Analysis of the use of CSK for Future GNSS Signals

GNSS signals are designed in order to fulfill the special needs of a GNSS system: to provide the receiver with precise synchronization or pseudo-range measurements and to broadcast little essential information such as the satellites ephemeris, clock error correction, etc. The combination of these two elements allows a GNSS system to provide the user with its PVT (position, velocity, time). The historical design choice for the GNSS system synchronization part consists in implementing spread spectrum direct sequences having a very narrow autocorrelation function. Additionally, the introduction of several almost orthogonal direct sequences, one for each satellite, was used to implement the simultaneous access of the original GPS system constellation satellites; this technique is known as Code Division Multiple Access (CDMA). Therefore, the use of direct sequences in a GNSS system has become a key element and an inherent part of the signal design. The historical design choice of the GNSS signal communication part is the implementation of a BPSK modulation (a BOC modulation is a BPSK modulation from the demodulation point of view). This choice was made in order to allow the easy implementation of the synchronization part: direct sequences. Moreover, the low bandwidth efficiency of a BPSK modulation (number of bits/second/Hz) did not present any limitation to the signal design: the low power of the received signal added to the little required information required imposed a low bit rate. However, nowadays this choice of hybrid signal structure can be questioned due to the introduction of a new dataless (pilot) channel as well as the extension of the GNSS user community with high expectations in terms of new services and positioning capabilities in more challenging environments. On one hand, the pilot channel introduction to a GNSS signal and the possibility for the receiver to generate pseudo-range measurements from this channel implies that the data channel is no longer necessarily restricted by the GNSS system special hybrid characteristics. Therefore, the data channel can be looked at as a more traditional communication channel. One example is the LEX signal of the Japanese QZSS system. Another example could be the GPS L1C signal: the 75% power allocation to the pilot channel could lead to receivers discarding the data channel for synchronization purpose. On the other hand, nowadays new applications and new services such as precise positioning, safety of life, integrity, etc, demand a much higher data rate. Moreover, a higher data rate can improve the signal demodulation performance by, for instance, means of increasing the transmitted information temporal diversity: more repetitions of the ephemeris data allow the receiver to obtain the information more quickly or to accumulate the information for a lower required demodulation C/N0 threshold. The main limitation of using a BPSK data modulation to increase the signal data rate is the signal design choice of employing direct sequences (necessary for Code Division Multiple Access and precise pseudo-range measurements). The PRN code is limited by the data symbol duration which must decrease in order to increase the data rate; consequently, the signal design is directed into two undesired solutions, either increase the chip rate, resulting in a wider spectrum, or decrease the PRN code length, resulting in a loss of PRN code isolation and orthogonality properties. Therefore, the introduction of alternative modulations on the data channel in order to increase the data rate should be analyzed. In this paper, the modulation known as Code Shift Keying (CSK), specially designed to increase the transmission rate of a spread spectrum signal, is inspected. The CSK modulation consists in circularly shifting each transmitted PRN code in order to represent with each PRN code shifted version a different CSK symbol mapping a set of bits. Therefore, if each data channel PRN code period is equal to the data symbol duration, the bit transmission rate is increased proportionally to the number of bits mapped by a CSK symbol. Previous analysis of the CSK technique introduction in a GNSS signal was already done by the authors and can be found in the literature. Besides, the LEX signal of the Japanese QZSS already implements this specific modulation. The aim of this paper is thus to deepen the analysis of the CSK modulation introduction in a GNSS signal and to compare it with the classical BPSK modulation. The study is split in 4 main parts: 1. The demodulation performance analysis of different decoding techniques based mainly on LDPC codes. Indeed, LDPC is proposed for GPS L1C signal and it is recognized as a reference for GNSS data transmission. The investigated LDPC decoding techniques are the typical belief-propagation algorithm, multi-stage decoding and channel code-modulation iterative decoding. Additionally, the implementation of a Reed-Solomon channel code as was done in LEX of QZSS is analyzed. A comparison of each one of these techniques with the traditional BPSK modulation is made and its gains are presented. In particular, it is shown that the iterative decoding and the multistage decoding bring decoding performance gain of 0.6 dB in terms of equivalent C/N0. 2. GNSS signal structure propositions implementing a CSK modulation are made aiming at two different objectives: increasing data rate with respect to the current and future GNSS signals, and maintain the same useful data rate while trying to increase the demodulation performance compared to the reference channel code. 3. The impact of the introduction of a CSK modulated signal on a GNSS receiver is analyzed: degradation of tracking and acquisition performance due to the impossibility of using the data channel is assessed. This analysis compares signals having a pilot and a CSK modulated data channel with signals having a pilot and a BPSK modulated data channel. Different PRN code lengths and different power allocations between pilot and data channels are assessed. The impact of the introduction of a CSK modulation on the receiver's complexity is also inspected. 4. Advantages and drawbacks of a CSK modulated signal with respect to a BPSK modulated signal are presented. Hybrid structures time that multiplex BPSK and CSK modulations are discussed.