post

Novel technology for monitoring nuclear weapons

In the future, this technology might help verify if countries abide by disarmament treaties.

An inter­na­tio­nal IT rese­arch team from Bochum, Prince­ton, and Har­vard has deve­lo­ped a tech­no­lo­gy that faci­li­ta­tes the moni­to­ring of chan­ges in nuclear silos without having to reveal secret infor­ma­ti­on about the stored wea­pons. In future, it is expec­ted to help veri­fy if coun­tries abi­de by dis­ar­ma­ment trea­ties. The rooms are phy­si­cal­ly moni­to­red with radio waves; a sophisti­ca­ted cryp­to­gra­phic tech­ni­que ensu­res that the pro­cess can­not be mani­pu­la­ted.

As far as rese­ar­chers are con­cer­ned, no chal­len­ge is grea­ter than moni­to­ring nuclear wea­pons: poten­ti­al atta­ckers in this case are ent­i­re nati­ons, rather than small groups of hackers or other cri­mi­nals. The coun­tries have almost unli­mi­ted finan­ci­al resour­ces at their dis­po­sal and have access to sta­te-of-the-art offen­si­ve tech­no­lo­gy.

In the inter­di­sci­pli­na­ry pro­ject, mem­bers of the Bochum-based Horst Görtz Insti­tu­te for IT Secu­ri­ty (HGI) col­la­bo­ra­te clo­se­ly with US-Ame­ri­can col­leagues from Prince­ton Uni­ver­si­ty and Har­vard Uni­ver­si­ty. A report about the work has been published in the sci­ence maga­zi­ne Rubin.

Radio wave map indi­ca­tes chan­ges

In order to iden­ti­fy chan­ges in a nuclear silo, rese­ar­chers deploy elec­tro­ma­gne­tic waves in the radio fre­quen­cy ran­ge. As they are reflec­ted by walls and objects, a uni­que radio wave map of the room can be gene­ra­ted. Every chan­ge – for examp­le if a war­head were to be remo­ved from the sto­ra­ge faci­li­ty – would chan­ge the refle­xi­on pat­tern and could thus be detec­ted. As a result, coun­try A could moni­tor the nuclear silos of coun­try B by reques­ting radio wave maps of the room in regu­lar inter­vals.

Howe­ver, we must make sure that a coun­try can­not gene­ra­te a radio wave map of a ful­ly sto­cked nuclear silo in advan­ce and then con­ti­nues to send it to coun­try A, even after the wea­pons had long been remo­ved,” exp­lains Dr Dr Ulrich Rühr­mair from HGI. To this end, the rese­ar­chers have inte­gra­ted a so-cal­led chal­len­ge into the sys­tem, i.e. a varia­ti­on in the request for a radio wave map bet­ween the coun­tries.

Pre­ven­ting decep­ti­on

In the room that has to be moni­to­red, 20 rota­ting mir­rors are instal­led, which can be remo­te­ly adjus­ted. The mir­rors reflect the radio waves, thus chan­ging the refle­xi­on pat­tern in the room, with each mir­ror set­ting crea­ting an indi­vi­du­al pat­tern. Pri­or to sen­ding the request, coun­try A would arran­ge the mir­rors in a cer­tain way. In reply, coun­try B would have to send the radio wave map of the room with the exact same mir­ror arran­ge­ment to coun­try A. This can be done only if coun­try B mea­su­res the room live with radio waves and the cur­rent mir­ror set­ting every time. Pre­vious­ly recor­ded radio wave maps would be useless.

Coun­try A can veri­fy the reply only if the refle­xi­on pat­terns for a num­ber of dif­fe­rent mir­ror set­tings were mea­su­red and saved when the tech­no­lo­gy was first imple­men­ted.

Mir­ror arran­ge­ment must not be pre­dic­ta­ble

The IT secu­ri­ty rese­ar­chers are cur­r­ent­ly tes­ting the sys­tem in a con­tai­ner at Ruhr-Uni­ver­si­tät, using dum­my war­heads and 20 mir­rors. This set­ting enab­les them to crea­te bil­li­ons of sextilli­ons dif­fe­rent mir­ror arran­ge­ments. “The chal­len­ge is to make sure that the moni­to­red coun­try doesn’t learn to pre­dict the next mir­ror set­ting over time,” says HGI rese­ar­cher Prof Dr Chris­tof Paar. Were this the case, the coun­try could gene­ra­te the requi­red radio wave map without scan­ning the room anew.

In order to pre­vent this sce­n­a­rio, the IT experts from Bochum deploy an unpre­dic­ta­ble cryp­to­gra­phic pro­to­col to align the mir­rors. “The important thing is to ensu­re that the cor­re­la­ti­on bet­ween the chal­len­ge and the reply can­not be descri­bed by a sys­tem of line­ar equa­ti­on,” says Zen­ger. “This is becau­se such sys­tems are rela­tively easy to figu­re out in mathe­ma­ti­cal terms.” The same app­lies to the phy­sics, i.e. the mir­ror mate­ri­als: their refle­xi­on pro­per­ties shouldn’t be line­ar eit­her.