https://wiki.hh.se/ceres/api.php?action=feedcontributions&feedformat=atom&user=Axelsson 2026-04-04T23:50:21Z User contributions MediaWiki 1.22.6 https://wiki.hh.se/ceres/index.php/Digital_forensics_investigations_of_IoT_using_electromagnetic_side_channels 2019-09-30T13:33:55Z

<p>Axelsson: Detect and analyse IoT devices through their EM radiation</p> <hr /> <div>{{StudentProjectTemplate<br /> |Summary=Use electromagnetic leakage from IoT devices to determine what software they're running<br /> |Keywords=electronics, computer architecture, security<br /> |TimeFrame=spring<br /> |References=http://dfrws.org/conferences/dfrws-usa-2019/sessions/leveraging-electromagnetic-side-channel-analysis-investigation<br /> |Prerequisites=electronics, electromagnetic field theory, physics, security<br /> |Supervisor=Stefan Axelsson, Mark Dougherty, Mohamed Eldefrawy <br /> |Author=Stefan Axelsson<br /> |Level=Master<br /> |Status=Open<br /> }}<br /> When computers run, they give rise to electromagnetic radiation that can be picked up by a nearby probe. It has long been known that this EM radiation can leak information about what the computer is doing, even up to the point of being able to determine (with statistical certainty) a particular cryptographic key that is in use. <br /> <br /> However, in digital forensics to come, police will arrive at a crime scene and not even know what devices are present, and what bearing that could have on the case. This due to IoT devices, which will probably be scattered around the landscape. Thus finding, and determining what these devices are doing is valuable from a crime fighting perspective. Determining whether the firmware an IoT device is running is the original, or has been hacked would be useful. <br /> <br /> We have already started work in this field, first by trying to replicate previous results, but there are many obvious new ways of taking this research, which particular ones are upp for discussion. As this is an area of current research here at Halmstad, we would aim for a result that is publishable.</div>

Axelsson https://wiki.hh.se/ceres/index.php/Radar_based_chronograph_(bullet_sensor) 2019-09-30T08:33:43Z

<p>Axelsson: Dopler radar (rifle) bullet speed measurement</p> <hr /> <div>{{StudentProjectTemplate<br /> |Summary=Develop a cheap chronograph to measure bullet speed using a radar sensor<br /> |Keywords=electronics, computer engineering, physics<br /> |TimeFrame=spring<br /> |References=https://mylabradar.com<br /> |Prerequisites=electronics, physics, computer engineering<br /> |Supervisor=Stefan Axelson, Pererik Andreasson<br /> |Author=Stefan Axelsson<br /> |Level=Flexible<br /> |Status=Open<br /> }}<br /> Measurement of bullet speed is important in many situations. From speed measurement when developing rifle ammunition (to achieve accuracy and ensure the round is legal to hunt with) to ballistic measurement of guns shots in crimes (or war zones).<br /> <br /> Most simple chronographs today use optical sensors, but these have the drawback that the bullet has to pass through a very well defined measurement zone (on the order of 20x20cm). This also means that there is a considerable risk to shoot the chronograph, and that one has to go down range to set it up (which is a safety hazard). <br /> <br /> There have however been one chronograph developed on the dopler radar principle. This is however rather expensive, while the radar modules etc. shouldn't really be. It also uses the 22GHz band, which is a bit low for the optimal detection of the base of a bullet (which is on the order of 7-8mm across). There are now many cheaper radar modules for the automotive market, one should be able to adapt one of these for this simple dopler case (i.e. dopler shift is the only measurement made, angular resolution is not needed). <br /> <br /> <br /> So this project entails developing a chronograph that is highly accurate (better than one part per thousand), cheap, based on the dopler radar principle, and that can be placed behind the muzzle of the rifle, and report speed to the shooter.<br /> <br /> Sponsorship and guidance from the hunting/gun shop just across the parking lot from the school has been secured in the form of an experienced former police officer with access to rifles and the local range. (Thus also ensuring the legality and safety of the project).</div>

Axelsson https://wiki.hh.se/ceres/index.php/Acoustic_bullet_detection_and_measurement 2019-09-30T08:25:42Z

<p>Axelsson: Develop bullet speed chronograph based on acoustic measurement</p> <hr /> <div>{{StudentProjectTemplate<br /> |Summary=Measure speed and direction of a supersonic rifle bullet with acoustic sensors<br /> |Keywords=electronics, physics, computer engineering<br /> |TimeFrame=spring<br /> |References=https://pdfs.semanticscholar.org/5ece/afa03dfb1a9233588d7c9077378c47d05183.pdf<br /> https://hal.archives-ouvertes.fr/hal-01852518/document<br /> |Prerequisites=electronics, computer engineering<br /> |Supervisor=Stefan Axelsson, Pererik Andreasson<br /> |Author=Stefan Axelsson<br /> |Level=Flexible<br /> |Status=Open<br /> }}<br /> Measurement of bullet speed and direction of travel is important in many situations. From speed measurement when developing rifle ammunition (to achieve accuracy and ensure the round is legal to hunt with) to ballistic measurement of guns shots in crimes (or war zones). <br /> <br /> Most simple chronographs today use optical sensors, but these have the drawback that the bullet has to pass through a very well defined measurement zone (on the order of 20x20cm). This means that down range (i.e. close to the target) measurements are not possible as the risk of missing (and shooting the measurement device) are too great. This is problematic in that Swedish legislation specify minimum energy for hunting ammunition at 100 meters, not the muzzle. <br /> <br /> However, the bow shock wave from a supersonic bullet is of very short duration and easy to detect. Hence the careful placement of microphones somewhere along the bullet path should lead to an accurate measure of bullet flight, both speed and direction. (Measuring where the bullet hit on a paper target should be possible as well, given the proper geometry of the microphones). <br /> <br /> So this project entails developing a chronograph that is highly accurate (better than one part per thousand) and that can be more freely placed down range, and report speed (and placement/direction if possible) to the shooter. <br /> <br /> Sponsorship and guidance from the hunting/gun shop just across the parking lot from the school has been secured in the form of an experienced former police officer with access to rifles and the local range. (Thus also ensuring the legality and safety of the project).</div>

Axelsson https://wiki.hh.se/ceres/index.php/FEM_simulation_of_bullet_sensor 2019-09-30T07:38:22Z

<p>Axelsson: Develop and test FEM model of bullet speed measuring sensor</p> <hr /> <div>{{StudentProjectTemplate<br /> |Summary=Simulate a magnetic reluctance sensor that detects speeding bullet<br /> |Keywords=electro magnetism, physics<br /> |TimeFrame=spring<br /> |References=https://magnetospeed.com<br /> https://patents.google.com/patent/US9709593B1/en<br /> |Prerequisites=electromagnetics<br /> |Supervisor=Stefan Axelsson, Struan Gray, Pererik Andreasson<br /> |Author=Stefan Axelsson<br /> |Level=Flexible<br /> |Status=Open<br /> }}<br /> A new entry on the market of ballistic chronographs (i.e. devices that measure the muzzle velocity of bullets) is Magnetospeed. While other, previous, sensors typically measure light occlusion, this chronograph works on the principle of magnetic reluctance (the same type of sensor that measures the rotational speed of the wheels in your car for the ABS system). Here a bullet, of a conducting but non-magnetic material, is shot into a static magnetic field. This field induces a current in the bullet, that in turn induces a magnetic field that is picked up by a coil. <br /> <br /> Muzzle velocity is important from a number of aspects, i.e. developing loads that shoot accurately, but also make sure that hunting ammunition fulfill legal energy requirements for particular types of hunting etc. <br /> <br /> I have already built such a sensor bar, and they are commercially available, but the geometry, field strengths etc. are not sufficiently well understood from an engineering perspective. An analytical solution is probably complex, but a FEM (Finite Element Modeling) solution is probably within reach. We have the COMSOL software at the school, but there are also free and open multi-physics simulation software available such as ELMER (from CSC in Finland). <br /> <br /> So this thesis would develop such a model based on real bullets, speeds etc. and calibrate/compare it with real field measurements (using a .223 and .308 caliber hunting rifle with moderator). Sponsorship and guidance from the hunting/gun shop just across the parking lot from the school has been secured in the form of an experienced former police officer with access to rifles and the local range. (Thus also ensuring the legality and safety of the project).</div>

Axelsson https://wiki.hh.se/ceres/index.php/Read_hard_drive_with_AFM 2019-09-30T07:19:55Z

<p>Axelsson: </p> <hr /> <div>{{StudentProjectTemplate<br /> |Summary=See if it's possible to read bits from a hard-drive with an Atomic Force Microscope<br /> |Keywords=electronics, physics, computer engineering, security, digital forensics<br /> |TimeFrame=spring<br /> |References=https://www.usenix.org/legacy/publications/library/proceedings/sec96/full_papers/gutmann/index.html<br /> |Prerequisites=An understanding of digital electronics and physics<br /> |Author=Stefan Axelsson<br /> |Supervisor=Stefan Axelsson, Struan Gray<br /> |Level=Master<br /> |Status=Open<br /> }}<br /> Hard drives store data by magnetizing a disk that spins by a read head at high speed. Hard disk density depending on how close together the tracks are (and how small the magnetic domains are) increases all the time. However, recovering data from broken hard drives (whether by accident, or someone put a sledge hammer to them to try and destroy the data) is always important. Both to understand what's possible when it comes to securely deleting data (and destroying data), and for e.g. criminal investigations. <br /> <br /> At the school we have a fairly good AFM - Atomic Force Microscope, that we'd hope to put to use to read the magnetic data regions on a disassembled, modern, hard drive. This is work that has been done before, but it needs to be done repeatedly as hard drives develop.</div>

Axelsson https://wiki.hh.se/ceres/index.php/Read_hard_drive_with_AFM 2019-09-30T07:16:23Z

<p>Axelsson: Read data directly from a hard drive surface using AFM microscopy</p> <hr /> <div>{{StudentProjectTemplate<br /> |Summary=See if it's possible to read bits from a hard-drive with an Atomic Force Microscope<br /> |Keywords=electronics, physics, computer engineering, security, digital forensics<br /> |TimeFrame=spring<br /> |References=https://www.usenix.org/legacy/publications/library/proceedings/sec96/full_papers/gutmann/index.html<br /> |Prerequisites=An understanding of digital electronics and physics<br /> |Supervisor=stefan axelsson, struan gray<br /> |Level=Master<br /> |Status=Open<br /> }}<br /> Hard drives store data by magnetizing a disk that spins by a read head at high speed. Hard disk density depending on how close together the tracks are (and how small the magnetic domains are) increases all the time. However, recovering data from broken hard drives (whether by accident, or someone put a sledge hammer to them to try and destroy the data) is always important. Both to understand what's possible when it comes to securely deleting data (and destroying data), and for e.g. criminal investigations. <br /> <br /> At the school we have a fairly good AFM - Atomic Force Microscope, that we'd hope to put to use to read the magnetic data regions on a disassembled, modern, hard drive. This is work that has been done before, but it needs to be done repeatedly as hard drives develop.</div>

Axelsson