Title: Direct-to-Cell Satellite Network without Satellite Navigation
Author: Wei Liu (Tsinghua University), Yuanjie Li (Tsinghua University & Zhongguancun Laboratory), Jingyi Lan (Tsinghua University), Hewu Li (Tsinghua University & Zhongguancun Laboratory), Yimei Chen, Lixin Liu, Jiabo Yang, Xi Long, Li Ouyang, Minghao Tang (Tsinghua University), Jianping Wu, Qian Wu, Jun Liu, Zeqi Lai (Tsinghua University & Zhongguancun Laboratory)
Scribe: Gao Han (Xiamen University)
Introduction
Current systems utilize GNSS-derived user location and timing information for wireless access, authentication, and authorization. However, GNSS signals are well-known to be unreliable and prone to mislocalization under interference or malicious manipulation. This cross-technology dependence propagates GNSS’s defects into direct-to-cell satellite networks, leading to intermittent connectivity, over/under-billing, unauthorized services, and service denials, even when satellites are accessible. Therefore, it is both important and timely to explore whether direct-to-cell satellite networks can function without external navigation. This paper examines the feasibility of enabling direct-to-cell satellite networks without external satellite navigation.
Key Idea and Contribution
The paper proposes SN2 (Self-Navigating direct-to-cell Satellite Network), a new paradigm that follows the “fate-sharing” principle to reuse direct-to-cell satellites themselves for navigation. Instead of striving for high-precision positioning, SN2 embraces a network-oriented navigation philosophy: accurate navigation is not always necessary, as long as “good enough” geolocation and timing information is available to guarantee functionally correct services.
SN2 leverages two key insights :
-
Direct-to-cell satellites as navigators: Similar to GNSS, satellites’ broadcast signals and predictable ephemeris can be reused for positioning and timing. LEO mega-constellations, with stronger signals and larger scale, are a reliable complement to GNSS.
-
Network-oriented navigation: Network functions such as radio access, authentication, and authorization tolerate a certain degree of inaccuracy. SN2 revisits these requirements and safely relaxes them, enabling correct services even when precise GNSS is unavailable.
The design reuses existing cellular signals and control messages, making it incrementally deployable in commodity phones/IoTs, satellites, and 3GPP NTN architecture.
Evaluation
Compared to baseline 3GPP NTN and operational systems like Iridium and Globalstar, SN2 improves network availability by 4.4–23.5× and reduces access latency by 1.9–12.3×. Even when GNSS is unavailable or under attack, SN2 guarantees that services remain available, while existing solutions often fail.
Question
Q1： Would your solution introduce too much overhead—for example, in terms of communication or computation?
A1: Our paper actually includes a detailed analysis of both communication and computation overhead. For communication, we don’t introduce new signaling—we just reuse existing messages for localization. Some extra signaling may happen when GNSS is unavailable, but this is common to all solutions, and we mitigate it by using more satellites for faster access. As for computation cost, the required operations are already supported in the baseband of current networks. We only reuse them for localization, so the additional overhead is negligible.
Q2: From a security perspective, you mentioned that the system is protected against spoofing. Does this include all kinds of spoofing, for example, GNSS spoofing or even direct fake cell signals?
A2: Our solution is robust as long as there is at least one legitimate signal, since we can fall back to single-satellite localization. If all authentic signals are blocked and only spoofed ones remain, then no solution can work. But in practice, full spoofing is very hard.
Personal Thoughts
This paper presents SN2, a network-oriented self-navigation scheme for highly available, reliable, and secure direct-to-cell satellite network services. SN2 shows how to strengthen network resilience without extra hardware or major protocol changes. However, the evaluation focuses mainly on small-scale experiments. Its performance in complex scenarios involving large-scale constellations, multi-user concurrency, and cross-border roaming strategies warrants further exploration. Additionally, SN2 may still require higher positioning accuracy for emergency services (e.g., SOS). For such services, the trade-off between accuracy and availability deserves further investigation.