Bluetooth Low Energy sensor networks for railway applications

Blue oo h Low Ene gy senso ne wo ks o ailway applica ions Ángela He nández, An onio Valdo inos A agon Ins i u e o Enginee ing Resea ch (I3A) Uni e si y o Za agoza Za agoza, Spain Da id Pe ez-Diaz-de-Ce io, Jose Luis Valenzuela Signal Theo y and Communica ions Depa men Uni e si a Poli ècnica de Ca alunya Cas ellde els, Spain Abs ac — This pape analyzes he iabili y o Blue oo h Low Ene gy (BLE) as he enabling echnology in ailway scena ios o allow senso da a ansmission om g ound o ain. This p oposal is based on using he b oadcas ing capabili ies in oduced by he Blue oo h speci ica ions since 4.0. The las e sion, 5.0, ex ends he ange he speed and he b oadcas capaci y. Howe e , due o i s ecen publica ion i is ha d o ind eal de ices wi h i implemen ed in he ma ke . The p oposed schema has been alida ed by means o expe imen al measu emen s using 4.2. Keywo ds— BLE; ehicula ne wo ks; ene gy consump ion; ailway communica ions I. INTRODUCTION In he las ew yea s, In e ne o he Things (IoT) has expe imen ed a echnological leap in senso applica ions. The new wi eless s anda ds allow o ansmi mo e in o ma ion wi h less ene gy consump ion and wi h mo e eliabili y. Hund eds o senso s and in o ma ion sys ems can be in eg a ed in he ailway o con inuously moni o hei pa ame e s c ea ing in elligen anspo sys ems. The use o wi ed and wi eless elemen s o sensing and ailway communica ions in ailway lines has many challenges. Ene gy consump ion, eliabili y and delay a e he main pa ame e s which limi he use o au onomous senso s. The Eu opean Rail T a ic Managemen Sys em (ERTMS) [1] o Communica ions-Based T ain Con ol (CBTC) [2] ha e been implemen ed o imp o e he in e ope abili y o sys ems and o c ea e a s anda d o ain con ol and command sys ems. Ne e heless, hese sys ems lack o lexibili y o ins all he senso s in de e mined condi ions. The main objec i e o his wo k is o in es iga e he easibili y o using Blue oo h Low Ene gy (BLE) wi eless echnologies o ain communica ion scena ios and hei u u e in eg a ion in he ailway communica ions sys ems. Many di e en senso s (Fig. 1) could make use o BLE as i s enabling echnology: ailway wo ke p o ec ion sys ems, collision a oidance, me eo ological senso s, ack moni o ing sys ems, ain de ec ion mechanisms, pan og aph and ca ena y senso s, e c. Gene ally, Blue oo h is uled ou as a sui able echnology o ehicula communica ions. The e is a much ex ended hough which says ha Blue oo h is limi ed o a e y sho ange and equi es a conside ably amoun o ime o s ablish a connec ion and ansmi da a. In [3], i was al eady demons a ed ha classic Blue oo h could be employed o ailway communica ion sys ems. Howe e , he in e nal ope a ion and he oles o he employed de ices is comple ely di e en when using BLE. In [3], he classic Blue oo h senso s equi ed o be cons an ly scanning o he a i al o he con oy, which was gene a ing emo e name eques s. When he senso ecei ed his eques i answe ed wi h he s o ed da a. This mechanism, al hough iable, consumes a huge amoun o ene gy on he senso side. Wi h he in oduc ion o BLE, we p opose o swi ch he oles and he place he scanne inside he ain engine. Thus, he powe consump ion o he senso s will be educed d as ically. Fig. 1. Example o a ailway scena io senso deploymen So, in his pape , we will demons a e ha BLE is app op ia e o his kind o communica ions imp o ing he pe o mance ob ained by classic Blue oo h in e ms o da a ecep ion la ency and ene gy consump ion wi h a simila ange. II. BLE OVERVIEW The BLE speci ica ion de ines a comple e wi eless a chi ec u e in ended o enable low-powe communica ions. BLE de ines 40 b oadcas and connec ion channels in he Indus ial Scien i ic Medical (ISM) 2.4 GHz band wi h a physical da a a e o 1 Mbps. Unlike classical Blue oo h, BLE de ines h ee special channels (37, 38 and 39), called ad e ising channels used o he p ocess o disco e ing, ini ia ing a connec ion and b oadcas ing da a. In his wo k, we used non-connec able undi ec ed ad e ising, which is one o he di e en ad e ising e en s p oposed by he s anda d. Thus, he ad e ise pe iodically gene a es e en s, which consis o a sequence o packe s in he h ee ad e ising channels. The pe iod be ween hese e en s is de e mined by he ad e ising in e al pa ame e (Tad In e al) plus a andom pe iod be ween 0 and 10 ms (τ andomDelay) o a oid consecu i e collisions among di e en ad e ise s. This wo k has been suppo ed in pa by he MINECO/FEDER unde he p ojec s TEC2014-58341-C4-2-R and TEC2014-60258-C2-2-R. The o al packe size is 47 by es, whe e he payload is a iable om eigh o 39 aw by es. Howe e , in ou expe imen s, he payload consis s o a wo-by e heade plus he ADV add ess, a h ee-by e lags ield o allow he eading o he ad e ising by s anda d mobile applica ions and he use da a i sel anging om 1 o 26 by es plus wo by es o leng h and ype o da a. III. EXPERIMENTAL PERFORMANCE OF BLE To e alua e he sui abili y o BLE o g ound o ain communica ions we analyzed i s ange, he da a ecep ion delay, i s pe o mance on high densi y senso ne wo ks and he ene gy consump ion. A. Range I is widely belie ed ha Blue oo h has a limi ed ange o ens o me e s. E en he Blue oo h SIG has ecen ly published a whi e pape o demys i y his ac [4]. The eal e ec i e ange is condi ioned by he ansmi ed powe , dis ance, p opaga ion condi ions, he an ennae and he ecei e sensi i i y. Ne e heless, a quick e alua ion could be done using he ee space p opaga ion model [5] o wi h simple measu es [3]. In equa ion 1, PR is he ecei ed powe in Wa s, λ is he elec omagne ic wa eleng h (0.125 m), G is he an enna gain, d is dis ance be ween ansmi e and ecei e . The subsc ip s T and R deno e each one o hem, espec i ely. Using he maximum equi alen iso opically adia ed powe (EIRP) in he 2.4 GHz band, 100 mW, combined bo h wi h high sensi i i y ecei e s (−93 dBm) and 3 dBi ecei e an enna, Blue oo h can heo e ically allow anges up o nea i e kilome e s: 𝑃𝑃𝑅𝑅=𝑃𝑃𝑇𝑇·𝐺𝐺𝑇𝑇·𝐺𝐺𝑅𝑅·�𝜆𝜆 4𝜋𝜋𝑑𝑑�2 (1) Howe e , in p ac ice, we ha e measu ed ha due o o he impai men s, e.g., he p esence o obs acles o he e ec o he g ound, wea he condi ions, he losses in oduced by bo h he an enna cable and he connec o s, e c., he ange is limi ed o abou wo kilome e s. B. Da a ecep ion delay and high densi y senso ne wo ks Ano he ac ha should be e u ed is ha Blue oo h in oduces oo much delay on he connec ion and da a ansmission which makes i un iable o ehicula communica ions whe e high speed mo ing nodes a e in ol ed. Wi h his objec i e in mind, we designed a senso ne wo k wi h up o 15 de ices ( om wo o 14 ad e ise s and one scanne ) and we measu ed he mean ime be ween consecu i e ad e isemen s coming om he same de ice. We made hese measu emen s o h ee use payloads (1, 10 and 26 by es) and h ee di e en ad e ising in e als (100, 300 and 500 ms). The esul s a e depic ed in Fig. 2. The e, we can app ecia e ha o a couple o de ices, he da a can be ob ained in he expec ed alue o Tad In e al + τ andomDelay. We can ex ac mo e conclusions om his igu e. Fo example, when he numbe o de ices inc eases, he collisions and, he e o e, he non-de ec ions also inc ease. Ne e heless, hey keep on a sus ainable a e. The packe size also has a small impac on he esul s, he la ge he packe , he mo e p obable a e he collisions. And, hus, he da a ecep ion ge s sligh ly delayed. (a) Tad In e al = 100 ms (b) Tad In e al = 300 ms (c) Tad In e al = 500 ms Fig. 2. Mean ime be ween consecu i e de ec ions So, he delay un il we ecei e a packe depends no only on Tad In e al, bu also on he numbe o elemen s o he senso ne wo k. Fig. 3 depic s he p obabili y densi y unc ion o ecei ing an ad e isemen in a wo s case scena io wi h 25 eal ad e ise s using a Tad In e al o 100ms and ansmi ing 26-by e use da a packe s. The e, we can see ha i he scanne could emain unde he co e age a ea o a leas 600 ms i will ecei e da a om all he ad e ise s o he senso ne wo k. Fig. 3. P obabili y densi y unc ion o ecei ing an ad e isemen (25 nodes) To complemen hese measu emen s, we designed an expe imen whe e he BLE scanne was placed in a ca a 120 km/h and hen i came in o he p oximi ies o a small senso ne wo k composed o i e de ices. The ad e ise s used a Tad In e al o 100ms. Fig. 4 depic s he esul s ob ained. Each poin in he igu e ep esen s he ecei ed signal s eng h indica o (RSSI) o each ecei ed ad e isemen . As can be seen, a his speed, he mobile node akes nea en seconds o a i e o he poin nea es he senso s since i ecei es he i s ad e isemen . Thus, combining hese esul s wi h he p e iously ob ained in Fig. 3, he mobile has enough ime o ecei e he da a om all he senso s and e en i would be possible o educe he ad e ising in e al o sa e ene gy. Fig. 4. RSSI o he ecei ed packe s wi h a high speed mobile node C. Ene gy consump ion Finally, one key aspec o discuss when alking abou BLE is he ene gy consump ion. Classic Blue oo h ene gy consump ion is in he mA o de o magni ude, o example, he SPBT3.0DP2 chipse om STMic oelec onics has a cu en consump ion in inqui y and page scan mode o 9 mA [6]. On he o he hand, o he p oposed sys em wi h BLE, he a e age cu en consump ion o he di e en employed pa ame e s is shown in Table I. I can be no iced ha he wi eless senso de ices need up o 500 imes less ene gy using BLE. IV. CONCLUSIONS In his wo k we conduc ed and discussed se e al eal expe imen s o demons a e ha Blue oo h is sui able o g ound o ain connec i i y in ailway communica ions. The p esen ed wo k alida es ha Blue oo h Low Ene gy can be used as a wi eless long ange sys em, and i is possible o ansmi da a o high speed mo ing ecei e s. We assumed ha a 26-by e use da a packe can ca y enough in o ma ion o ailway senso applica ions and, hus, he ad e ising p ocedu e is p oposed agains o he connec ion mechanisms. Using his p ocedu e, a message o up o 26 by es can be ansmi ed in a e age a ound 100 ms wi h a e y low ene gy consump ion. On he o he hand, i has been demons a ed ha his da a exchange can be comple ed wi hin wide co e age a eas, up o wo kilome e s, wi h s anda d and comme cial de ices equipped wi h omnidi ec ional an ennas. Al hough he p esen ed esul s we e pe o med in nea op imal condi ions such as line o sigh , we ha e e i ied ha Blue oo h sys ems also ope a e p ope ly ansmi ing wi h a leaky eede o in bad wea he condi ions. Mo eo e , Blue oo h echnology can use mesh ne wo k p o ocols ha can ex end da a ansmission o e ens o kilome e s. The esul s p esen ed in his pape a e ob ained using eal deploymen s. So, we can conclude ha Blue oo h is a iable wi eless echnology o designing senso s used in complex ailway communica ions. TABLE I. AVERAGE CURRENT CONSUMPTION Ad e ising in e al (ms) Use da a payload (by es) Redbea Nano ( µ A) Cyp ess PSoC4-A BLE (µA) 100 1 17.3 235 100 10 18.5 268 100 26 20 327 300 1 10.4 79.24 300 10 10.8 90.32 300 26 11.5 110 500 1 9.1 48 500 10 9.2 54 500 26 9.5 66 REFERENCES [1] S. K. Abed, “Eu opean Rail T a ic Managemen Sys em - An o e iew,” 2010 1s In . Con . Ene gy, Powe Con ol, 2010. [2] IEEE s anda ds, “IEEE S anda d o Communica ions-Based T ain Con ol (CBTC) Pe o mance and Func ional Requi emen s.” . [3] J. Higue a, E. Ka sakli, J. L. Valenzuela, L. Alonso, A. Laya, R. Ma inez, and A. Aguila , “Expe imen al S udy o Blue oo h, ZigBee and IEEE 802.15.4 Technologies on Boa d High-Speed T ains,” in 2012 IEEE 75 h Vehicula Technology Con e ence (VTC Sp ing), 2012, pp. 1–5. [4] C. Coupe, S. Hegende e , and S. Schmid , “Debunking he my h: Blue oo h® Range.” p. 8, 2016. [5] H. T. F iis, “A No e on a Simple T ansmission Fo mula,” P oc. IRE, ol. 34, no. 5, pp. 254–256, May 1946. [6] STMic oelec onics, “SPBT3.0DP2 Blue oo h® Classic module Da ashee .” 2017.