Tampere University Wireless Communications and RF Systems Master's Thesis Guide: 30 ECTS, RF Measurement and Trepo

Quick answer: what governs the WCRF thesis?

Tampere University’s Wireless Communications and RF Systems (WCRF) study option is a 120 ECTS, two-year Master of Science (Technology) option in Computing Sciences and Electrical Engineering. The current specialization object is STEM-WCRFY Wireless Communications and RF Systems, labelled at least 120 credits for 2026–2027. That label must not be read as 120 ECTS of coursework plus another thesis: the degree itself is 120 ECTS, while the Technology master’s thesis itself is 30 ECTS and uses the 0–5 grading scale.

The current applicant-facing programme spans four overlapping technical domains: wireless communication systems including 5G/6G, radio-based positioning and sensing/radar, radio-frequency integrated-circuit design, and RF/antenna engineering. That breadth makes one programme-level guide defensible, but it also means thesis methods vary substantially. A radio-network simulation, antenna prototype, RF measurement campaign, positioning algorithm, radar experiment and RFIC study cannot be evaluated with the same evidence checklist.

Current course pages such as COMM.SYS.300, COMM.SYS.750, COMM.SYS.760, COMM.RF.401, COMM.RF.410 and COMM.RF.710 show the present technical environment. This guide uses them as current methodological evidence, not as a claim that every WCRF student must take all of them. The public STEM-WCRFY page does not expose a compulsory course list.

1. Treat STEM-WCRFY as the current programme object, not a credit equation

STEM-WCRFY is current through 2026–2027 and is shown as an advanced-studies module with an extent of at least 120 credits. Because the applicant-facing WCRF degree is itself 120 ECTS, do not add the module label to the 30 ECTS thesis and conclude that the programme contains 150 credits. Use Sisu to see how the degree structure, recognition of prior studies, joint studies, specialization content and thesis are represented in your own plan.

For thesis planning, the important institutional rule is simpler: WCRF leads to Master of Science (Technology), so Tampere’s Technology thesis process applies. The programme module defines the academic domain; the Technology thesis page defines supervision, research-plan, originality, submission and assessment procedures.

2. What the 30 ECTS Technology thesis means

Thirty ECTS is the thesis requirement, not a mandatory hardware deliverable. Tampere allows empirical research, planning and implementation, and literature review within the Technology framework. In WCRF, that can become link-level or system-level simulation, algorithm development, radio/RF measurement, antenna design, circuit analysis, signal-processing implementation, positioning experiments, radar sensing, prototype work or a rigorous literature-based study.

A technically impressive artefact is not automatically an academic thesis. A working antenna, RF board, simulator, positioning pipeline or radar demonstrator still needs a bounded question, justified method, evidence, analysis, limitations and conclusions. The contribution should be stated at the level actually demonstrated.

3. Choose the technical centre of gravity first

The current programme page explicitly names wireless systems, positioning/sensing/radar, RFIC design and RF/antenna engineering. Decide which of these carries the primary research question. Cross-layer theses are possible, but one layer should remain the centre of evaluation.

“6G”, “RFIC”, “antenna”, “radar” or “positioning” is a field label, not a research question. Define the system boundary, operating conditions, comparison, metric and evidence. If the thesis compares two beamforming methods, specify the array, channel or measurement setting and performance objective. If it studies an RF circuit, define frequency, gain/noise/linearity/power targets and the evidence level used.

4. Wireless-system simulation: make assumptions visible

Current COMM.SYS.300 and COMM.SYS.750 show that communication theory and advanced wireless-system analysis remain part of Tampere’s 2026–2027 environment. Simulation-heavy work should therefore be methodologically explicit. Record the channel or propagation model, bandwidth, waveform/numerology or protocol assumptions, topology, user distribution, traffic, mobility, scheduler/resource-allocation settings, random seeds and repetition strategy when they affect the result.

Do not report only a best run. If randomness matters, show variability or confidence information. Keep comparisons fair by using equivalent workloads and constraints. Throughput, latency, outage probability, spectral efficiency, energy use, fairness and control overhead answer different questions; select metrics because they test the claim.

If the conclusion changes strongly with user count, SNR, mobility or channel model, that sensitivity is part of the result. A narrow conclusion that states its operating range is stronger than a universal claim unsupported by the model.

5. Standards and 5G/6G claims need version control

Modern wireless research often depends on standards, simulator implementations and rapidly changing releases. Separate a formal standard requirement from a research proposal, vendor behaviour or your own configuration. State the 3GPP release, IEEE version, open-source stack, simulator version or dataset version when it materially affects interpretation.

A “6G” label does not establish novelty. If the work evaluates an emerging waveform, sensing concept, positioning method or resource-allocation scheme, define the baseline and why the comparison is meaningful. Avoid describing a research prototype as an adopted standard feature unless the source actually supports that claim.

6. RF and antenna measurements: calibration is part of the method

Current COMM.RF.410 Basic RF Measurements covers spectrum analyzers, signal generators, network analyzers, calibration, impedance matching and antenna parameters. That makes calibration and reference-plane control directly relevant methodological concerns for WCRF measurement theses.

Record instrument model/settings when material, cable/fixture configuration, calibration method, frequency span, resolution bandwidth, averaging, power levels and reference plane. For S-parameter work, explain what has been de-embedded or included. For over-the-air measurements, document geometry, antenna orientation, polarization, distance, environment and uncertainty sources.

A clean graph is not enough if the measurement chain is ambiguous. Unexpected ripple, cable loss, connector repeatability, mismatch or chamber/room reflections can dominate a small claimed improvement. Include uncertainty or repeatability evidence when the conclusion depends on small differences.

7. Antenna-design theses: separate simulation, construction and measurement

Current COMM.RF.710 Antenna Design explicitly combines numerical simulation with construction, testing and interpretation of measured results. A thesis should preserve those evidence levels. A simulated resonance, a fabricated antenna and a measured radiation pattern answer different questions.

Document solver/tool version, mesh or convergence choices, boundary conditions, substrate/material parameters, feed model and nearby structures when they matter. For fabricated prototypes, record dimensions, material tolerances and connector/feed implementation. For measurements, explain calibration, fixture/chamber arrangement and whether the measured environment matches the simulation assumptions.

If simulation and measurement disagree, do not hide the discrepancy. Investigate fabrication tolerance, dielectric uncertainty, connector/parasitic effects, cable routing, nearby objects and measurement setup. Explaining the disagreement can be a stronger research contribution than forcing visual agreement.

8. RFIC-oriented theses: state the evidence level explicitly

The current programme page includes radio-frequency integrated-circuit design. RFIC research can involve architecture, transistor-level design, matching networks, noise, linearity, stability, power consumption, layout/parasitics, process-voltage-temperature variation and measurement. Not every thesis reaches manufactured silicon.

If the work uses schematic/pre-layout simulation, say so. If it includes extracted or post-layout simulation, identify that level separately. If measured silicon is available, distinguish measured results from simulated prediction. Do not describe post-layout simulation as silicon validation.

For comparisons, use consistent process assumptions, supply voltage, frequency, load, bias and performance definitions. PDK, model and foundry information can be confidential; describe enough method for academic assessment without publishing restricted process files or proprietary design-kit content.

9. Radar, sensing and joint communications-sensing studies

Current COMM.SYS.760 covers radar signal processing, detection/estimation, tracking, radar imaging, array processing, MIMO/phased arrays, beamforming and joint communications-radar concepts. A sensing thesis should define the scenario as carefully as a communications thesis defines its channel.

Specify waveform, carrier/bandwidth, antenna or array configuration, geometry, target properties, motion, clutter/interference assumptions, sampling and processing chain. Define ground truth and the evaluation metric: detection probability, false-alarm rate, range/velocity/angle error, tracking error or classification performance, for example.

If machine learning is used for radar or sensing, separate data leakage control, train/validation/test split and scenario generalization from the signal-processing pipeline. A model that recognizes the measurement setup rather than the target does not support the intended claim.

10. Positioning and localization theses

The programme explicitly includes radio-based positioning. Positioning results are sensitive to geometry, synchronization, multipath, obstruction, device clock behaviour, antenna characteristics and ground-truth quality. State which positioning regime is studied: GNSS, cellular, Wi-Fi, ranging, angle-based, fingerprinting, sensor fusion or another method.

Do not rely only on mean error. Median, percentile, availability, outage and spatial distribution can reveal failure modes hidden by an average. Define whether the evaluation is 2D/3D, static/mobile, line-of-sight/non-line-of-sight, indoor/outdoor and device-specific or general.

If ground truth comes from another positioning system, discuss its own uncertainty. The reference system is not automatically exact.

11. Cross-layer WCRF projects need one primary claim

A thesis may connect an RF front end to a communication algorithm, an antenna array to beamforming, or radar sensing to communications resource management. Cross-layer work is valuable when the interaction is the research problem. It becomes weak when it accumulates several unrelated engineering tasks.

Write down which variable crosses the layers and what outcome is evaluated. For example, if antenna-pattern constraints affect link performance, define the mapping from measured/simulated antenna behaviour to system-level metrics. If hardware nonlinearity affects waveform performance, connect circuit behaviour to communication-level evidence rather than presenting two separate mini-projects.

12. ITC.CEE.800 and the current seminar boundary

Current ITC.CEE.800 Tools for theses is 5 ECTS, pass/fail, with research methods, research ethics, information searching skills and career planning. Tampere recommends completing it before the thesis.

However, the current public STEM-WCRFY page does not establish ITC.CEE.800 as compulsory in every WCRF study plan. Verify its placement in your own Sisu plan. Likewise, Tampere’s university-wide Technology thesis framework requires seminar/presentation activity, but the current public WCRF module does not expose a separate current WCRF-specific thesis-seminar course code. Follow current Sisu/Moodle instructions rather than reviving an old seminar code.

13. Topic, supervisor and research environment

Begin topic searching early enough to match the supervisor to the technical centre of gravity. RFIC, antennas, radio systems, positioning and radar can require very different expertise, tools and laboratory access. A company supervisor can provide engineering context, but Tampere retains the academic supervision and examination responsibilities.

Before committing to an industry topic, confirm what can be published, what equipment/data can be accessed and whether the proposed evidence can survive confidentiality constraints. A thesis cannot depend on results that examiners are not allowed to see.

14. Use the Thesis Supervision Plan as a project-control document

The Technology thesis process requires a Thesis Supervision Plan. Agree on question, scope, method, schedule, meetings, feedback, responsibilities and examiner arrangements. For WCRF work, add dependencies such as anechoic or RF laboratory access, VNAs/spectrum analyzers, antenna fixtures, SDRs, private 5G systems, licensed EDA tools, compute resources, PDK access, fabricated hardware and datasets.

Identify fallback paths. If a chip, PCB, antenna prototype or measurement campaign is delayed, decide whether validated simulation, an existing platform or a reduced scope could still answer the question. A fallback should preserve the research question rather than silently replacing it.

15. Write the research plan before the setup becomes expensive

Define the research question, literature basis, system boundary, variables, baseline, metrics, data/experiment plan, analysis strategy, ethics/data issues and limitations before committing to a long fabrication or measurement cycle. Map each major claim to the evidence needed to support it.

For experimental RF work, create a test matrix before final hardware arrives. For simulation, freeze a parameter table and reproducible configuration. For positioning/radar data collection, define ground truth and failure criteria before collecting the main dataset.

16. Reproducibility and configuration control

Wireless/RF results can change with small configuration differences. Preserve software/firmware commits, HDL or DSP versions, simulation configuration, instrument scripts, calibration files, board/antenna revision, dataset version and analysis code when relevant.

If results are manually copied between tools, create a traceable naming convention. A figure should be traceable back to the exact configuration and raw data that produced it. For confidential projects, the repository may remain private, but internal reproducibility still matters.

17. Data, privacy, spectrum and confidential systems

Radio traces, network logs, device identifiers, location data and sensing datasets may contain personal or security-sensitive information. Collect only what the research requires and follow Tampere data-protection guidance. If people are identifiable or personal data are processed, agree on the lawful and ethical handling with the supervisor before collection.

Company projects may involve proprietary RF schematics, PDKs, source code, spectrum plans, customer data or security details. Tampere theses are public documents. Keep confidential material outside the evaluated public thesis while retaining enough method and evidence for academic scrutiny.

18. AI use in WCRF research

Tampere permits AI use under its current study guidance, but the student remains responsible for the work. AI can assist with coding, text improvement or preliminary exploration, but generated equations, RF assumptions, standards claims, register settings, citations or simulation code must be verified.

Do not upload confidential PDK material, company source code, credentials, unpublished measurement data or personal data to an external AI service unless the applicable rules allow it. Acknowledgement requirements apply when AI materially contributes.

19. Write while simulation, fabrication or measurement continues

Start the manuscript before the final experiment. Keep architecture diagrams, parameter tables, calibration records, experiment logs and result indexes current. Separate system description, method, verification, results and discussion so readers can see what was assumed, what was built or configured, what was measured and what the evidence supports.

Do not delete failed results simply because they are inconvenient. Failed calibration, unstable circuit behaviour, unexpected propagation, convergence problems or measurement disagreement can reveal limitations that belong in the thesis discussion.

20. Maturity test and language route

The master’s degree includes the maturity requirement. For international Technology master’s students, the current thesis guidance commonly uses the thesis abstract as the maturity test, while individual language-history situations can require a different route. Confirm your own case rather than copying another student’s process.

The maturity route can affect final processing, so resolve it before the submission deadline.

21. Turnitin and originality

After supervisor permission, the final thesis goes through Tampere’s Turnitin originality process. Similarity percentage is not an automatic plagiarism verdict. Standard terminology, quoted material, references and earlier drafts can affect the report, and the supervisor interprets the result academically.

Resolve unattributed reuse, copied method descriptions and source attribution before final submission. AI-assisted text does not remove the student’s responsibility for originality and factual accuracy.

22. Trepo, PDF/A and publicity

The final thesis is deposited through Tampere’s process in Trepo and permanently archived electronically. The archival file must be valid PDF/A. Test conversion before the deadline because equations, fonts, figures and embedded objects can cause errors.

The thesis is a public document. Commercial sensitivity does not make the evaluated thesis confidential. Structure industry work so confidential appendices, source code, process information or customer details stay outside the public manuscript.

23. Assessment, examiner timing and finality

WCRF uses the Technology thesis 0–5 scale. The normal examiner assessment window is 21 days after the relevant Trepo/submission process; the current guidance extends it to 28 days where the applicable maturity-test route also has to be completed. These are examiner windows, not guaranteed total graduation-processing times.

The examiners’ proposed assessment is sent through the formal process. Students can respond under Tampere’s rules, and once the thesis is approved it is final rather than freely resubmittable. Leave time for examination and Dean/faculty processing when planning graduation.

24. Final WCRF checklist

Before starting, verify the current STEM-WCRFY placement in Sisu, the 30 ECTS Technology thesis, supervisor, topic, ITC.CEE.800 placement and current seminar/presentation implementation. Freeze the primary technical layer, research question, evidence level, baseline, metrics and reproducibility plan.

Before submission, verify that simulation and measurement claims are not mixed, document calibration and configuration where relevant, discuss uncertainty and limitations, remove confidential material, complete the maturity requirement, follow Turnitin, create valid PDF/A, submit through Trepo and complete the separate graduation application.

Before freezing the final manuscript, perform one last traceability pass: every headline result should point to a known configuration, every comparison should use a defensible baseline, and every limitation that could change the interpretation should be visible to the examiner. For wireless-system work, recheck whether the channel, traffic, mobility and standard-version assumptions are visible beside the results they support. For RF and antenna work, verify that calibration, fixture, reference-plane and hardware-revision information can be reconstructed. For radar and positioning work, confirm that ground truth, geometry and scenario labels follow each dataset through analysis. For RFIC work, make sure every table clearly states whether values are simulated, post-layout or measured, and that process, supply and frequency conditions are comparable. A reader should never have to guess which validation level produced a headline number. This final check is particularly important in wireless and RF research because apparently small differences in calibration, channel assumptions, geometry, process models or software versions can materially change a conclusion.

The safest principle is to keep three layers separate: Tampere’s institutional thesis rules, the current WCRF academic domain, and your specific evidence model. A strong WCRF thesis does not need to cover every wireless/RF topic. It needs a bounded technical question and evidence whose validation level matches the conclusion.