QuaSim

QuaSim: A Virtual Quantum Cryptography Educator

QuaSim is a virtual learning environment that turns introductory quantum cryptography concepts into interactive photon-manipulation puzzles. The paper describes an Unreal Engine 4 implementation, an architecture for adaptive guidance, and three modules covering polarization, superposition, and measurement. It also introduces the supporting QooSim quantum simulation library, with particular attention to transferring simulated qubits and maintaining entanglement relationships across networked systems. The contribution is a system and educational design illustrated through game screenshots; the paper does not report an evaluation of students learning with QuaSim.

Background and motivation

The authors address an existing educational problem: students can learn isolated mathematical definitions and cryptographic procedures without understanding how quantum components operate within a larger communication network. They argue that lecture-based teaching tends to present concepts sequentially, whereas a project can introduce concepts when students need them to solve a practical problem. Short Capture The Flag competitions offer interactive cybersecurity experience, but the authors consider their typical one-to-three-day format insufficient for sustained theoretical development across a semester. In the paper's 2017 context, expensive quantum key distribution equipment and limited opportunities for field experience further restricted hands-on teaching. QuaSim targets junior and senior undergraduates and graduate students, while its introductory modules assume no previous optics knowledge.

Related work includes puzzle-based cybersecurity education and serious games such as CyberCIEGE, CyberProtect, SimBLEND, and OpenSim-based security exercises. These systems provide precedents for role playing, resource management, feedback, competition, and contextual learning. The authors argue that existing cybersecurity games do not adequately cover quantum cryptography or provide the intended progression toward a complete project involving quantum and classical components. They also review Libquantum, Q++, Quantum::Entanglement, and other quantum simulators, identifying network communication, serialization, distributed entanglement, maintainability, and memory consumption as obstacles to their intended application. These are the authors' assessments of the versions examined at the time, rather than a current comparison of those tools.

Adaptive problem solving and system architecture

QuaSim frames learning as moving from an initial situation to a goal through a sequence of actions. Domain experts specify problems, available resources, security requirements, and constraints, potentially allowing several valid solutions with different costs. The platform description includes roles such as network engineer, project manager, and system administrator, as well as team competition and sharing situation-specific learning experiences. The illustrated modules focus more narrowly on individual photon puzzles, so the screenshots do not establish the operation or educational value of this broader collaborative workflow.

The guidance architecture combines a learning engine, an evaluation and guidance engine, a consistency engine, and a knowledge repository of domain facts and problem-solving tactics. The authors describe encoding facts in first-order logic extended with arithmetic and recursion, attaching expert explanations, and using abductive theorem proving to formulate possible solutions. The learning engine constructs partial or complete action plans and ranks them using time, cost, and other expert constraints. Plans share common structure and associate actions with explanations of rationale, pitfalls, and rollback options. A plan can become a replay capsule that guides solution discovery, while the evaluation and guidance engine assigns interaction scores. The paper presents this architecture at a conceptual level without reporting theorem-prover performance or a quantitative assessment of its guidance.

An Oracle character connects this architecture to the game experience. Figure 1 shows the hub as a first-person three-dimensional scene containing the Oracle and a menu for teleporting to the three modules. Players may follow recommendations based on their role and prior achievements or select another module. Within exercises, the system is described as comparing performance with expected and target success rates for novice, intermediate, and expert users. Difficult categories receive additional practice; continued difficulty prompts hints, instructional videos, or reading material. Successful shots increase score, unsuccessful shots reduce health, and requesting Oracle assistance also carries a health penalty. Watching instruction or completing practice can restore health.

Visual encoding and interaction in the learning modules

The interface embeds abstract quantities in objects that players can aim at and manipulate. A photon gun occupies the foreground, circular targets display polarization orientations, and numerical angle labels support reading the target state. A heads-up display shows health and score, while controls near the gun expose angle values and learning resources. These are screenshots of the game interface, rather than schematic illustrations of a proposed visualization or plots of evaluation results.

The retained excerpt from printed page 604 contains Figures 2 and 3 on polarization and Figure 4 on superposition. Orientation is shown through directional marks inside circular targets and large degree labels; the superposition display adds basis axes and horizontal and vertical component information.

The polarization module introduces three problem sets: Anglematch, Orthogonal, and Quadrant Equivalency. Players first orient and fire photons to match a target angle, learning both the controls and the spatial representation of linear polarization. They then select a polarization orthogonal to the presented angle and learn the equivalence of linear polarization directions separated by $180^\circ$. For example, the paper treats $30^\circ$ and $210^\circ$ as the same linear polarization direction. Figures 2 and 3 show large angle labels, directional marks, a target angle, remaining attempts, and controls for switching problem sets. The task therefore links a visible orientation, a numerical angle, and an action whose success contributes to game feedback.

The superposition module asks students to compose photons using different strengths of two basis components. It starts with a rectilinear basis and later asks students to work relative to another supplied basis. Figure 4 shows a circular basis display with perpendicular axes, orientation labels, component percentages, and an on-screen calculator. This makes the basis and component decomposition explicit within the shooting interface. The text describes monitoring responses to build an error model and adjust practice toward a target success rate, but does not give measured learning outcomes or a detailed algorithm for that model.

The measurement module places three filters at initially random rotations between the photon gun and a final green target. Figures 5 and 6 show the scene and an active puzzle, including the filters' orientations, their degree labels, and the final target above them. The player chooses an initial polarization that maximizes the probability of reaching the target after the intervening measurements. The probability of passage depends on the relative photon and filter angles, and each measurement changes the photon state, requiring reasoning about the state entering and leaving successive filters. Figure 6 also depicts photons traveling between the targets; this animation should be understood as the game's representation of the exercise, rather than a detailed physical optical apparatus. Together, the modules teach prerequisites for the BB84 quantum key distribution protocol, but the paper does not demonstrate a complete BB84 gameplay sequence.

QooSim and distributed simulation

QooSim is a separate C++ library supporting the simulation requirements behind the educational platform. Its initial implementation encapsulated Libquantum functionality, followed by changes to storage and communication. Protocol Buffers serialize data structures, and gRPC provides distributed services. A System class contains the gRPC object and manages quantum registers, supporting protocols whose simulated sender and receiver operate across a TCP/IP network. For entanglement, an EntangledRegister records its operation history, while an Entanglement object coordinates interacting registers and retains their original entangled state, described as an EPR probability matrix. This is a more specific mechanism than simply transferring independent qubit objects between machines.

The paper reports storing two complex floats per qubit and describes these values as probabilities of measuring $0$ and $1$, contrasting this with the much larger register representation used in its Libquantum experiments. It acknowledges a processor-efficiency tradeoff but supplies no memory or runtime benchmark for QooSim. The representation is described too briefly to establish a general complexity result, and the text's use of complex storage values and measurement probabilities leaves details of amplitude and phase handling unclear. Consequently, this account supports a claim about the authors' proposed storage design, not a conclusion that arbitrary entangled quantum states can be simulated faithfully with only two values per qubit.

Contributions, evidence, and remaining work

The main contribution is a game-based introduction to quantum cryptography that connects polarization, superposition, and measurement through related visual tasks and a shared interaction environment. The accompanying architecture explains how expert knowledge, solution planning, learner profiles, and feedback could support adaptive project-based education. QooSim supplies the distributed simulation mechanisms needed for the larger network-oriented ambition. The paper documents these ideas with implementation descriptions and six interface screenshots, rather than a controlled educational experiment.

No participant sample, learning test, usability study, comparative classroom result, or quantitative performance evaluation of QuaSim is reported. Results from earlier game-based learning research motivate the design but do not demonstrate its effectiveness. Similarly, the paper does not benchmark QooSim's memory and processor tradeoffs, establish the full scope of its entanglement representation, or demonstrate that its adaptive guidance improves learning. The authors conclude by restating the platform and its three modules rather than presenting a separate future-work agenda. Evaluating learning, documenting the guidance mechanisms, and validating the simulator's correctness and performance are therefore unresolved questions exposed by the reported evidence, not completed results or explicitly promised extensions.