Tampere University Communication Systems and Networks Master's Thesis Guide: 30 ECTS, COMM-S01 and Trepo

Quick answer: what exactly governs this thesis?

Communication Systems and Networks (CSN) is a 120 ECTS, two-year Master of Science (Technology) study option at Tampere University within Computing Sciences and Electrical Engineering. For a student planning the thesis in the current 2026-2027 curriculum, the most important distinction is simple: the thesis follows Tampere University’s Technology/Architecture master’s-thesis framework, so the thesis itself is 30 ECTS and is graded on the 0-5 scale. The programme-specific curriculum then gives the technical context in which that thesis topic should sit.

The current programme module is TTEM-CSNY Communication Systems and Networks. Its advanced-studies layer is COMM-S01 Advanced Studies in Communication Systems and Networks, at least 60 ECTS, with two documented focus areas: Networking and Radio Systems and Networks. COMM-S01 can also be taken as an extended 80 ECTS module. These focus areas help define a defensible thesis domain; they do not create different thesis regulations or different grading systems.

A second current course matters for thesis preparation: ITC.CEE.800 Tools for theses, 5 ECTS, pass/fail. Its current public 2026-2027 structure has four compulsory parts: research methods, research ethics, information searching skills and career planning. Older Tampere pages show historical seminar/presentation implementations, but those are not safe to present as today’s CSN requirement unless your current Sisu or Moodle implementation explicitly assigns them.

1. Start from the current curriculum, not an old telecom template

Communications engineering changes quickly, and university curricula change too. A thesis plan copied from an older student can therefore be wrong in two different ways: the technical baseline may be obsolete, and the administrative course structure may have changed. For current planning, anchor the work to the 2026-2027 CSEE curriculum, TTEM-CSNY, COMM-S01 and the current Technology thesis instructions.

COMM-S01 is deliberately broad. The Networking focus includes packet-switched wired and wireless networks, TCP/IP, MAC-level mechanisms, network planning, traffic modelling, performance analysis, simulation, future cellular systems, IoT and nanonetworking. Radio Systems and Networks develops deeper expertise in radio-system optimization and planning, propagation, system-level RF issues, modulation, detection, coding, topology and radio-resource management, including modern cellular and Wi-Fi systems. A thesis can sit clearly within one focus area or connect them if the research question genuinely requires it.

2. What the 30 ECTS thesis means

Thirty ECTS is the credit value of the thesis, not a promise that every thesis uses the same method or produces the same artefact. Tampere’s Technology framework permits empirical research, planning and implementation, and review of existing research literature. In CSN this can translate into simulation, measurement campaigns, protocol or network-software implementation, analytical modelling, performance evaluation, radio/network planning, data-driven analysis, testbed experimentation, or a carefully designed literature-based thesis.

The method has to answer the research question. Building a simulator, configuring a 5G stack, collecting packet traces or measuring a radio channel can be useful, but none of those activities is automatically a thesis contribution. The academic contribution comes from the bounded problem, justified method, evidence, analysis, limitations and conclusions.

3. Choose a topic that is narrow enough to evaluate

Topics such as “6G networks”, “IoT security”, “5G performance” or “AI for wireless networks” are themes, not thesis questions. Convert the theme into a system boundary and an evaluable claim. A stronger topic specifies the environment, mechanism, workload or channel condition, metric and comparison.

For example, instead of asking whether a scheduling algorithm is “better”, define the traffic pattern, topology, channel or mobility assumptions, baseline scheduler, target metrics and operating range. Instead of promising to “optimize a network”, define what is optimized, which constraints apply, and whether the output is a model, algorithm, configuration strategy or empirical comparison.

4. Networking-focus theses: make the experiment reproducible

Networking work often uses simulation, emulation, packet captures, real testbeds or software prototypes. Record topology, protocol versions, link characteristics, queue configuration, traffic generators, offered load, packet sizes, run duration and failure assumptions when they affect the result. If random processes are involved, document seeds or repetition strategy and report variation rather than only the best run.

Select metrics because they answer the research question. Throughput, latency, jitter, packet loss, fairness, availability, energy consumption and control overhead represent different properties. Reporting ten metrics without explaining their relevance usually produces a weaker thesis than reporting three well-justified metrics with uncertainty and a clear interpretation.

5. Radio Systems and Networks theses: control the measurement context

Radio-system work can be especially sensitive to hidden conditions. Propagation environment, antenna placement, frequency band, bandwidth, transmit power, receiver settings, synchronization, mobility, interference and hardware calibration can materially change a result. Document the parameters that a reader needs to understand the experiment.

If you use simulation, state the channel model, propagation assumptions, network geometry, user distribution, traffic assumptions, scheduler, coding/modulation settings and repetitions that matter. If you use field or laboratory measurements, record hardware and software versions, calibration procedure, geometry and environmental conditions when relevant. Measurement uncertainty and missing data should be discussed rather than silently removed.

6. 5G and 6G topics need a baseline, not hype

COMM-S01 explicitly reaches modern and emerging cellular systems, but using “5G” or “6G” in a title does not establish novelty. A thesis still needs a bounded research problem and a comparison. Possible baselines might be a reference algorithm, standard configuration, analytical expectation, previous implementation or conventional network architecture.

Be careful with standards language. Separate what is formally specified, what is a research proposal and what is your own experimental configuration. If you rely on a particular 3GPP release, simulator, open-source stack or vendor implementation, state the version. That prevents later readers from interpreting a version-specific result as a timeless property of the technology.

7. IoT, Industrial Internet and autonomous-system topics

The programme explicitly links communication infrastructure to IoT, Industrial Internet, robotization and autonomous systems. These topics often combine networking with application constraints such as reliability, latency, energy consumption, scalability or positioning. Decide which layer the thesis actually studies. Otherwise one project can expand uncontrollably across sensors, radio, networking, cloud, cybersecurity and application logic.

If the contribution is communication performance, keep the application as context and define communication metrics. If the contribution is architecture, justify the architecture and evaluate relevant trade-offs. If the contribution is an implemented system, plan validation that separates “it works in our demo” from stronger claims about robustness, scalability or general performance.

8. Simulation and modelling: avoid one-run evidence

Simulation is common in communications research because many network scenarios are expensive or difficult to reproduce physically. Its weakness is that a convincing graph can still be produced from weak assumptions. Define the model before interpreting the result. Record parameter ranges, scenario generation, repetitions and stopping conditions. Compare against a baseline and include variability or confidence information where meaningful.

Sensitivity analysis can be more informative than adding another algorithm. If the conclusion reverses when traffic load, mobility, propagation loss or user count changes slightly, that is an important finding. Explain the range in which your conclusion is supported.

For comparative studies, keep the comparison fair. Use the same workload, topology, radio assumptions and evaluation window unless the difference itself is part of the research question. Tune baselines according to documented practice rather than leaving them artificially weak. If different schemes require different computational or signalling resources, report that cost instead of comparing only the headline performance metric.

When results depend on statistical aggregation, state whether a plotted value is a mean, median, percentile, confidence interval or single run. Tail latency and outage probability can matter more than average performance in reliability-sensitive communications. Averages should not hide rare but consequential failures if those failures are relevant to the system objective.

9. Software and protocol implementation theses

A software artefact can be central to a CSN thesis, but implementation effort is not the same as research evidence. Before building, specify what technical claim the artefact enables you to test. Define functional verification separately from performance evaluation. Unit tests or successful packet exchange show that an implementation works; they do not establish efficiency, scalability, reliability or superiority.

For reproducibility, record repository or version identifiers, dependencies, configuration, deployment environment and test procedure when publication and confidentiality rules permit. If company code cannot be published, explain enough about the architecture, interfaces, evaluation method and limitations for academic assessment without exposing confidential material.

10. ITC.CEE.800 Tools for theses: what is current

The current 2026-2027 ITC.CEE.800 Tools for theses is 5 ECTS and pass/fail. Tampere lists four parts: research methods, research ethics, information searching skills and career planning, and recommends completing the course before starting the thesis. All parts of its completion option are compulsory.

This is a useful example of why current-source checking matters. Older curriculum pages show a presentation/seminar component and old CSN-targeted seminar implementations. The current public course description no longer lists that component. Therefore this guide does not claim that every current CSN student must complete the historical TST-01908 or an old one-credit seminar. Follow your current Sisu study plan and Moodle implementation if programme staff assign a presentation or seminar locally.

11. Starting requirements, topic and supervisor

Tampere’s Technology thesis instructions place the thesis near the end of MSc-level studies. Before starting, the student should have sufficient subject competence and must have completed the bachelor’s degree or received the required approval concerning the bachelor’s thesis. Begin searching for a topic early enough to identify an appropriate supervisor and realistic scope.

Topics may arise from university research, a company commission or the student’s own proposal. A company project remains an academic thesis. The university appoints the supervisor or supervisors, and academic suitability, scope and evaluation remain university responsibilities.

12. Use a Thesis Supervision Plan

The Technology thesis process uses a Thesis Supervision Plan. Treat it as a project-control document, not a formality. Agree on the topic, objectives, scope, intended methods, timetable, meeting rhythm, responsibilities and feedback process before the work becomes difficult to change.

For communications projects, also agree on access to testbeds, licensed software, network equipment, measurement hardware, datasets, compute resources and company systems. If equipment or access can disappear, identify a fallback method early.

If the project depends on a shared laboratory, spectrum licence, private 5G environment, restricted cloud account or company network, identify who can grant access and how long approvals take. Build setup and calibration time into the schedule. A fallback can be a smaller testbed, trace-driven evaluation or validated simulation, but changing method late may also change the research question, so discuss that possibility with the supervisor before it becomes necessary.

13. Write a research plan that separates system building from research

A useful CSN research plan explains the problem, research question, prior literature, system boundary, method, data or experiment design, evaluation metrics, expected limitations, ethics/data issues and schedule. If you build something, describe the artefact and the evaluation as separate work packages.

A common failure is to spend most of the thesis period implementing a platform and only then decide how to evaluate it. Reverse that order conceptually: define the evidence needed to answer the research question, then build only what is necessary to produce that evidence.

For a two-focus-area project, write down which part is primary. A study of radio-resource management over a packet-network architecture may touch both Networking and Radio Systems and Networks, but the thesis still needs one coherent problem statement. Use secondary concepts to support the main question rather than opening a second independent thesis inside the same project.

Also separate engineering constraints from research variables. A device limit, fixed spectrum allocation or company deployment rule may be a boundary condition rather than something you are testing. State such constraints explicitly so readers can distinguish unavoidable system context from variables whose effects you actually evaluate.

14. Data, privacy and confidential company systems

Network traces, device identifiers, logs and operational telemetry can contain personal, security-sensitive or commercially confidential information. Decide what data is actually necessary. Minimise collection, control access, document preprocessing and avoid copying production data into personal or public services without authorization.

Tampere theses are public documents. Confidential business information therefore cannot simply be embedded in the evaluated thesis. If an industry partner supplies sensitive topology, source code, customer data or security details, agree before data collection what can appear in the public manuscript and what must remain outside it.

15. AI tools in a communications-engineering thesis

Tampere permits AI use within its current study guidance, but the student remains responsible for the submitted work. If AI materially assists coding, literature exploration, language editing or other work, follow the current acknowledgement rules. Do not treat generated explanations, citations, protocol details or code as verified evidence.

Never paste confidential company information, unpublished datasets, credentials or protected source code into an AI service unless the applicable data and service rules permit it. For AI-assisted coding, test generated code and document the actual implementation used in the experiments.

16. Writing and presenting results

Write while the experiments are running. Maintain a parameter table, experiment log and result index so that figures can be traced back to configurations. A results chapter should distinguish observation from interpretation. Report failures or unexpected behaviour when they affect the research question.

A strong discussion explains why the result occurred, what alternative explanations remain, where the result generalizes and where it does not. Communications results are often scenario-dependent; saying so accurately is stronger than claiming universal superiority from a narrow test.

17. Maturity test and language route

A master’s degree also requires the maturity test under Tampere’s current rules. The applicable route depends on whether the student has already demonstrated the required language proficiency and on the language situation defined by the university. Do not assume another student’s maturity-test route applies to you.

Because maturity requirements interact with thesis assessment timing, confirm your own route early enough that it does not become an administrative surprise at submission.

18. Turnitin, originality and final submission

The final Technology thesis goes through Tampere’s originality process using Turnitin before final archival submission. Similarity is an academic-review signal, not an automatic plagiarism verdict. Quotes, references, technical terminology, standard descriptions and prior drafts can all affect similarity and need interpretation.

After the required checks and supervisor permission, the final thesis is submitted through the university’s process and archived in Trepo. Prepare the final file carefully; the archived thesis is the formal public document.

19. Trepo, PDF/A and publicity

Tampere requires permanent electronic archiving and uses PDF/A for the archival thesis file. Check conversion before the deadline because figures, fonts, equations and embedded objects can cause PDF/A problems.

The thesis is public. If a company project includes confidential material, restructure the project so that confidential details stay outside the evaluated thesis while the public document still contains enough method and evidence to support academic assessment.

20. Assessment and grading

The Communication Systems and Networks thesis follows the Technology thesis 0-5 grading scale. Tampere’s assessment looks at the thesis project as an academic whole, including the quality of the work and the written report under the university’s criteria. The university uses examiners and the formal approval process defined for Technology theses.

Keep the difference between course grading and thesis grading clear. ITC.CEE.800 is pass/fail; the 30 ECTS thesis is 0-5. Completing the preparation course does not determine the thesis grade.

21. Examiner timelines and graduation planning

Tampere’s thesis instructions set assessment timelines, but those timelines are not a guarantee of graduation on a particular date. The normal examiner period and the maturity-test route can affect the formal schedule, and summer has its own practical timing constraints. Work backwards from the desired graduation date and leave time for supervisor review, originality checking, PDF/A correction, final submission, examination and the separate graduation application.

A technically finished experiment is not the same as a submitted and approved thesis. Protect the last weeks for writing, checking and administration.

22. A practical CSN thesis checklist

Before starting: confirm that your Sisu plan, COMM-S01 focus and thesis topic align; identify the supervisor; complete or schedule current ITC.CEE.800; define data/equipment access; and agree on the Thesis Supervision Plan. Before experiments: freeze the research question, baseline, metrics and reproducibility plan. Before submission: verify claims against the actual evidence, complete the maturity requirement, run Turnitin as required, obtain supervisor permission, create a valid PDF/A, submit through the current process and complete the separate graduation steps.

The safest principle is to separate programme facts, current university rules, and your research design. Tampere determines the degree, 30 ECTS thesis process, assessment and archival requirements. COMM-S01 defines the communications-engineering academic domain. Your thesis then earns its strength from a focused question, controlled evidence and conclusions that stay within what the experiment or analysis actually supports.

If your current Sisu plan, supervisor instructions or course implementation differs from a public page, resolve that discrepancy before submission and retain the current written instruction you relied on.