions that the defense community is just now embarking on operationally oriented studies and activities addressing the efficacy of various denial techniques. Not surprisingly, those closer to the operational side are more doubtful than those in the development community that highly surgical



    The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
    Copyright © National Academy of Sciences. All rights reserved.
    Terms of Use and Privacy Statement



    Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
    Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

    Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

    OCR for page 249
    --> Appendix J Selective Denial of Civilian GPS Signals by the Military The recommended policy on GPS signal denial, in lieu of the use of SA, will force our military to take a variety of steps to deny local GPS access. Therefore, in considering signal structure enhancements of C/A-code emissions on or near L2 or transmissions at a new frequency in the L-band, the NRC committee was mindful of the need to maintain flexible selective denial options. In all cases, the issues of whether a C/A-code signal on L1 or L2 could be selectively denied without severely impacting the existing military receiver inventory, or whether a modified or enhanced military receiver would be needed were addressed. The preliminary assessment of these issues involved discussions with appropriate experts in the military GPS community (Captain Jay Purvis, National Air Intelligence Center; Mr. John Clark, the Aerospace Corporation; and Mr. William Delaney, MIT Lincoln Laboratory) as well as computer modeling of the selective denial jammer problem. The following three questions were posed to the cognizant GPS experts: (1)    Is it feasible to employ a noise jammer that covers a 2-Mhz C/A-code bandwidth, -   without unduly impacting local friendly Y-code users; -   without modification to existing military receivers; -   with modification to existing military receivers? (2)    Are there more sophisticated jamming signals that could render C/A-code receivers ineffective without unduly impacting friendly Y-code (and C/A-code) reception. For example, encrypted pseudo-noise jamming, which could be removed by friendly receivers? (3)   Are there high-confidence deceptive spoofing techniques which, over operationally useful areas, could render a sophisticated aided C/A-code receiver inoperative? It was clear from discussions on these and related questions that the defense community is just now embarking on operationally oriented studies and activities addressing the efficacy of various denial techniques. Not surprisingly, those closer to the operational side are more doubtful than those in the development community that highly surgical

    OCR for page 249
    --> jamming techniques for denial or spoofing of C/A-code reception on either the L1 or L2 frequency can be deployed. One of the major concerns is operational flexibility in the field, with or without modified GPS receivers. To further assess these concerns, a measure of operational flexibility was developed for quantifying the relative impact of denial jamming on friendly and unfriendly forces. One such measure is the post-correlator signal-to-noise ratio (SNR) advantage for friendly forces. A directly related measure is the relative operating distance (ROD) to the denial jammer such that friendly and unfriendly units obtain equal signal-tracking margins: ROD = (enemy distance to denial jammer)/(friendly distance to denial jammer)equal correlator SNR This function was computed assuming ''friendly" Y-code receivers at L2 (unmodified and modified variants) and "enemy" C/A-code receivers at or offset from L2. Using frequency domain convolution techniques and two different postulated denial jammer spectra, the SNR after baseband code wipeoff was computed. For the modified friendly receiver variant, a bandstop filter was incorporated for further suppression of the selective-denial jammer. First, the problem of signal denial precisely at L1 or L2 was considered. As an optimistic bound on possible performance, it was assumed for this case, an ideal high-pass filter cutting off above the first zero crossing of the C/A-code spectrum. This could be implemented by digital filtering of baseband sample prior to code correlation wipeoff. With this cutoff, a 1.5 dB loss in useful Y-code power is incurred. Following code wipeoff, the respective SNRs for the C/A-code and Y-code receivers differ by 31.5 dB. This translates into an ROD distance ratio of 37.5. Operationally, the ROD could be exploited by a commander in several ways. Ideally the denial jammer would be so situated that the near-far or relative-distance geometry would be favorable, with the jammer located on board an aircraft at the battlefield periphery. The limiting case would be a space-born jammer, but this would require a sizable L-band antenna to meet the link budget. A more realistic scenario in a tactical situation would be a denial jammer close at hand. Suppose that jammer power has been set by a field commander to deny those C/A-code users within a radius of 10 kilometers of the jammer site. Friendly receivers equipped with comparable aiding and antenna augmentations would fail to operate within a radius of about 266 meters. This is the most optimistic scenario and requires substantial modification to Y-code receivers. Any ROD advantage would be eroded if inertial aiding and/or nulling antennas were employed by hostile units. Therefore, the NRC committee concluded that surgical jamming of the C/A-code centered at L1 or L2 would cause operationally unacceptable consequences for Y-code users. Assuming heavy jamming of the existing C/A-code at L1, with unacceptable impact on Y-code at that frequency, attention focused on a new civilian frequency, L4. The goal was to provide a sufficiently large separation from L1 for civilian ionospheric correction and adequate separation from L2 to permit effective selective-denial jamming. In support of this new transmission frequency, a selective-denial jamming analysis was carried out for narrow-

    OCR for page 249
    --> band C/A-like code transmissions offset from L2, as well as a P-code like wider civilian signal offset from L2. The first and third nulls of the L1 Y-code, at 1237.6 MHz and 1257.6 MHz respectively, were examined. Table J-1 summarizes the ROD ratio and SNR advantage under the noted assumptions. The first two options were covered above. Options 3 through 5 are with the new narrow-band civilian signal located at the first null. The most interesting result is for the shape II jammer spectrum. Without receiver modification the ROD distance ratio for this type of spectrum is 91.2 (39.2 dB SNR advantage). A modified receiver operating with the same jammer gives a ROD of 167. In Option 6 the narrow-band civilian signal is placed at the third zero crossing of the L2 Y-code, and is denied with shaped noise jamming. Without receiver modification, the ROD is extremely large, and there is no difficulty in isolating the narrow-band jamming signal from L2 Y-code. Table J-1 Relative Operating Distances and Signal-to-Noise Advantage for Selective Denial Jamming Alternatives Selective Denial Jamming Option ROD (relative operating distance ratio) dB Post-Correlator Advantage Option 1 Narrow-band code on L2; shape II jammer spectrum; no receiver modification 3.2 10 Option 2 Narrow-band code on L2; shape II jammer spectrum; ideal high-pass filter in receiver 37.5 2 Option 3 Narrow-band code at first null of L2; shape I jammer spectrum; no receiver modification 31.6 30 Option 4 Narrow-band code at first null of L2; shape II jammer spectrum; no receiver modification 91.2 39 Option 5 Narrow-band code at first null of L2; shape II jammer spectrum; fourth order band-stop filter 167 45 Option 6 Narrow-band code at third null of L2; shape II jammer spectrum; no receiver modification 7,080 77 Option 7 Wide-band code at third null of L2; shape II jammer spectrum; no receiver modification 63 36 Option 8 Wide-band code at third null of L2; shape II jammer spectrum; receiver low pass modification 630 56

    OCR for page 249
    --> a. Both narrow- and wide-band codes have sinc2 spectrum and 1 MHz and 10 MHz chipping rates, respectively. b. Shape I jammer follows sinc2 spectrum. c. Shape II jammer follows MSK (minimum shift key) spectrum. Wide-band civilian transmissions at the third null were examined in Options 7 and 8. Neglecting receiver radio frequency/intermediate frequency selectivity, the isolation advantage from the code wipeoff process alone is 36 dB, corresponding to an ROD of 63. Typical Y-code receivers have substantial intermediate frequency attenuation 20 MHz to 30 MHz above L2. Modeling this as an ideal low-pass starting at the 20 MHz point gives a 56 dB advantage and a corresponding ROD 630, which are adequate for any operational scenario. The other alternatives discussed with GPS experts, but not analyzed due to time constraints, were pseudonoise jamming and spoofing. These techniques could obviously be applied in conjunction with the jamming techniques above. Pseudonoise jamming requires a modified receiver that coherently estimates and subtracts denial jamming prior to the code-correlation process. Note that this technique might offer the distinct advantage of C/A-code operation for friendly forces under certain circumstances, while denying C/A-code to the adversary. Such a technique fits well with the digital band-stop filtering incorporated in the above analysis or could be introduced at an existing receiver's radio frequency input. While it may be possible to subtract much of the pseudonoise from friendly receivers, perhaps assisted by known selective-jammer location and known user motion, experts expressed concern with null depths and the ability to rapidly adapt to multiple jammers. Once again, the sophisticated user could employ antenna nulling and receiver aiding techniques to greatly diminish the effectiveness of this kind of selective denial. GPS signal spoofing of the so-called "denial" type, in which individual tracking loops are forced back into reacquisition mode, also was a technique discussed with the GPS experts. It was possible to postulate a number of techniques that would reduce its effectiveness; therefore, this technique, taken by itself, was not considered as adequate for selective denial. The above techniques are illustrative of the potential denial techniques that could be applied operationally. Denial jamming of an offset L2 frequency offers clear advantages over the other techniques. However, further in-depth study may suggest ways to combine these techniques for greater operational effectiveness and flexibility.